Drinking Water Criteria Document for Haloacetonitriles Final Draft


United States	Office of Science

Environmental Protection and Technology
Agency	Washington, D.C.

June 30, 2002
EPA-822-R-03-020

£ EPA Office of Water

Drinking Water Criteria Document
for Haloacetonitriles

Final Draft

Prepared for

Health and Ecological Criteria Division
Office of Science and Technology
401 M Street, SW
Washington, DC 20460

-------
Drinking Water Criteria Document for Haloacetonitriles

TABLE OF CONTENTS

Table of Contents	i

Chapter!	Executive Summary	 1-1

Chapter II.	Physical and Chemical Properties 	II-1

Chapter III.	Toxicokinetics 	Ill-1

A.	Absorption	Ill-1

B.	Distribution 	III-3

C.	Metabolism	TTT-9

D.	Excretion	Ill-18

E.	Bioaccumulation and Retention 	111-20

F.	Summary 	111-20

Chapter IV.	Human Exposure 	IV-1

Chapter V.	Health Effects in Animals	V-l

A.	Short-Term Exposure	V-l

B.	Long-Term Exposure	V-16

C.	Reproductive and Developmental Effects	V-26

D.	Mutagenicity and Genotoxicity	V-49

E.	Carcinogenicity	V-60

F.	Summary 	V-65

Chapter VI.	Health Effects in Humans	VI-1

Chapter VII.	Mechanisms of Toxicity and Sensitive Subpopulations 	 VII-1

A.	Biochemical Basis of Toxicity	 VII-1

B.	Mechanisms of Carcinogenesis	 VII-11

C.	Interactions and Suceptibilities	 VII-14

D.	Summary 	 VII-16

Chapter VIII.	Quantification of Toxicological Effects	VIII-1

A.	Introduction to Methods	VIII-1

B.	Noncarcinogenic Effects	VIII-8

C.	Carcinogenic Effects	VIII-42

D.	Characteristics of Uncertainties and Data Groups 	VIII-44

Chapter IX.	References	IX-1

Appendix A.	Benchmark Dose Modeling Results 	A-l

Appendix B.	Benchmark Dose Modeling Output (Available only in electronic form)

EPA/OW/OST/HECD	i	Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Acknowledgements

This document is an update and expansion of the Rough Final Draft for the Drinking Water
Criteria Document on Haloacetonitriles, Chloropicrin and Cyanogen Chloride (U.S. EPA,
1987) and the 1993 Draft Drinking Water Health Advisory for Haloacetonitriles, from EPA's
Office of Water. This document includes an evaluation of literature on the HANs resulting from a
full literature search for toxicity data conducted in December 1999, and exposure data in May
2002. Key newer studies identified after the literature search date have been included as available
at the time of document preparation.

Chemical Manager:

Nancy Chiu, Ph.D.

Contractor Authors:

Andrew Maier, Ph.D., C.I.H. (TERA)

Claudine Kasunic (GRAM, Inc.)

Lynne T. Haber, Ph.D. {TERA)

Joan S. Dollarhide, M.S., M.T.S.C., J.D. {TERA)

Bruce Allen, M.S. (ENVIRON)

Nathan Bowles (GRAM, Inc.)

External Peer Reviewers

Patricia McGinnis, Ph.D., DABT, Syracuse Research Corporation

John Reif, M.Sc. (Med), D.V.M., Colorado State University

Alan Stern, Dr.P.H., DABT, New Jersey Department of Environmental Protection

EPA/OW/OST/HECD

ii

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
Chapter I. Executive Summary

A. Introduction

Haloacetonitriles (HANs) are derivatives of acetonitrile (CH3CN), in which one to three
halogen atoms are substituted for hydrogen. The four halogenated acetonitriles selected for
consideration in this document are bromochloroacetonitrile (BCAN), dibromoacetonitrile
(DBAN), dichloroacetonitrile (DCAN), and trichloroacetonitrile (TCAN). These HANs were
selected for inclusion in this document in consideration of the prevalence of individual HANs in
drinking water, and the availability of toxicity data. This document includes an evaluation of
literature on the HANs resulting from a full literature search for toxicity data conducted in
December 1999, and exposure data in May 2002. Key newer studies identified after the literature
search date have been included as available at the time of document preparation.

B. Human Exposure

The Information Collection Rule (ICR) database (U.S. EPA, 2002a) contains extensive
information on concentrations of BCAN, DBAN, DCAN, and TCAN in drinking-water systems,
and on how those concentrations vary with input-water characteristics and treatment methods.
The database contains information from six quarterly samples from 7/97 to 12/98, from
approximately 300 large systems covering approximately 500 plants. The mean concentrations of
BCAN were 0.73 and 1.14 //g/L in groundwater and surface water, respectively. The mean
concentrations of DBAN were 0.82 and 0.75 //g/L in groundwater and surface water,
respectively. The mean concentrations of DCAN were 0.87 and 2.21 //g/L in groundwater and
surface water, respectively. The mean concentrations of TCAN were 0.14 and 0.03 //g/L in
EPA/OW/OST/HECD 1-1 Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
groundwater and surface water, respectively. The median concentrations of BCAN, DBAN,
DC AN, and TCAN were less than their means in groundwater and surface water.

HANs are produced during water chlorination or chloramination from naturally occurring
substances, including algae, humic acid, fulvic acid, and proteinaceous material. Reckhow el al.
(1990) found that disinfection of water containing humic acids resulted in higher concentrations of
HANs than disinfection of water containing the corresponding fulvic acids.

The disinfection process producing the highest concentration of HANs was chlorination.
Chloramine produced lower levels of HANs. Most investigators (Boorman et al., 1999;
Richardson 1998; Lykins et al., 1994; Jacangelo et al., 1989) found that the formation of HANs
when ozonation was followed by chlorine or chloramine was less than when chlorine or
chloramine was the sole disinfectant. Interestingly, Miltner et al. (1990) reported that the
formation of DBAN in simulated distribution water was higher (p = 0.05) when ozonation was
combined with chlorination or with chloramination than when chlorination was used alone. In
addition, Miltner et al. (1990) found that ozonation had no statistically significant effect on the
formation of BCAN, DCAN, or TCAN. Richardson (1998) found that BCAN, DCAN, and
TCAN were not produced in measurable quantities by ozonation or chlorine dioxide. However,
DBAN was formed by ozone in the presence of elevated bromide, but not by chlorine dioxide
disinfection.

Ambient bromide levels appear to influence, to some degree, the speciation of HANs.
DCAN is by far the most predominant HAN detected in drinking water from sources with

EPA/OW/OST/HECD

1-2

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

bromide levels of 20 |ig/L or less. In treated water from sources with higher bromide levels
(50-80 //g/L), BCAN was the second most prevalent compound (WHO, 2000). Richardson
(1998) found that when bromide was present in the source water, DBAN concentrations were
greater than those of chloroform or dichloroacetic acid, which normally predominate.

In general, increasing temperature and/or decreasing pH has been associated with
increasing concentrations of HANs (AWWARF, 1991; Siddiqui & Amy, 1993). Although HANs
form rapidly, they decay in the distribution system as a result of hydrolysis. HANs hydrolyzed at
pH levels >9.0 and continued to degrade in the distribution system (Arora el al.,\ 997), The
relative stability of individual HANs appears to be dependent on the specific source water
(AWWARF, 1991).

In general, there were no clear trends of the concentrations of HANs with season.
However, among 35 water treatment facilities investigated, Krasner et al. (1989) found that at the
facility with the highest bromide level (~ 3 mg/L bromide), there was a shift in the distribution of
HANs from chlorinated HANs to brominated HANs.

TCAN has been used as an insecticide (Budavari et al., 1989). No data were located on
exposure to BCAN, DBAN, DCAN, and TCAN in food, air, or via dermal exposure when
showering or swimming.

EPA/OW/OST/HECD

1-3

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

C. Toxicokinetics

Limited data are available on the toxicokinetics of the HANs, with a comprehensive
toxicokinetic study for oral dosing available only for DCAN, However, the existing toxicokinetic
data suggest that HANs can be rapidly and nearly completely absorbed following oral dosing
(Roby et al., 1986; Roth et al., 1990). Systemic toxicity data suggest that HANs are absorbed by
the dermal route. Once absorbed, HANs appear to be widely distributed. The two compounds
tested, DCAN (Roby et al., 1986) and TCAN (Lin et al., 1992), were widely distributed
following oral dosing, with no clear preferences in tissue distribution apparent based on the
limited data. No data were available on tissue-dependent metabolism, but an overall metabolic
scheme for HANs involving an initial oxidative dehalogenation step has been proposed based on
the ability of these compounds to form cyanide and metabolism studies for other nitriles (Pereira
et al., 1984). Proposed intermediate metabolites have not been measured directly, and the
identity of enzymes responsible for steps in the pathway have not been identified. Conjugation
with glutathione (GSH), at least at high doses, might be a second important route of metabolism
for HANs (Ahmed et al., 1989; Lin and Guion, 1989; Ahmed et al., 1991; NTP, 2002).

Excretion of HANs is nearly complete over a period of days, largely in urine and in exhaled air.
The rate of excretion may differ across species, since mice excrete DCAN more rapidly than rats
(Roby et al., 1986). Differences in urinary excretion of thiocyanate for different HANs was
observed by Pereira et al. (1984), with TCAN being excreted as thiocyanate to a lesser degree
than the other HANs. The results of Roby et al. (1986) that showed relatively rapid excretion of
DCAN-associated radioactivity suggests limited potential for the bioaccumulation of the HANs.
However, no studies were located that provided data on long-term accumulation and retention of
any of the HANs.

EPA/OW/OST/HECD

1-4

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
D. Health Effects In Animals

The toxicity data on the HANs are summarized in Tables V-15 through V-18. Overall,
very little data are available evaluating the non-cancer effects of the HANs. Acute oral LD50
values for DBAN, DC AN, and TCAN in rodents have been reported to range from 50 to 361
mg/kg. DBAN and TCAN have been reported to be irritants. DBAN causes eye, nasal, and
respiratory tract irritation following inhalation, and skin irritation following dermal exposure.
TCAN also causes skin irritation following dermal exposure. No data on the acute toxicity of
BCAN are available, and no subacute or subchronic studies of either BCAN or TCAN are
available. No chronic studies have been conducted on any of the HANs.

EPA/OW/OST/HECD

1-5

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
basis for the quantitative dose-response assessment. Taken together, the data suggest that
decreased body weight appears to be the primary indicator of toxicity for DBAN, Overall, male
rats appear to be more sensitive than female rats for DBAN. For DC AN, consistent, compound-
related, dose-dependent effects were limited to decreased body weight and increased liver weight
(Hayes et al., 1986). In this case, the observed liver weight increases were supported by changes
in serum biochemistry parameters suggestive of liver damage. No histopathological evaluation
was done in the key study for DCAN (Hayes et al., 1986), so the degree, if any, of liver damage
can not be confirmed. The data for TCAN and BCAN are too limited to identify with confidence
any potential target organs.

EPA/OW/OST/HECD

1-6

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

conducted using tricaprylin as a vehicle, because these compounds are very miscible in this
vehicle. However, in these studies, comparison of tricaprylin versus water-treated controls
revealed increased embryotoxicity due to tricaprylin. A recent study by Christ el al. (1996)
indicates that tricaprylin also influences the pattern of malformations observed in fetuses caused
by TCAN. For TCAN in corn oil, the malformations were primarily cranio-facial in nature while
for TCAN in tricaprylin the malformations were primarily cardiovascular and urogenital in nature.
Therefore, the use of data from studies in which tricaprylin was used as the vehicle is not
appropriate for risk assessment purposes. In the one developmental toxicity study that used a
vehicle other than tricaprylin (Christ et al., 1996), maternal toxicity was observed at lower doses
than developmental effects.

Overall, the data suggest that HANs can directly damage DNA as evaluated by a wide
array of assays (summarized in Table V-12). The weight of the evidence varies for the each
compound. BCAN has yielded positive results in all assays tested. DBAN yielded negative
results in S. typhimurium mutation assays and failed to form DNA adducts in vivo. DBAN
appears to induce DNA strand breaks and yield positive results in assays that reflect responses to
DNA damage (i.e. SCE, gene conversion, and SOS assays). The results for DCAN and TCAN
are less consistent. DCAN yielded positive results in S. typhimurium mutation assays and assays
reflecting DNA recombination, but the reason for absence of significant effects in the DNA strand
break assay is not clear. For TCAN, the weak responses in S. typhimurium mutation assays did
not correspond well with the observed in vivo formation of DNA adducts, although the positive
results in the DNA strand break assays and SCE assay were consistent.

EPA/OW/OST/HECD

1-7

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

The evidence for the induction of chromosome damage by HANs is less compelling, due
to the limited number of studies available for evaluation and the inconsistent results. In the single
study that used a standard assay protocol to evaluate induction of micronuclei, no effect was
observed for any of the HANs, although is was not clear that sufficiently high doses were tested.
In contrast, positive results for micronuclei formation were reported for all four compounds in a
less well characterized newt larvae system. DCAN, but not DBAN induced aneuploidy in
Drosophila melanogaster assay system.

The existing data provide at best only marginal support for the conclusion that HANs are
carcinogenic. The evidence is stronger for BCAN, which increased tumor yields in both lung
tumor and dermal screening assays (Bull and Robinson, 1985; Bull et al., 1985). DBAN was
positive at non-ulcerative doses in the dermal screening assay. TCAN was positive only in the
lung assay, and DCAN treatment did not increase either lung or skin tumors. Opposing these
positive findings are the negative results for DBAN, DCAN, and TCAN in the
gamma-glutamyltranspeptidase (GGT) foci assay (Herren-Freund and Pereira, 1986). Overall, the
data are insufficient to qualitatively or quantitatively assess the carcinogenic potential of any of
the HANs. The positive results in two tumor screening assays, together with positive bacterial
gene mutation results, suggest that it would be worthwhile to conduct a full 2-year bioassay for
BCAN. DBAN is currently on test for a full cancer bioassay (NTP, 2002). Results for the other
HANs are more mixed, with inconsistencies between the genotoxicity and tumor screening data.

EPA/OW/OST/HECD

1-8

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

E. Health Effects in Humans

Human epidemiology data on the toxicity of the HANs are lacking. Most of the human
health data for HANs are as components of complex mixtures of water disinfection byproducts.
These complex mixtures of disinfection byproducts have been associated with increased potential
for adverse effects on reproduction (reviewed by Nieuwenhuijsen et al., 2000). Although most
studies of human health effects following exposure to water disinfectant byproducts have used
total trihalomethanes as the exposure metric, Klotz and Pyrch (1999), conducted a case-control
study on the relationship between neural tube defects and drinking water exposure to
trihalomethanes, HANs, and haloacetic acids. The specific compounds that were measured as
part of the total HAN exposure estimate were not identified. Based on the results of the study,
the authors concluded that the HANs did not exhibit a clear association with neural tube defects.

No epidemiological studies have evaluated directly the carcinogenic potential of HANs in
humans. Rather, studies have evaluated the carcinogenic potential of chlorinated versus
unchlorinated drinking water or the presence of trihalomethanes as a marker of chlorination by-
products (IARC, 1999; Mills et al., 1998). Many of these studies have shown an association
between chronic exposure to chlorinated water and increased risks of bladder, rectal, or colon
cancers (Mills et al., 1998; WHO, 2000).

F. Mechanisms of Toxicity and Sensitive Subpopulations

The HANs induce general systemic toxicity. Decreased body weight and a variety of
organ weight changes occur following oral dosing, and the testes (NTP, 2002) and liver might be
particularly sensitive (Hayes et al., 1986), although the reported effects in these organs in

EPA/OW/OST/HECD

1-9

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
available studies are fairly limited. The HANs also induce developmental effects (Smith el al.,
1986; Smithed al., 1987; Smithed al., 1988; Smithed al., 1989; Christen al., 1995; Christen al.,
1996). The mechanism(s) of toxicity are not known, but several possibilities have been described.
HANs may act through direct interactions with cellular macromolecules such as DNA (Daniel el
al., 1986; Lin et al., 1992; Nouraldeen and Ahmed, 1996). HAN toxicity might be secondary to
GSH depletion (Ahmed et al., 1991) or oxidative stress (Ahmed el al., 1999; Mohamadin and
Abdel-Naim, 1999). Formation of cyanide from HAN might be another important mechanism of
toxicity, although important systemic effects that are sensitive indicators of cyanide toxicity have
not been fully examined.

The role of cyanide in the developmental toxicity of HANs has received much attention.
Some studies suggest that metabolites other than cyanide play a critical role (Smith et al., 1986),
and implicated glutathione depletion as an important factor (Christ et al., 1995; Abdel-Aziz el al.,
1993). Although some indirect data supports a role of cyanide (Moudgal et al., 2000; Saillenfait
and Sabate, 2000), evaluation of the available developmental toxicity studies of cyanide itself do
not support this hypothesis (U.S. EPA, 2002c).

The ability of the HANs to bind to cellular macromolecules (Daniel et al., 1986; Lin et al.,
1992; Nouraldeen and Ahmed, 1996), as well as generally positive results in genotoxicity assays,
supports direct DNA damage as the mode of action for the tumorigenicity observed in cancer
screening studies (Bull et al., 1985; Bull and Robinson, 1985). However, the carcinogenic
potential of the HANs is unknown, since epidemiology studies are not available and standard
cancer animal bioassays of HANs have not been conducted.

EPA/OW/OST/HECD	1-10	Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Identification of potential susceptible subpopulations is hampered by the incomplete
characterization of HAN metabolism or identification of the toxic moiety. Although a metabolic
pathway for the HANs has been proposed (Pereira el al., 1984), the enzymes important for
catalyzing HAN metabolism are unknown. In addition, no studies on age-dependent differences
in metabolism or toxicity were identified, although one study demonstrated that HANs may bind
more greatly to fetal DNA than to DNA in maternal tissues (Abdel-Aziz et al., 1993). Analysis of
the developmental toxicity studies for TCAN revealed a lower maternal than developmental
NOAEL, which does not suggest that fetuses are more susceptible than adults.

G. Derivation of the Health Advisories

Health Advisory values (HA) for BCAN, DBAN, DCAN, and TCAN are summarized in
Table 1-1 and the derivation of these values is shown in Chapter VIII. Based on the clear
limitations in the database and gaps in understanding of the mechanisms of toxicity for HANs, the
derived RfD and HA values are best characterized as low in confidence. For BCAN, no suitable
studies were identified for derivation of any HAs. For DBAN, no suitable studies were located
for derivation of a One-day HA. A NOAEL of 12 mg/kg/day for decreased body weight,
decreased testes weight and testes atrophy in male F344 rats in a 14-day drinking water study
(NTP, 2002) was used to derive a Ten-day HA value of 1 mg/L for a 10-kg child. This Ten-day
HA was used as a conservative value for the One-day HA. A NOAEL of 11.3 mg/kg/day was
also identified in a parallel 90-day drinking water study in male F344 rats (NTP, 2002). The
NOAEL value was used to calculate the Longer-term HA value of 0.4 mg/L for a 10-kg child and
1 mg/L for a 70-kg adult. No chronic study of DBAN toxicity was located, so the subchronic

EPA/OW/OST/HECD

1-11

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

(13-week) NOAEL was employed with an extra uncertainty factor to account for extrapolation
from subchronic to chronic exposure to calculate a Life-time HA value of 0.08 mg/L (80 |ig/L).

Table 1-1. Summary of Health Advisory Values for Drinking Water(a)







Longer-Term HA



Chemical

One-Dav HA

Ten-Dav HA

Child

Adult

Life-Time HA

BCAN

_(b)

-

--

-

-

DBAN

1

1

0.4

1

0.08

DCAN

0.4 (0.5C)

0.4 (0.5C)

0.03 (0.04c)

0.09 (0.1c)

0.02 (0.009c)

TCAN

2(2C)

2(2C)

--

--

--

a mg/L

b No value calculated due to lack of suitable toxicological data.
c Value calculated from the study BMDL.

For DC AN, no suitable studies were located for derivation of a One-day HA. A LOAEL
of 12 mg/kg/day for increased relative liver weight in male rats greater than 10% in a 14-day
gavage study (Hayes et al., 1986) was used to derive a Ten-day HA value of 0.4 mg/L for a 10-
kg child. Further analysis of these data yielded a BMDL of 5 mg/kg/day as the critical effect level
for this same effect. When derived on the basis of the BMDL, the Ten-day HA value is 0.5 mg/L.
The Ten-day HA was used as a conservative value for the One-day HA. A LOAEL of 8
mg/kg/day was identified for increased relative liver weight, supported by clinical chemistry
findings at higher doses in a 90-day study in male and female rats (Hayes et al., 1986). Further
analysis of these data yielded a BMDL of 4 mg/kg/day as the critical effect level. The LOAEL
and BMDL were used to calculate the Longer-term HA value. When derived on the basis of the
LOAEL, the Longer-term HA value was 0.03 mg/L for a 10-kg child and was 0.09 mg/L for a 70-

EPA/OW/OST/HECD

1-12

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
kg adult. When derived on the basis of the BMDL, the Longer-term HA value was 0.04 mg/L for
a 10-kg child and 0.1 mg/L for the 70-kg adult. No chronic study of DC AN toxicity was located,
so the subchronic (90-day) LOAEL and BMDL were employed with an extra uncertainty factor
to calculate the Life-time HA values of 0.02 mg/L (based on the LOAEL) and 0.009 mg/L (based
on the BMDL).

For TCAN, no suitable studies were located for derivation of a One-day HA. A NOAEL
of 15 mg/kg/day for absence of a decrease in maternal body weight gain in pregnant rats was used
to derive a Ten-day HA value of 2 mg/L for a 10-kg child. Further analysis of these data yielded
a BMDL of 17 mg/kg/day as the critical effect level. When derived on the basis of the BMDL,
the Ten-day HA value is 2 mg/L. The Ten-day HA was used as a conservative value for the One-
day HA. No suitable subchronic or chronic toxicity data were located for derivation of Longer-
term or Life-time HAs.

Due to lack of adequate dose-response information, calculations of carcinogenic risk have
not been performed for any of the haloacetonitriles. The limited short-term data from the mouse
lung and skin assays, as well as QSTR analyses, indicate that BCAN, DBAN, and TCAN may
have some carcinogenic potential in animals. In addition, data suggest that haloacetonitriles may
induce genotoxicity through direct interactions with DNA. Although the available data provide at
least limited indications of potential carcinogenicity of HANs, these data are not adequate to
demonstrate carcinogenicity in animals. Following EPA's 1986 Guidelines for Cancer Risk
Assessment (U.S. EPA, 1986), BCAN, DBAN, DCAN, and TCAN are appropriately classified as
Group D - Not Classifiable as to Human Carcinogenicity. This classification is appropriate

EPA/OW/OST/HECD

1-13

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

when there is inadequate evidence of carcinogenicity in humans or animals. Following EPA's
Draft 1999 Guidelines for Cancer Risk Assessment (U.S. EPA, 1999), the data for the HANs can
best be described as Data Are Inadequate for an Assessment of Human Carcinogenic Potential.

EPA/OW/OST/HECD

1-14

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Chapter II. Physical and Chemical Properties

Nitriles are organic compounds that contain a cyanogen moiety (-CN) as the characteristic
functional group. Acetonitrile is the compound CH3-CN, and halogenated acetonitriles (HANs)
are compounds of this structure in which one to three halogen atoms (e.g., chlorine or bromine)
are substituted for hydrogen atoms on the methyl carbon.

Four halogenated acetonitriles have been selected for consideration in this document.
These are bromochloroacetonitrile (BCAN), dibromoacetonitrile (DBAN), dichloroacetonitrile
(DCAN), and trichloroacetonitrile (TCAN). These HANs were selected for inclusion in this
document in consideration of the prevalence of individual HANs in drinking water, and the
availability of toxicity data.

Available data on the physical and chemical properties of these compounds are
summarized in Table II-1, and the structural formulas are provided in Figure II-1.

Table II-1. Physical and Chemical Properties of Haloacetonitriles.



Bromochloro-

Dibromo-

Dichloro-

Trichloro-

Property

acetonitrile

acetonitrile

acetonitrile

acetonitrile



(BCAN)

(DBAN)

(DCAN)

(TCAN)

Chemical Abstracts

83463-62-1

3252-43-5

3018-12-0

545-06-2

Registry Services No.









Formula

CHBrClCN

CHBr2CN

CHC12CN

CC13CN

Molecular weight

154.4

198.9

109.9

144.4

Appearance

liquid

liquid

liquid

liquid

Density (g/mL)

1.68

2.30

1.37

1.44

Melting point (°C)

-

-

-

-42

Boiling point (°C)

125-130

67.69

112-113

84.6

EPA/OW/OST/HECD

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Property

Bromochloro-
acetonitrile
(BCAN)

Dibromo-
acetonitrile
(DBAN)

Dichloro-
acetonitrile
(DCAN)

Trichloro-
acetonitrile
(TCAN)

Solubility
Water
Alcohol

soluble

-

soluble

-

Adapted from O'Neil (2001), Lide (1992), Hechenbleikner (1946).

BCAN
Br

2

H	C—

EN

CI

DBAN
Br

H-

EN

Br

DCAN

?'

2

H	C—

EN

CI

TCAN
CI

CI-

EN

CI

Figure II-1. Chemical structures of the haloacetonitriles addressed.

No data were found on commercial uses of the selected haloacetonitriles, except for
TCAN, which has been used as an insecticide (Budavari et al., 1989; O'Neil et al., 2001).

Dihaloacetonitriles (BCAN, DBAN, and DCAN) are reportedly produced during water
chlorination from naturally occurring substances, including algae, fulvic acid and proteinaceous
material (Bieber and Trehy, 1983; Oliver, 1983; Reckhow and Singer, 1990; Reckhow et al.,
1990). Residues of proteinaceous material such as aspartyl residues are a potential source of

EPA/OW/OST/HECD	II-2	Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

dihaloacetonitriles via a stepwise halogenation degradation (Bieber and Trehy, 1983). The ICR
database (U.S. EPA, 2002a) contains extensive information on concentrations of BCAN, DBAN,
DCAN, and TCAN in drinking-water systems, and on how those concentrations vary with input-
water characteristics and treatment methods. These occurrence data are described in detail in
Chapter IV.

Haloacetonitriles may also be formed in vivo following ingestion of chlorinated water.
DCAN was detected in the stomach contents of nonfasted Sprague-Dawley rats following oral
administration of sodium hypochlorite (Mink et al., 1983). The authors attributed the formation
of DCAN to direct chlorination of organic material in the stomach. In a related test from the
same study, Mink et al. (1983) detected DBAN and DCAN in the stomach contents of rats after
oral gavage with NaOCl/KBr solutions.

Bieber and Trehy (1983) reported that DHANs undergo hydrolysis in water to nonvolatile
products. Half-lives of dihaloacetonitriles in water are presented in Table II-2.

Table II-2. Half-Lives of Dihaloacetonitriles in Water at Several pH Values (25 °C).

Half-life (hours)

Compound pH 7.4 pH 8.3 pH 9.0 pH 9.77

BCAN

-

35

-

—

DBAN

500

85

19

—

DCAN

-

30

-

0.75

Adapted from Beiber and Trehy (1983).

EPA/OW/OST/HECD

II-3

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
Chapter III. Toxicokinetics

Limited data are available on the toxicokinetics of the haloacetonitriles (HANs). No data
were located on the absorption or distribution of BCAN following oral exposure. A comparative
toxicokinetics and metabolisms study in mice and rats has been conducted for DBAN (NTP,
2002). However, this study was not available for review at the time this document was prepared.
No studies were located on the absorption, distribution, metabolism, or excretion of any of the
HANs following inhalation or dermal exposure, although some qualitative information can be
inferred from toxicity studies.

A. Absorption

Roby et al. (1986) administered single oral gavage doses of either [1-14C]- (labeled on the
cyanide group) or [2-14C]-DCAN (labeled on the dichloromethyl group) in water to male F344
rats and B6C3F1 mice. The administered doses were 0.2, 2, or 15 mg/kg for the rats and 2 or 15
mg/kg for the mice. Appearance of label in feces, urine and expired air was monitored until at
least 70% of the radioactivity had been recovered. The amount of time required to recover 70%
of the administered radioactivity differed across species. In rats, the collection of data continued
for 6 days for [1-14C]- DCAN and for 2 days for [2-14C]-DCAN. In mice, collection was
terminated at 24 hours for both positions of radiolabel, since at least 70% of the dose had been
collected. These results indicate that DCAN in water is well absorbed (at least 80% to 90%) from
the gastrointestinal tract, since only 8% to 20% of the total dose was excreted in feces. The rate
of absorption was not determined, and data on the blood concentrations of radiolabel over time

EPA/OW/OST/HECD

III-l

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
were not reported. The appearance of radiolabel in urine and exhaled air at 24 hours suggests,
however, that DCAN was absorbed rapidly.

Roth et al. (1990), in a published abstract, reported on a study designed to test whether
differential absorption and distribution kinetics of TCAN in tricaprylin versus corn oil was
responsible for observed differences in developmental toxicity studies. Pregnant rats (strain and
number not specified) were administered a single oral gavage dose of 55 mg [uC]-TCAN/kg in
either tricaprylin or corn oil on gestation day 10 or 11. Levels of radiolabel were followed in the
maternal stomach and intestinal contents, blood, liver, spleen, heart, adipose tissue, and embryonic
tissue. Maximal blood and tissue levels were observed 4 to 6 hours following exposure,
indicating rapid absorption kinetics. The choice of solvent vehicle did not affect the absorption
kinetics. No data on the degree of absorption were provided in this published abstract. A
published version of this study was not located, and the data presented did not allow an
independent verification of the results.

The lethality observed in acute dermal toxicity tests for BCAN (Eastman Kodak Co.,
1992) and TCAN (Smyth et al., 1962) indicates that systemic exposure to these compounds
occurred, demonstrating that HANs can be absorbed dermally. These studies are not adequate to
estimate the rate and degree of absorption by the dermal route.

In summary, the existing data suggest that HANs can be absorbed following either oral or
dermal administration. HANs are rapidly absorbed following oral administration, based on the
observed peak blood concentrations 4 to 6 hours after dosing reported by Roth et al. (1990). The

EPA/OW/OST/HECD

III-2

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
strength of this conclusion is limited, however, since only a published abstract is available. The
degree of absorption following oral dosing is nearly complete, based on the small fraction of the
administered radioactivity observed in feces (Roby et al., 1986). HANs can be absorbed through
the skin, but the existing data are insufficient to estimate the rate or degree of absorption. No
data are available to determine whether HANs can be absorbed following inhalation exposure.

B. Distribution

Roby et al. (1986) studied the tissue distribution of radiolabel following administration of
single oral gavage doses of 0.2 to 15 mg/kg of [1-14C]- or [2-14C]-DCAN in water to rats and
mice. Daily excreta, including exhaled volatile organic compounds and C02, were analyzed until
at least 70% of the radioactivity was recovered. The amount of time required to recover 70% of
the administered radioactivity differed across species: 6 days following oral administration of
[1-14C]- DCAN and 48 hours for [2-14C]-DCAN in rats, and 24 hours regardless of the position of
the radiolabel in mice. After at least 70% of the administered dose had been excreted, the animals
were sacrificed and tissues were collected. Label was detected in all tissues tested (see Table III-
1), although the residual tissue levels represented a small portion of the administered dose. Six
days after oral administration of [1-14C]-DCAN to rats, the tissue distribution of the label as
percent of the administered dose was: blood (4.1-7.9%), muscle (3.9-7.9%), skin (3.3-6.3%) and
liver (1.9-2.6%)) for the three dose groups. For [2-14C]-DCAN, the liver retained the largest
amount of radiolabel (approximately 5% of the administered dose 2 days after treatment),
followed by muscle (2.7-4.8%), blood (2-4.6%) and skin (0.9—1.0%). Most other tissues
contained less than 1% of the dose. These tissue distribution data for rats are presented in Table
III-1. In mice, the tissue distribution did not differ greatly between the alternately labeled

EPA/OW/OST/HECD

III-3

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

compounds. The largest amount of radioactivity was present in the liver, 3.5-4.3% of the
administered dose for [1-14C]-DCAN and 5.1-5.4% for [2-14C]-DCAN. The muscle and skin also
contained appreciable amounts of radioactivity as shown in Table III-2.

EPA/OW/OST/HECD

III-4

Final Draft

-------
m

>
o

o

C/2

H

O

a

S3

P

Dose

0.20 mg/kg	 	2.0 mg/kg	 	15 mg/kg

Tissue

d-14c)b

14 c
(2- C)

(i-14c)b

(2-lV

1

¦o
o

(2-UC)C

Blood

4.1211.36

2.47±0.19

7.8911.01

2.02+0.60

7.2211.44

4.5711.99

Liver

2.13±0.33

5.70+0.45

2.6010.35

5.23+0.32

1.8810.34

5.6710.54

Muscle

7.88±1.68

4.78±0.51

6.88+0.76

2.7410.96

3.88+0.47

3.3711.07

Skin

6.13±0.24

1.61±0.06

6.2510.92

0.9410.02

3.3210.37

1.62+0.24

Kidney

0.37+0.01

0.47±0.03

0.4410.04

0.37+0.01

0.35+0.05

0.47+0.04

Adipose

0.63+0.08

0.93+0.27

0.57+0.06

0.3510.12

0.4810.08

0.3110.02

Gastrointes t inal

1.91±0.28

0.7610.09

1.60+0.17

0.5910.04

0.97+0.24

0.7110.13

Brain

0.05±0.01

0.0810.00

0.06+0.01

0.0510.01

0.6010.01

0.0810.02

Lung

0.17±0.04

0.10+0.02

0.20+0.06

0.0710.01

0.1410.02

0.1210.01

Testes

0.23+0.06

0.12±0.01

0.2710.05

0.0710.01

0.1410.03

0.1010.01

Total tissue

23.62

17.02

26.76

12.43

18.98

17.02

retention

3Values are expressed as mean percentage of administered dose of	equivalents (±S D

= 3).

Animals sacrificed six days after dosing.

Animals sacrificed two days after dosing.

Adapted from Roby et al. (1986).

-------
Drinking Water Criteria Document for Haloacetonitriles

Table III-2. Tissue Levels of DCAN One Day After Oral
Administration to Mice3.

Dose



2.0 mg/kg

15 i

mg/kg

Tissue

(i-14c)

(2-UC)

(1-,4C)

(2-UC)

Blood

1.11±0.28

0.29±0.07

0.95+0.19

0.27±0.08

Liver

3.4810.43

5.12±1.14

4.25±0.35

5.37±0.43

Muscle

2.11+0.53

1.03±0.06

1.56±0.11

0.79±0.21

Skin

2.02+0.55

0.45+0.04

1.69+0.21

0.52+0.09

Kidney

0.38+0.10

0.56±0.15

0.46±0.02

0.50±0.04

Adipose

1.22±0.55

0.62±0.05

0.89±0.23

0.38+0.11

Gastrointestinal

1.46+0.27

0.65±0.12

1.59±0.27

0.80±0.26

Brain

0.04+0.02

0.04±0.01

0.03±0.01

0.03±0.01

Lung

0.05±0.03

0.04±0.01

0.07+0.01

0.03±0.01

Testes

0.05±0.03

0.02+0.00

0.04±0.02

0.05±0.05

Total tissue
retention

11.95

8.37

11.53

8.74

Y^lues are expressed as mean percentage of administered dose of
C equivalents (±S.D., N = 3).

Adapted from Roby et al. (1986).

EPA/OW/OST/HECD

III-6

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

These data do not provide clear evidence for significant differences in distribution for the dimethyl
and cyanide carbons, since only small differences were apparent in the rat study, and no
difference was observed for the alternatively labeled compounds in mice.

A study to assess the potential of TCAN to form protein and DNA adducts provides
qualitative evidence for wide tissue distribution of HANs. Lin el al. (1992) administered single
oral gavage doses ranging from 7.2 to 69.3 mg/kg of either [1-14C]- or [2-14C]-TCAN in
tricaprylin to male F344 rats. DNA was isolated from liver, stomach, and kidney, and several
proteins were isolated from blood. The tissues were analyzed from 4 to 48 hours following
dosing. More radiolabel was associated with DNA when the trichloromethyl carbon [2-14C] was
labeled than when the cyanide group carbon [1-14C] was labeled. The study authors hypothesized
that the adducts resulted from the reaction of DNA with single carbon metabolites formed by the
cleavage of TCAN. DNA binding was highest in the stomach, followed by liver and kidney.
Adducts with globin, albumin, and globulins were also identified, and similar levels were observed
with the label at either position. In addition, the HPLC elution profiles for the radioactivity was
the same regardless of the position of the label, leading the authors to suggest that the protein
adducts are formed from either "2-carbon metabolites (unsplit) or by direct reaction with TCAN."
Adduct studies provide only limited information on tissue distribution, since macromolecular
binding may also be dependent on metabolism of the compound. However, the observed
formation of serum protein adducts and the appearance of DNA adducts in all three tissues
measured suggests wide distribution of TCAN.

EPA/OW/OST/HECD

III-7

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Additional studies have been conducted to address the potential for solvent vehicle to alter
the distribution of TCAN and have been reported in two published abstracts. Roth el al. (1990)
administered a single dose of 55 mg [uC]-TCAN/kg to pregnant rats (strain and number not
specified) in either tricaprylin or corn oil on gestation day 10 or 11. The radiolabel (the identity
of the labeled carbon was not specified) was followed for 48 hours in the maternal stomach and
intestinal contents, blood, liver, spleen, heart, and adipose tissue, and in embryos. In the blood,
most counts were bound to red blood cells, and in the plasma up to 50% of the radiolabel was
protein-bound. Tissue levels were highest in the liver (approximately 40 |ig TCAN equivalents/g
at 6 hours post-exposure). Radiolabel was also detected in embryos. According to the study
authors, there were no solvent-related differences reported for any of the toxicokinetic parameters
evaluated for TCAN when administered in tricaprylin versus corn oil.

In a follow up study to assess the impact of repeated dosing on the effect of solvent
vehicle that was reported in a published abstract, Gordon et al. (1991) administered one, two, or
three successive daily doses (dose not specified) of [1-14C] or [2-14C]-labeled TCAN in tricaprylin
or corn oil to groups of pregnant rats (strain not specified) in mid-gestation. The animals were
sacrificed on gestation day 13 and maternal and embryo levels of radiolabel were evaluated. The
14C levels in the embryos from the tricaprylin groups were described as much greater than in the
corn oil vehicle group following 3 daily doses, but no quantitative estimate was reported. After
three doses, maternal blood 14C levels were higher for the tricaprylin vehicle group compared to
the corn oil group. The abstract did not report the effects of solvent vehicle on the embryo or
maternal blood levels of TCAN-associated radioactivity after one or two daily doses, precluding
an analysis of trends in the relationship between the number of days of dosing and tissue

EPA/OW/OST/HECD

III-8

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

accumulation. In the maternal liver, 14C accumulation of both [1-14C] and [2-14C] was greater in
the tricaprylin group than in the corn oil group after two daily treatments, but after three doses
with [2-14C]-labeled TCAN, accumulation was higher from corn oil. While solvent-related
differences in the accumulation of radiolabel were reported in this study, the degree of difference
between solvent vehicle groups was not provided for any of the findings and apparent
inconsistencies observed across the dosing regimens were not adequately explained. Based on
inconsistent results, the absence of quantitative data, and the lack of peer review, these data
should be viewed as preliminary. In addition, the higher embryonic accumulation of TCAN with
tricaprylin in this study appears to be inconsistent with the results of Roth et al. (1990) with
TCAN in tricaprylin and corn oil, although the latter study used a shorter dosing regimen.
Therefore, it remains unclear if solvent vehicle affects tissue distribution. Nevertheless, both
abstracts provide qualitative evidence for wide tissue distribution of TCAN, in support of the
better-documented study on DCAN by Roby el al. (1986).

In summary, the two compounds tested, DCAN and TCAN, are widely distributed
following oral dosing. Radiolabeled parent compound or metabolites have been identified to
varying degrees in blood and a host of tissues, including in embryos, with no single tissue
dominating HAN uptake. The role of solvent vehicle on distribution of HANs also remains
unresolved, with contradictory results being reported within a single published abstract (Gordon
et al., 1991). (See Chapter VII for additional analysis of potential solvent vehicle effects.) No
data were identified to evaluate distribution of HANs following dermal or inhalation exposure.

EPA/OW/OST/HECD

III-9

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

C. Metabolism

Pereira et al. (1984) administered a single gavage dose of 0.75 mmol/kg BCAN (116
mg/kg), DBAN (149 mg/kg), DCAN (82 mg/kg), or TCAN (108 mg/kg) in tricaprylin to male
Sprague Dawley rats. Urinary thiocyanate, the only metabolite measured, accounted for 2.25% to
12.8% of the dose by 24 hours. The excretion of thiocyanates in the urine was in the order of
BCAN>DCAN>DBAN>TCAN, The authors also measured the effects of the HANs on
dimethylnitrosamine demethylase (DMN) activity (a measure of CYP2E1 activity) as a marker of
protein binding. Based on dose-responses from in vitro incubations with rat liver microsomes,
DBAN and BCAN were more potent inhibitors of this enzyme than DCAN or TCAN. In contrast
to the in vitro results, a single gavage dose of 0.75 mmol/kg TCAN, but not 0.75 mmol/kg DBAN,
significantly inhibited DMN activity in liver microsomes of rats sacrificed 3 or 18 hours after
dosing. Results for other HANs were not presented. The mechanism of inhibition was considered
to be noncompetitive or uncompetitive (noncompetitive inhibition is characterized by an altered
ratio ofK^V^ with a decrease in Vmax, while uncompetitive inhibition is characterized by a
constant ratio of	with a decrease in the Vmax). The results suggest that HANs are not

competing for the active site of CYP2E1. Although Pereira et al. (1984) suggest that oxidative
dehalogenation is an initial step in the metabolism of HANs, the data are inadequate to identify the
specific enzymes involved. Based on their results and earlier work on formation of cyanide from
nitrile compounds presented by Silver et al. (1982), Pereira et al. (1984) proposed a metabolic
scheme to explain the formation of thiocyanates from HANs (Figure III-l).

EPA/OW/OST/HECD

111-10

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Figure III-l Proposed Metabolism of Haloacetonitriles

Figure Legend. Proposed metabolic pathways for HANs. Two distinct pathways are proposed:
conjugation with glutathione and oxidative metabolism. The glutathione pathway is shown with dashed lines
to indicate that direct identification of these conjugates or other downstream metabolites has not been
demonstrated in vivo, as described further in the text. It is not clear if the proposed glutathione conjugation
is catalyzed by glutathione-V-transfcrases (GST) or is nonenzymatic. For the oxidative pathway,
dehalogenation is thought to be catalyzed by CYPs, although the isoforms that mediate this reaction have
not been identified.

1	2

5

Intermediate metabolites labeled in the figure with numbers 1 through 6 are as follows: 1) haloacetonitrile;
2) halocyanomethanol; 3) haloformaldehyde; 4) cyanoformaldehyde; 5) halocyanoformaldehyde; 6)
unidentified glutathione conjugates. Presentation of the oxidative metabolism pathway adapted from Pereira
etal. (1984).

EPA/OW/OST/HECD

III-11

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

As shown in this figure, metabolism of HANs is hypothesized to occur by oxidative
dehalogenation to yield halocyanomethanols. These reactions are catalyzed by mixed-function
oxidases such as cytochrome p450s (CYP), although the identity of the isozyme(s) that carries out
the individual reactions has not been determined. The halocyanomethanols are proposed to then
dehydrate to form halocyanoformaldehydes or lose cyanide to form haloformaldehydes, including
phosgene. The cyanoformaldehydes undergo further oxidative metabolism to form C02 and
cyanide (which can then be further metabolized to thiocyanate).

Pereira et al. (1984) did not measure the intermediate metabolites proposed in their
metabolic pathway. However, a study by Roby et al., (1986) provides additional indirect evidence
for the oxidative metabolism of HANs. Roby et al. (1986) administered single oral doses of 0.2, 2,
or 15 mg/kg [1-14C]- or [2-14C]-DCAN to rats, and 2 or 15 mg/kg of these compounds to mice. In
both rats and mice, labeling of the cyanide group carbon resulted in higher amounts of radiolabel in
urine than in expired air (i.e., as C02), while radioactivity excreted via both routes was nearly equal
when the dichloromethyl group was labeled. Marginal dose-dependent changes in the amount of
the radiolabel recovered in feces and expired air were reported, but the pattern of these changes
were not consistent across species or with position of the radiolabel, making the findings difficult
to interpret. The pattern of label distribution in tissues and excreta indicated to the authors that
DCAN would be oxidized to dichlorocyanomethanol, consistent with the metabolic scheme
proposed by Pereira et al. (1984). Dichlorocyanomethanol could then either dehydrate to
chlorocyanoformaldehyde or lose cyanide to form phosgene, leading to terminal degradation
products including chlorine, formic acid, C02 and cyanide. The authors did not, however, directly
measure these metabolites.

EPA/OW/OST/HECD

III-12

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Conjugation with glutathione (GSH) also appears to be an important pathway in the
metabolism of HANs as shown in Figure III-1. Lin and Guion (1989) investigated the ability of
BCAN, DBAN, DCAN, and TCAN to interact with GSH and glutathione-s-transferases (GST) in
a series of in vitro and in vivo experiments. The in vitro experiments tested the effect of various
incubation conditions on the direct reactivity of HANs with GSH as measured by the loss of GSH
in the incubation mixture. Direct incubations (in the absence of GST) revealed that HANs have the
potential to bind to GSH. The relative reactivity toward GSH was DBAN>BCAN»TCAN. No
detectable binding with DCAN was observed. Addition of bovine serum albumin to the
incubations reduced the degree of GSH removal by TCAN, but had no modifying effect on DBAN,
suggesting that at least for DBAN, binding to GSH is somewhat specific (no results for BCAN
were reported). The presence of cytosol (a cell fraction containing GST activity) did not alter
GSH loss by DBAN or TCAN (no results for DCAN or BCAN were reported). Therefore, for at
least these two HANs, GSH conjugation appears to be nonenzymatic. The presence of
microsomes decreased GSH loss with DBAN and TCAN, leading the authors to suggest that
microsomal metabolism of these HANs results in the formation of metabolites that are less GSH-
reactive than the parent compounds. No results were presented for incubation of BCAN with
microsomes. Even though the presence of cytosol alone did not alter GSH loss, incubation of both
cytosol and microsome with DBAN or TCAN decreased GSH loss compared to that observed
with microsomes alone. Taken together, these results show that the GSH reactivity of HANs varies
greatly among the HANs under review in this document. These in vitro findings suggest that
BCAN, DBAN, and TCAN appear to react with GSH in a nonenzymatic fashion as the parent
compounds, while DCAN shows little propensity for binding to GSH.

EPA/OW/OST/HECD

III-13

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

In a second part of this study, Lin and Guion (1989) tested the ability of HANs to inhibit
GST activity (as measured by l-chloro-2,4-dinitrobenzene (CDNB) conjugation with GSH) in
vitro and in vivo. All the HANs decreased GST activity in vitro in the order of
TCAN>BCAN=DBAN>DCAN, with a 4-fold difference separating the level of GST activity in
the presence of TCAN and DCAN, For the in vivo studies, male Fischer 344 rats were
administered single gavage doses of 0.75 mmol/kg BCAN (116 mg/kg), DBAN (149 mg/kg),
DCAN (82 mg/kg), or TCAN (108 mg/kg) in tricaprylin. These doses represented 10 to 30% of
reported LD50 for the individual compounds. Liver GST activities and GSH concentrations were
measured 1,3, and 18 hours after dosing. GST activity was slightly decreased at 3 hours by
DBAN and TCAN, and was significantly decreased by DBAN at 18 hours (data not shown). The
authors noted that there are several potential mechanisms for inhibition of GST activity by HANs,
including depletion of GSH through direct conjugation, by competing with GSH for GST catalytic
sites (competitive inhibition), or through noncompetitive protein binding to GST. To test this first
possibility, the authors also measured liver GSH levels following in vivo administration of HANs
using the same dosing protocol as for the GST activity measurements. For BCAN, DBAN, and
DCAN, GSH levels were decreased at 1 hour post-treatment, recovered to control levels by 3
hours, and were elevated at 18 hours. For TCAN, liver GSH levels were unchanged at 1 hour, and
were elevated at 3 and 18 hours post-treatment. These results show that initial decreases in GSH
levels are transient, and that GSH levels return to control levels shortly after cessation of exposure,
with a rebound to higher GSH levels within a day post-treatment.

The effects of HANs on GST and GSH levels have also been investigated by Ahmed and
colleagues (Ahmed et al., 1989; Ahmed et al., 1991). In an in vitro study, DBAN, DCAN, and

EPA/OW/OST/HECD

III-14

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
TCAN significantly inhibited GST activity (CDNB conjugation with GSH). IC50 (the
concentration of inhibitor resulting in 50% inhibition) values for these three HANs were 0.82,
2.49, and 0.34 mM, respectively. This result suggests that TCAN binds more readily to GST than
DBAN or DCAN, The observed inhibition of GST was reversible upon dialysis of the enzyme,
suggesting reversible binding of HANs to GST (or other mechanisms not involving direct protein
reactivity). Incubation with DBAN or DCAN decreased both the apparent Iv and the Vmax of
GST activity toward GSH, while incubation with TCAN increased the apparent Iv and Vmax.

Based on the patterns of activity of GST toward GSH, kinetic interactions were described as
intermediate between uncompetitive and noncompetitive for DBAN, as uncompetitive for DCAN,
and as competitive for TCAN. The effect of HANs on GST activity toward its substrate CDNB
was also evaluated. The pattern of HAN inhibition of GST-dependent conjugation of CDNB was
described as mixed by the study authors (i.e., showing aspects of competitive, uncompetitive, and
noncompetitive inhibition). These complex patterns of inhibition suggest that HANs might interact
with GST through multiple mechanisms. For example, HANs might interact at the catalytic site of
the enzyme (consistent with competitive inhibition) as well as other protein sites (consistent with
uncompetitive, and noncompetitive inhibition).

In an in vivo study by Ahmed and colleagues (1991), GST activity and GSH levels were
determined in time-course and dose-response experiments with DBAN. For the time-course
experiment, male Sprague Dawley rats were administered a single oral dose of 75 mg/kg DBAN in
dimethyl sulfoxide (75% of the LD50), and aortal blood, liver, stomach, and kidney were harvested
at 0.5, 1, 2, or 4 hours after treatment. For the liver, GSH levels were significantly (p<0.05)
decreased at 0.5 hours to roughly 60% of controls (as read from a figure in the paper), recovered

EPA/OW/OST/HECD

III-15

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
to control levels by 2 hours, and were increased above control levels at 4 hours. The mild increase
at 4 hours was not statistically significant. For the stomach, GSH levels were nearly completely
depleted by 0.5 hours (6% of control levels), and remained decreased for up to 4 hours. No
significant change in blood or kidney GSH levels was detected. The effects of DBAN on GST
activity closely paralleled the effects on GSH levels, with significant decreases beginning at 0.5
hours for the liver and stomach, and no significant effect in the kidney.

For the dose-response experiment, Ahmed et al. (1991) administered single oral gavage
doses of 0, 25, 75, or 100 mg/kg DBAN in dimethyl sulfoxide to male Sprague Dawley rats and
harvested blood and tissues 0.5 hours after dosing as described for the time-course experiment.
Hepatic and gastric GSH levels were decreased in a dose-dependent fashion, and were significantly
(p<0.05) decreased beginning at 25 mg/kg. For the liver, GSH levels were decreased by 23% at
25 mg/kg, by 38% by 50 mg/kg, by 46% at 75 mg/kg, and by 57% at 100 mg/kg. GSH levels in
the stomach tissue were decreased by 43% at 25 mg/kg, by 75% at 50 mg/kg, by 84% by 75
mg/kg, and 86% by 100 mg/kg. No significant effect on blood or kidney GSH levels was
observed. Liver and stomach GST activities were also decreased in a dose-dependent fashion, but
this effect was less severe than the decreases in GSH levels. The degree of inhibition (enzyme
activity decreased to 60% of control levels in the liver, and decreased to 71% of control levels in
the stomach) was statistically significant (p<0.05) beginning at 50 mg/kg. The authors suggested
that the likely mechanism for GSH depletion was direct conjugation with HANs due to their
electrophilic nature. They further noted that the depletion of GSH, coupled with protein binding
to the GST enzyme itself, could lead to the observed inhibition of GST activity.

EPA/OW/OST/HECD

III-16

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

The results of Lin and Guion (1989) and Ahmed et al. (1989; 1991) suggested that HANs
inhibit liver GST activity. However, NTP (2002) reported an increase in liver GST activity in male
F344 rats following exposure to DBAN in their drinking water for 14 days. The increase of 126%
over controls was only significant in the male rats exposed to drinking water containing 200 mg/L
DBAN (18 mg/kg/day). One possible explanation for the opposite effects of HANs on GST
activity reported among these studies is the duration of dosing that was employed in each study.
The studies by Lin and Guion (1989) and Ahmed et al. (1989; 1991) were single dose gavage
studies. Even in these studies a rebound in GSH levels or GST activity was noted within a period
of hours that often exceeded control levels. Therefore, it is possible that longer-term exposure
(such as in the 14-day study) enhances GST activity due to a rebound effect after an initial
decrease. A time course experiment measuring GSH levels and GST activities over acute and
subchronic periods would be needed to determine if this is the case.

The results of Lin and Guion (1989), Ahmed et al. (1989; 1991), and NTP (2002) suggest
that conjugation with GSH may be an important source of HAN detoxification in animal toxicity
studies. The in vitro results suggest that BCAN, DBAN, and TCAN are conjugated with GSH in
a non-enzymatic fashion, and at least for DBAN, this interaction is somewhat selective. The
decreases in GSH levels following oral dosing of rats with HANs further supports the in vitro
findings. In in vitro experiments HANs also appear to inhibit GST activity, although a rebound
effect after longer periods of exposure could be possible. Multiple mechanisms are likely involved
in the observed inhibition following acute dosing. It is noteworthy that the concentrations of
HANs used in these studies to demonstrate GSH depletion and altered GST activity are orders of
magnitude greater than measured human exposures to these compounds in drinking water. Since

EPA/OW/OST/HECD

III-17

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

GSH depletion is clearly dose-dependent (Ahmed et al., 1991), the importance of this pathway in
human exposure situations is probably minimal.

D. Excretion

Pereira et al. (1984) studied urinary excretion of thiocyanate in rats following single oral
doses of 0.75 mmol/kg of several HANs in tricaprylin. The percentages of the administered dose
excreted as thiocyanate after 24 hours were 12.8%, 7.67%, 9.28%, and 2.25% of BCAN, DBAN,
DCAN, and TCAN, respectively. No data were presented for other urinary metabolites; thus the
total contribution of urinary excretion cannot be estimated. This issue was, however, more fully
evaluated in the kinetics study of Roby et al. (1986), in which single oral doses of [1-14C]- or
[2-uC]-DCAN were administered in water to male rats and mice. Urine, feces, and expired air
were collected and the radioactive content measured until at least 70% of the label had been
recovered. In rats, this required 6 days for [1-14C]-DCAN and 2 days for [2-14C]-DCAN. In mice,
excretion was more than 70% complete within 24 hours for both locations of the label. In both
animal species, the label of [1-14C]-DCAN was excreted mostly in the urine (42-70%), with lower
amounts in feces (9-20%, some of which may have been unabsorbed) and expired air (3—8%). For
[2-'4C]-DCAN, label was excreted both as carbon dioxide in air (33—37%) and in urine (35—43%),
with 8-13%) in feces (Shown in Table III-3 measured as radiolabel excretion).

EPA/OW/OST/HECD

III-18

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Table III-3. Excretion of DCAN in Rats and Mice.



Percentage of total dose

excreted

Excretion



-DCAN

[2-1*

C]-DCAN

route

Rat

Mouse"

RatC

Mouse"

Urine

42-45

64-70

35-40

42-43

Feces

14-20

9-13

10-13

8-11

Expired air

3-8

5-6

33-34

31-37

Total excreted

62-73

83-85

82-86

84-88

^The period of collection was 6 days.
cThe period of collection was 24 hours.
The period of collection was 2 days.

Adapted from Roby et al. (1986).

Based on these data, excretion of HANs is nearly complete over a period of days. The rate
of excretion may differ across species, since mice excrete DCAN more rapidly than rats.
Differences in excretion of thiocyanate for different HANs was observed by Pereira et al. (1984),
with TCAN being excreted as thiocyanate to a lesser degree than the other HANs. It is not clear if
this reflects differences in metabolism or differences in excretion, since total urinary excretion of
the radiolabel was not determined. The dispensation of the two carbons also differs (Roby et al.,
1986). The cyanide group tends to be more readily excreted in the urine and the halomethyl
carbon is excreted nearly equally in expired air and in urine. Excretion in the feces appears to be
limited.

EPA/OW/OST/HECD

III-19

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

E. Bioaccumulation and Retention

No studies were located that provided data on long-term accumulation and retention of
BCAN, DBAN, DCAN, or TCAN in the body. The existing kinetic studies did not determine half-
lives and were not conducted for sufficiently long periods to evaluate long-term accumulation.
However, the results of Roby et al. (1986) that showed relatively rapid excretion of DCAN-
associated radioactivity (at least 70% of the administered dose excreted within 6 days in rats or
within 24 hours in mice) suggests limited potential for the bioaccumulation of the HANs.

F. Summary

Limited data are available on the toxicokinetics of the HANs, with a comprehensive
toxicokinetic study for oral dosing available only for DCAN. However, the existing toxicokinetic
data suggest that HANs can be rapidly and nearly completely absorbed following oral dosing
(Roby et al., 1986; Roth et al., 1990). Systemic toxicity data suggest that HANs are absorbed by
the dermal route. Once absorbed, HANs appear to be widely distributed. The two compounds
tested, DCAN (Roby et al., 1986) and TCAN (Lin et al., 1992), were widely distributed following
oral dosing, with no clear preferences in tissue distribution apparent based on the limited data. No
data were available on tissue-dependent metabolism, but an overall metabolic scheme for HANs
involving an initial oxidative dehalogenation step has been proposed based on the propensity for
these compounds to form cyanide and metabolism studies for other nitriles (Pereira et al., 1984).
Proposed intermediate metabolites have not been measured directly, and the identity of enzymes
responsible for steps in the pathway have not been identified. Conjugation with GSH, at least at
high doses, might be a second important route of metabolism for HANs (Ahmed et al., 1989; Lin
and Guion, 1989; Ahmed et al., 1991; NTP, 2002). Excretion of HANs is nearly complete over a

EPA/OW/OST/HECD

111-20

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
period of days, largely in urine and in exhaled air. The rate of excretion may differ across species,

since mice excrete DCAN more rapidly than rats (Roby et al., 1986). Differences in urinary

excretion of thiocyanate for different HANs was observed by Pereira et al. (1984), with TCAN

being excreted as thiocyanate to a lesser degree than the other HANs. The results of Roby et al.

(1986) that showed relatively rapid excretion of DCAN-associated radioactivity suggests limited

potential for the bioaccumulation of the HANs. However, no studies were located that provided

data on long-term accumulation and retention of any of the HANs.

EPA/OW/OST/HECD

111-21

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
Chapter IV. Human Exposure

A. Drinking Water Exposure

BCAN, DBAN, DCAN, and TCAN have been identified as drinking-water disinfection
byproducts under the Information Collection Rule (U.S. EPA, 1994) and are being assessed for
regulatory consideration in the Stage 2 Disinfectants/Disinfection Byproducts Rule to be
promulgated. Therefore, this section will examine the occurrence of these compounds in drinking
water.

A.l National Occurrence Data for BCAN, DBAN, DCAN, and TCAN

This section presents the data collected from the Information Collection Rule (ICR)
database, which provides data from surface- and ground-water systems serving at least 100,000
persons. This database includes information gathered for 18 months from July 1997 to December
1998.

Section A. 1.1 describes the ICR data set and analysis techniques used to present the data
for the plants that submitted data under the ICR. The data in Sections A. 1.1 and A. 1.2 were taken
from the online version of the ICR database (U.S. EPA, 2002a), and the explanation of the
methods used was taken from the Draft EPA Document on Stage 2 Occurrence and Exposure
Assessment for Disinfectants and Disinfection Byproducts (D/DBPs) in Public Drinking Water
(U.S. EPA, 2000a).

EPA/OW/OST/HECD

IV-1

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

A.1.1 ICR Plants

The ICR generated plant-level sets of data that link water quality and treatment from
source to tap, and aid in understanding the seasonal variability in these relationships. The database
contains information from 18 monthly or six quarterly samples from 7/97 to 12/98 from
approximately 300 large systems covering approximately 500 plants. These samples were tested
for influent and finished water-quality parameters (e.g., TOC, temperature, pH, alkalinity), DBP
levels, and disinfectant residuals. Samples were collected at several locations throughout the
distribution system to cover the entire range of residence times during which DBPs can form in
the finished water. Over the 18-month period, approximately 1470 samples were taken from 305
plants with surface water as their source, and approximately 580 samples were taken from 123
plants with groundwater as their source. For more detailed information, such as sampling
locations and frequencies, refer to the ICR Data Analysis Plan (U.S. EPA, 2000b).

A. 1.2 Quarterly Distribution System Average and Highest Value for BCAN, DBAN,

DCAN, and TCAN

This section describes the data-analysis techniques employed for the analysis of observed
data for water-quality parameters, and for BCAN, DBAN, DCAN, and TCAN concentrations. All
data are categorized according to the types of source water - surface or ground. Plants having
both surface- and ground-water sources (mixed) or that purchase water are included in the
surface- water category. Quarterly Distribution System Average and Highest Value for the HANs
are presented in Table IV-1. Data presented in the table have been taken from the ICR database

EPA/OW/OST/HECD

IV-2

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
as provided to avoid misrepresentation or misinterpretation. Therefore, although all data in the

table are presented with two decimal points (as provided in the ICR database), this does not

necessarily represent the actual precision of the data.

The quarterly distribution-system average is an average of the following four distinct
locations in the distribution system.

•	Distribution System Equivalent (DSE) location;

•	Average 1 (AVG 1) and Average 2 (AVG 2) locations: Two sample points in the
distribution system representing the approximate average residence time as
designated by the water system; and

•	Distribution System Maximum: Sample point in the distribution system having the
highest residence time (or approaching the longest time) as designated by the
water system

The quarterly distribution-system highest value is the highest of the four distribution-
system samples collected by a plant in a given quarter.

EPA/OW/OST/HECD

IV-3

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Table IV-1. Haloacetonitriles
Quarterly Distribution System Average and Highest Value

Source

Quarterly
Dist. Sys.

Plants

N

PctND

%

Mean
jig/L

Median
jig/L

STD
jig/L

Min
jig/L

Max
jig/L

plO
jig/L

p90
jig/L

BCAN

SW

Average

304

1411

20.55

1.14

0.88

1.21

0.00

13.13

0.00

2.63



High

304

1411

20.55

1.40

1.05

1.41

0.00

13.40

0.00

3.20

GW

Average

108

524

50.38

0.73

0.00

1.10

0.00

7.38

0.00

2.13



High

108

524

50.38

0.99

0.00

1.48

0.00

13.00

0.00

2.70

DBAN

SW

Average

304

1397

43.88

0.75

0.27

1.12

0.00

7.78

0.00

2.25



High

304

1397

43.88

0.96

0.60

1.36

0.00

11.30

0.00

2.70

GW

Average

108

523

37.86

0.82

0.43

1.17

0.00

8.93

0.00

2.15



High

108

523

37.86

1.10

0.80

1.39

0.00

10.00

0.00

2.60

DCAN

SW

Average

304

1406

10.10

2.21

1.74

2.09

0.00

17.13

0.00

4.53



High

304

1406

10.10

2.72

2.20

2.52

0.00

24.60

0.00

5.40

GW

Average

110

524

57.63

0.87

0.00

2.08

0.00

17.65

0.00

2.35



High

110

524

57.63

1.25

0.00

2.72

0.00

21.00

0.00

3.70

TCAN

SW

Average

304

1393

96.91

0.03

0.00

0.30

0.00

7.28

0.00

0.00



High

304

1393

96.91

0.07

0.00

0.74

0.00

20.50

0.00

0.00

GW

Average

110

515

97.67

0.14

0.00

2.07

0.00

39.60

0.00

0.00



High

110

515

97.67

0.17

0.00

2.20

0.00

41.54

0.00

0.00

Source:	SW - Surface Water, GW - Groundwater

Quarterly Dist. Sys:	Quarterly Distribution System (DS) Samples.

Average - quarterly average of 4 locations in DS.

High - highest of 4 locations in DS.

Plants:	Number of plants sampled

N:	Number of samples

PctND:	Percent samples nondetect (detection limits not provided)

Mean:	Arithmetic mean of all samples

Median:	Median value of all samples

STD:	Standard deviation

Min:	Minimum Value

Max:	Maximum Value

plO:	10th percentile

p90:	90th percentile

ND:	Nondetected

EPA/OW/OST/HECD

IV-4

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

The median concentrations for all four chemicals were less than the corresponding mean
concentrations, for both surface water and groundwater. The mean concentrations of BCAN
(averaged across the four sampling locations) were 0.73 and 1.14 //g/L in groundwater and
surface water, respectively. The median concentrations of BCAN were 0.00 and 0.88 //g/L in
groundwater and surface water, respectively. The mean concentrations of DBAN (averaged
across the four sampling locations) were 0.82 and 0.75 //g/L in groundwater and surface water,
respectively. The median concentrations of DBAN were 0.43 and 0.27 //g/L in groundwater and
surface water, respectively. The mean concentrations of DCAN (averaged across the four
sampling locations) were 0.87 and 2.21 //g/L in groundwater and surface water, respectively. The
median concentrations of DCAN were 0.00 and 1.74 //g/L in groundwater and surface water,
respectively. The mean concentrations of TCAN (averaged across the four sampling locations)
were 0.14 and 0.03 //g/L in groundwater and surface water, respectively. The median
concentrations of TCAN were 0.00 //g/L in both groundwater and surface water. The lowest
mean concentrations are associated with the highest percentage of nondetects, which are treated
as zero in the calculation of the mean, median, standard deviation, and plO values (U.S. EPA,
2000a).

A.2 Factors Affecting the Relative Concentrations of BCAN, DBAN, DCAN, and TCAN

in Drinking Water

Sections A.2.1 - A.2.5 contain investigational information on the effects of disinfection
chemicals, influent bromide concentration, influent total organic carbon (TOC) concentration,

EPA/OW/OST/HECD

IV-5

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
temperature and pH, and seasonal shifts, respectively in BCAN, DBAN, DCAN, and TCAN

concentrations.

A.2.1 Disinfection Treatment

Chlorination has been the predominant water-disinfection method in the United States.
However, water utilities are considering a shift to alternative disinfectants. Therefore, there is a
need to understand the occurrence of DBPs in drinking water and the factors that may influence
their formation. Several published studies (Boorman et al., 1999; Richardson, 1998; Lykins et
al., 1994; Jacangelo et al., 1989; Miltner et al., 1990) reported on the formation of HANs and
other DBPs under different disinfection conditions.

In a review on drinking-water disinfection byproducts, Boorman et al. (1999) compared
the concentrations of different drinking-water disinfection byproducts, including BCAN, DBAN,
DCAN, and TCAN, formed by chlorination, ozonation, chlorine dioxide, and chloramination.
Most of the data that were available were from surface-water systems that used chlorination. For
the systems using chlorination, DCAN was present at the highest concentrations, with a median
and a maximum concentration of 2.1 and 10 |ig/L, respectively. The median and maximum
concentrations of BCAN were 0.6 and 1.1 |ig/L, respectively. The median concentrations of
both DBAN and TCAN in chlorinated water were less than their limits of detection at < 0.5 and
<0.02 |ig/L, respectively. The maximum concentrations of DBAN and TCAN in chlorinated
water were 9.4 and 0.02 |ig/L, respectively. The principal products formed by chloramination

EPA/OW/OST/HECD

IV-6

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
were similar to those formed by chlorination; additional information was not provided. HANs

were not detected when the water was treated with ozone. Boorman et al. (1999) reported that

chlorine dioxide formed oxidation by-products similar to those formed by ozonation; additional

details were not provided.

Richardson (1998) compared the relative concentrations of DBPs in drinking water using
different treatment methods, and also reported that chlorination produced the highest
concentration of DBPs, including BCAN, DBAN, DCAN, and TCAN, Compared to chlorine
treatment, chloramine produced 3% to 20% lower levels of by-products, including HANs.
Richardson (1998) speculated that, as with the other halogenated by-products of chlorination, the
formation of the HANs may be caused by residual chlorine in the chloramination process, rather
than by the chloramine itself. Richardson (1998) found that BCAN, DCAN, and TCAN were not
produced by ozonation or chlorine dioxide in measurable quantities. However, DBAN was
formed by ozone in the presence of elevated bromide, but not by chlorine dioxide disinfection.
When ozone was the primary disinfectant (i.e., ozone followed by chlorine or ozone followed by
chloramine) the formation of DBPs (including HANs) was less than when chlorine or chloramine
was the sole disinfectant. This was believed to be due to the destruction by ozone of DBP
precursor material. In addition, lower levels of chlorine or chloramine are required when ozone is
used as the primary disinfectant, and this also leads to lower levels of DBPs.

EPA/OW/OST/HECD

IV-7

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Lykins et al. (1994) investigated the formation of halogenated DBPs in the water-
distribution system, by predisinfecting and postdisinfecting the water with either chlorine or
chloramine and holding the water for five days. Similar to the other investigators, Lykins et al.
(1994) also found that the use of chlorine produced the highest concentration of halogenated
DBPs and that, in general, the concentrations were less when chloramine was used or when ozone
was used as a predisinfectant followed by either postchlorination or postchloramination. Lykins et
al. (1994) found that relatively low concentrations of HANs were formed. The highest average
concentration of total HANs (3.1 //g/L) was when chlorine was used, with or without ozonation.
The total HANs concentrations observed with the other process streams was < 1 //g/L. DCAN,
with a concentration of approximately 1.9 //g/L, was the predominant HAN when chlorine (with
or without ozone) was used, followed by BCAN (0.6 /ig/L), DBAN (~ 0.4 /ig/L), and TCAN
(0.1 ££g/L). In contrast, the average concentrations of BCAN, DBAN, DCAN, and TCAN were
0 £ig/L when the water was treated with chloramine or with ozone followed by chloramine.

Jacangelo et al. (1989) examined the impact of ozonation on the formation and control of
DBPs in drinking water at four utilities. Treatment modifications were made on the process train
at each full or pilot-scale plant to incorporate ozone in the treatment process. The disinfection
schemes that employed ozonation followed by chloramines as a disinfectant resulted in large
decreases (> 90%) in HAN formation relative to chlorination and > 25% to > 90% decreases in
HAN formation relative to chloramines only. For two utilities that measured individual HANs,
preozonation followed by chlorination decreased the total HANs by 29-35%, when compared

EPA/OW/OST/HECD

IV-8

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
with chlorination only. The concentrations of BCAN and DCAN decreased with ozonation, while

the concentrations of DBAN increased at one facility and showed little change at the second

facility. The concentrations of TCAN were less than the limit of detection (<0.012 //g/L) with

and without preozonation.

Miltner et al. (1990) studied DBP formation and control in three surface water pilot plants
employing three different disinfectant methods (chlorine, ozone followed by chlorine, and ozone
followed by chloramine). In an examination of the data using the Student's t-test, the authors
found that ozonation had no effect (at p = 0.05) on the formation of BCAN, DCAN, or TCAN in
simulated finished water and distribution water, and had no effect on the formation of DBAN in
simulated finished water. However, the formation of DBAN in simulated distribution water was
higher (at p = 0.05) when ozonation was combined with chlorination or with chloramination than
when chlorination was used alone.

A.2.2 Bromide Concentration

Ambient bromide levels appear to influence, to some degree, the speciation of HANs
(WHO, 2000). DCAN is by far the most predominant HAN detected in drinking water from
sources with bromide levels of 20 |ig/L or less. In treated water from sources with higher
bromide levels (50-80 //g/L), BCAN was the second most prevalent HAN. None of the treated
water from any of these sources had a DBAN concentration exceeding 0.5 |ig/L, including treated
water from one source that had a much higher bromide level, 170 |ig/L (WHO, 2000). However,

EPA/OW/OST/HECD

IV-9

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
Richardson (1998) found that when bromide was present in the source water, DBAN

concentrations were greater than those of chloroform or dichloroacetic acid, which normally

predominate.

A.2.3 Total Organic Carbon (TOC) Concentration

Many researchers have documented that chlorine reacts with natural organic matter,
including algae, humic acid, fulvic acid, and proteinaceous material to produce a variety of DBPs,
including HANs (Bieber & Trehy, 1983; Oliver, 1983; Reckhow and Singer, 1990; Reckhow et
al., 1990). Reckhow el al. (1990) found that the disinfection of water containing humic acids
resulted in higher concentrations of HANs than disinfection of water containing the corresponding
fulvic acids.

A.2.4 Temperature and pH

In general, increasing temperature and/or decreasing pH has been associated with
increasing concentrations of HANs (AWWARF, 1991; Siddiqui & Amy, 1993). Dihalogenated
acetonitriles (BCAN, DBAN, DC AN) are reported to undergo hydrolysis in water (Bieber &
Trehy, 1983). Arora el al. (1997) analyzed results of a DBP survey and a two-year DBP-
monitoring study of more than 100 treatment plants of the American Water System (a large water
utility) from 1989 to 1991, and found that HANs hydrolyzed at pH levels > 9.0 and continued to
degrade in the distribution system.

EPA/OW/OST/HECD

IV-10

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Different trends were observed in the HAN concentrations of different source waters. For
two source waters, HAN levels formed rapidly for the first eight hours and continued to increase
slowly or leveled off after 96 hours (AWWARF, 1991). DBAN levels remained relatively stable
over the 96 hours, as did BCAN and DCAN levels. For other sources, levels of HANs consisting
mostly of DCAN increased rapidly up to 4-8 hours and began to decline by the end of the 96-
hour period. For these sources, BCAN appeared to be slightly more stable than DCAN
(AWWARF, 1991).

A.2.5 Seasonal Shifts

Seasonal shifts in HANs were investigated by Krasner et al. (1989). In September 1987,
the US EPA's Office of Drinking Water entered into a cooperative agreement with the
Association of Metropolitan Water Agencies (AMWA) to perform a study of the occurrence and
control of DBPs. The AMWA contracted with the Metropolitan Water District of Southern
California (MWD) to provide management services for the project and to perform the DBP
analysis. In addition, the State of California Department of Health Services (CDHS), through the
California Public Health Foundation (CPHF), contracted with MWD to perform a similar study in
California. Baseline data were gathered on 35 water-treatment facilities, including 25 water
utilities across the United States in the U.S. EPA study and 10 California water utilities in the
CDHS study. Levels of BCAN, DBAN, DCAN, and TCAN were measured.

EPA/OW/OST/HECD

IV-11

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

During the first quarter (spring 1988), a good correlation was found between the HANs

and trihalomethanes, another class of disinfectant byproduct. In addition, Krasner et al. (1989)
reported that relatively high levels of the measured brominated DBPs were detected at some of
the utilities. These findings suggested that the influence of bromide in the raw water should be
evaluated. Therefore, chloride and bromide analyses were added to the protocol, beginning with
the second quarter (summer 1988) of sampling. Among the 35 facilities, bromide levels ranged
from < 0.01 to 3.00 mg/L. At the utility with the highest bromide levels (~ 3 mg/L bromide) there
was a shift in the distribution of DBPs from the chlorinated DBPs to the brominated DBPs,
resulting in DBAN as the major HAN detected. This is in apparent contrast to the findings of
WHO (2000), which found that at high bromide levels, BCAN was the second most prevalent
compound, following DC AN. While there were no clear trends of the concentrations of bromide
ions or brominated acetonitriles with season in the composite analysis, DBAN levels were higher
in the fall in the utility with the highest bromide levels. Some shifts in DBPs formed were also
seen as the result of drought conditions and saltwater intrusion.

B. Exposure to Sources Other Than Drinking Water

TCAN has been used as an insecticide (Budavari et al., 1989). No data were located on
exposure to BCAN, DBAN, DCAN, and TCAN in food, air, or via dermal exposure when
showering or swimming. Therefore, no assessment of overall exposure to any of the HANs can
be performed.

EPA/OW/OST/HECD

IV-12

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

C. Body Burden

No data could be located on body burden. However, the results of Roby et al. (1986) that
showed relatively rapid excretion of DCAN-associated radioactivity (at least 70% of the
administered dose excreted within 6 days in rats or within 24 hours in mice) suggests limited
potential for the bioaccumulation of the HANs.

D. Summary

The ICR database (U.S. EPA, 2002a) contains extensive information on concentrations of
BCAN, DBAN, DCAN, and TCAN in drinking-water systems, and on how those concentrations
vary with input-water characteristics and treatment methods. The database contains information
from six quarterly samples from 7/97 to 12/98, from approximately 300 large systems covering
approximately 500 plants. The mean concentrations of BCAN were 0.73 and 1.14 //g/L in
groundwater and surface water, respectively. The mean concentrations of DBAN were 0.82 and
0.75 £ig/L in groundwater and surface water, respectively. The mean concentrations of DCAN
were 0.87 and 2.21 //g/L in groundwater and surface water, respectively. The mean
concentrations of TCAN were 0.14 and 0.03 //g/L in groundwater and surface water,
respectively. The median concentrations of BCAN, DBAN, DCAN, and TCAN were less than
their means in groundwater and surface water.

HANs are produced during water chlorination or chloramination from naturally occurring
substances, including algae, humic acid, fulvic acid, and proteinaceous material. Reckhow el al.

EPA/OW/OST/HECD

IV-13

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
(1990) found that disinfection of water containing humic acids resulted in higher concentrations of

HANs than disinfection of water containing the corresponding fulvic acids.

The disinfection process producing the highest concentration of HANs was chlorination.
Chloramine produced lower levels of HANs. Most investigators (Boorman et al., 1999;
Richardson 1998; Lykins et al., 1994; Jacangelo et al., 1989) found that the formation of HANs
when ozonation was followed by chlorine or chloramine was less than when chlorine or
chloramine was the sole disinfectant. Interestingly, Miltner et al. (1990) reported that the
formation of DBAN in simulated distribution water was higher (at p = 0.05) when ozonation was
combined with chlorination or with chloramination than when chlorination was used alone. In
addition, Miltner et al. (1990) found that ozonation had no statistically significant effect on the
formation of BCAN, DCAN, or TCAN. Richardson (1998) found that BCAN, DCAN, and
TCAN were not produced in measurable quantities by ozonation or chlorine dioxide. However,
DBAN was formed by ozone in the presence of elevated bromide, but not by chlorine dioxide
disinfection.

Ambient bromide levels appear to influence, to some degree, the speciation of HANs.
DCAN is by far the most predominant HAN detected in drinking water from sources with
bromide levels of 20 |ig/L or less. In treated water from sources with higher bromide levels
(50-80 jj,g/L), BCAN was the second most prevalent compound (WHO, 2000). Richardson

EPA/OW/OST/HECD

IV-14

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
(1998) found that when bromide was present in the source water, DBAN concentrations were

greater than those of chloroform or dichloroacetic acid, which normally predominate.

In general, increasing temperature and/or decreasing pH has been associated with
increasing concentrations of HANs (AWWARF, 1991; Siddiqui & Amy, 1993). Although HANs
form rapidly, they decay in the distribution system as a result of hydrolysis. HANs hydrolyzed at
pH levels >9.0 and continued to degrade in the distribution system (Arora el al.,\ 997), The
relative stability of individual HANs appears to be dependent on the specific source water
(AWWARF, 1991).

In general, there were no clear trends of the concentrations of HANs with season.
However, among 35 water treatment facilities investigated, Krasner et al. (1989) found that at the
facility with the highest bromide level (~ 3 mg/L bromide), there was a shift in the distribution of
HANs from chlorinated HANs to brominated HANs. At this facility DBAN was the major HAN
detected, with the highest DBAN levels detected in the fall. This is in apparent contrast to the
findings of WHO (2000), which reported that at high bromide levels, BCAN was the second most
prevalent compound following DCAN.

TCAN has been used as an insecticide (Budavari et al., 1989). No data were located on
exposure to BCAN, DBAN, DCAN, and TCAN in food, air, or via dermal exposure when
showering or swimming.

EPA/OW/OST/HECD

IV-15

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

No data could be located on body burden. However, the results of Roby et al. (1986) that
showed relatively rapid excretion of DCAN-associated radioactivity (at least 70% of the
administered dose excreted within 6 days in rats or within 24 hours in mice) suggests limited
potential for the bioaccumulation of the HANs.

EPA/OW/OST/HECD

IV-16

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
Chapter V. Health Effects in Animals

Limited toxicity testing data are available for the HANs. Tables V-15 through V-18 at the
end of the chapter provide a summary of the toxicity studies for BCAN, DBAN, DCAN, and
TCAN for noncarcinogenic endpoints.

A. Short-Term Exposures

Acute oral LD50 values for DBAN, DCAN, and TCAN in rodents have been reported by
several investigators, as summarized in Table V-l. Acute oral LD50 values for DBAN and DCAN
were reported to range from 245 to 361 mg/kg in male and female CD rats and mice given single
oral doses dissolved in corn oil. Ataxia, depressed respiration, depressed activity, and coma
preceded death. No consistent, compound-related, gross pathological effects were observed at
necropsy (Hayes et al., 1986). In a limited report on the acute toxicity of DBAN, groups of 20
rats and mice (sex and strain not specified) were given single gavage doses of 10% DBAN in corn
oil ranging from 25 to 1,600 mg/kg in rats and from 25 to 3,200 mg/kg in mice. The LD50 was
estimated as 50 to 100 mg/kg in rats and 50 mg/kg in mice with slight to moderate convulsions
the only symptom reported (Eastman Kodak Co., 1992). Smyth et al. (1962) reported an oral
LD50 of 0.25 mL/kg (360 mg/kg) in male Wistar rats for TCAN (vehicle was not specified).

EPA/OW/OST/HECD

V-l

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Table V-l. Acute Oral Lethality of Haloacetonitriles.

Reference

Species

LD50 (mg/kg)
Male Female

DBAN

Hayes et al. (1986)

Mouse

289

(253-324)a

303

(269-342)

Hayes et al. (1986)

Rat

245

(210-286)

361

(320-410)

Eastman Kodak Co., 1992

Mouse

50 (sex not specified)

Eastman Kodak Co., 1992

Rat

50 - 100 (sex not specified)

DCAN

Hayes et al. (1986)

Mouse

270

(241 - 303)

279

(263 - 296)

Hayes et al. (1986)

Rat

339

(298 - 387)

330

(300 - 500)

TCAN

Smyth et al. (1962)

Rat

360



'95% Confidence limits.

Lin and Guion (1989) reported clinical signs of toxicity in a mechanistic study to evaluate
the ability of HANs to deplete liver GSH and inhibit GST activity. Male Fischer 344 rats were
administered single gavage doses of 0.75 mmol/kg BCAN (116 mg/kg), DBAN (149 mg/kg),
DCAN (82 mg/kg), or TCAN (108 mg/kg) in tricaprylin. These doses represented 10 to 30% of
the reported LD50 for the individual compounds. The incidence of deaths at 18 hours was 2 of 20
for TCAN and 1 of 10 for DBAN. No deaths were reported for the other HANs. Overt signs of
acute toxicity, including gasping and salivation, were observed for TCAN and DBAN.

EPA/OW/OST/HECD

V - 2

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

The acute toxicity of DBAN has also been evaluated by the inhalation route. Sprague-
Dawley rats (10 animals/sex/dose) were exposed for 6 hours to air concentrations of 3.5, 9.5,
17.9, or 41 ppm DBAN (28.5, 77.3, 145.6, or 333.5 mg/m3)(Dow Chemical Co., 1992). Slight
transient eye and nasal irritation was noted at the lowest concentration and increased in severity
with increasing concentration. In addition, respiratory tract irritation was observed at 9.5 ppm
and increased in severity with increasing concentration. Three males and two females did not
survive following exposure to 9.5 ppm; all males and 6 females in the 17.9 ppm exposure group
died. The LC50 was reported as 9.6 ppm for males and 14.4 ppm for females, respectively,
calculated using a moving average method. In the low exposure group, no treatment-related
effects were observed at necropsy. In the 9.5 ppm group, nasal irritation and distended stomach
were the principal findings in the animals that did not survive; no effects were observed in the
surviving animals. In the two high-exposure groups, necropsy revealed distended stomachs, and
congested liver and lungs. No treatment-related lesions were observed following gross
pathological examination in the four females that survived the 17.9 ppm exposure. In the study by
Smyth et al. (1962), no mortality was observed following a single 4-hour exposure to 125 ppm
(738 mg/m3) of TCAN in a group of six albino rats (sex and strain not specified), indicating that
the 4-hour LC50 would be well above 125 ppm. Data were not reported for end points other than
mortality in this study. Although the different exposure durations and strains used preclude a
direct comparison, the finding that the 4-hour LC50 for TCAN was likely much higher than the 6-
hour LC50 for DBAN suggests that DBAN has higher acute inhalation toxicity. No inhalation
studies were identified for BCAN or DCAN.

EPA/OW/OST/HECD

V - 3

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

In a poorly-reported skin irritation study, undiluted DBAN was applied to the skin of
guinea pigs (3 animals/dose) at doses ranging from 0.01 to 1.0 mL/kg (23 to 2,300
mg/kg)(Eastman Kodak Co., 1992). The estimated LD50 was 0.1 to 1.0 mL/kg (230 to 2,300
mg/kg). Irritant symptoms were recorded as moderate to gross edema, with necrosis of the entire
patch area and hemorrhagic periphery at 24 hours. At 1 week, the authors reported heavy eschar
that was broken at the edges with secondary eschar beneath. At 2 weeks, small secondary eschar
was noted, with heavy scarring and alopecia. The LD50 value for a single dermal application of
TCAN to male New Zealand rabbits was reported to be 0.90 mL/kg (1,296 mg/kg)(Smyth el al.,
1962). TCAN is extremely irritating; 0.5 mL of a 1% solution produced a severe eye burn in
rabbits, and 0.01 mL of the material caused necrosis when applied to the clipped skin of rabbits.
The study descriptions did not indicate if the TCAN was undiluted or applied in a solvent vehicle.
No dermal studies were identified for BCAN or DCAN.

In addition to acute lethality and irritation, other effects of short-term exposure to DBAN
and DCAN have been evaluated. The evaluation of several endpoints for DBAN toxicity have
been conducted or are being planned. DBAN has been selected for a neurotoxicity study (NTP,
2002), has been tested in a short-term reproductive and developmental toxicity screening study
(R.O.W. Sciences, 1997), and has been evaluated in 14-day dose-range finding studies (NTP,
2002). These latter two studies are described in this section.

EPA/OW/OST/HECD

V - 4

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

The short-term effects of DBAN were evaluated as part of dose-range finding studies
conducted for a reproductive and developmental toxicity screening study (R.O.W. Sciences,
1997). In the initial dose-range finding study, male and female Sprague-Dawley rats (5/sex/dose)
were exposed to drinking water containing 0, 250, 500, 1000, or 2000 ppm DBAN (doses
resulting from these exposures were not reported by the study authors). Significant decreases in
body weight relative to controls and decreased food consumption were observed within 4 days of
exposure, beginning at 250 ppm in males and 500 ppm in females. Water consumption was
significantly decreased at the lowest concentration tested (250 ppm) in males and females. The
onset of these effects led the authors to terminate the study, since the lowest concentration used
in this first range-finding study was judged to be too high for use in the main study. Therefore,
the rats were allowed to recover to control body weights, and were exposed to a lower set of
range-finding concentrations of 0, 7, 20, 70, or 200 ppm in drinking water for 2 weeks. The rats
were evaluated for clinical signs of toxicity, body weight, and feed and water consumption.

Based on measured water consumption and body weights, the estimated dose levels in the second
range-finding study were 0, 0.7, 2.2, 5.8, and 13.2 mg/kg/day for the males and 0, 0.8, 2.4, 6.8,
and 17.9 mg/kg/day for the females. No clinical signs of toxicity were observed in any of the
exposure groups. A significant decrease in body weights of roughly 10% was observed in males
exposed to 20 or 200 ppm DBAN after 4 days of treatment. However, no similar decrease was
observed in the males exposed to 70 ppm, and the mean final weight and total weight gain at the
end of the study were not affected in any dose group. No significant effects on feed consumption
were observed. Water consumption was significantly decreased (p<0.05) on day 11 in females

EPA/OW/OST/HECD

V - 5

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

exposed to 70 ppm (to 71% of controls). Water consumption was significantly decreased at each
time point examined (study days 4, 8, 11, and 15) for males and females in the 200-ppm group.
The maximum decrease in water consumption was to 50% of controls for males and to 67% of
controls for females. The absence of clinical signs of toxicity or body weight changes indicates
that the highest concentration tested (200 ppm in the second range-finding study) is a study
NOAEL. Decreased water consumption is not judged to be an adverse toxicological effect, as
this result may reflect poor palatability of the DBAN-containing water. Based on these
considerations, the study NOAEL is the highest dose tested (13.2 mg/kg/day in males; 17.9
mg/kg/day in females) and no LOAEL was determined.

In the main study (described in more detail in the Reproductive and Developmental
Toxicity Section), Sprague-Dawley rats were given 0, 15, 50, or 150 ppm DBAN in their drinking
water. Male rats (10 animals/dose) were given treated water on study days 6 through 35, and
were then examined for clinical pathology (hematology and clinical chemistry), body and organ
weight changes (liver, right kidney, spleen, thymus, right testis, right epididymis, right cauda
epididymis), sperm analyses, and histopathology. Estimated DBAN doses in males were 0, 1.4,
3.3, and 8.2 mg/kg/day as calculated by the study authors from body weight and water
consumption data. Separate groups of female rats were exposed to the same concentrations as
males on study days 1-34 (Group A) or from gestation day 6 to postnatal day 1 (Group B).
Estimated DBAN doses were 0, 1.8, 5.1, and 10.9 mg/kg/day for Group A females, and 0, 1.9,
5.3, and 10.8 mg/kg/day for Group B females. For both groups of females, clinical signs of

EPA/OW/OST/HECD

V - 6

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

toxicity, body weight, and feed and water consumption were determined at various intervals, and
general pathology was evaluated at necropsy. No compound-related effect was observed for any
of these parameters. The only biologically-significant effect that was treatment related for males
or females was a decrease in water consumption in the mid- and high-concentration groups, which
might have been related to decreased water palatability. Since the males were examined for more
sensitive endpoints than females (i.e. clinical chemistry and histopathology were evaluated), the
NOAEL for systemic effects in this study was 8.2 mg/kg/day, the highest dose level in males, and
no LOAEL was identified.

The subacute toxicity of DBAN has also been evaluated by the NTP (2002) in B6C3F1
mice and F344 rats as part of initial dose-range finding studies in support of chronic exposure
studies that are currently in progress. For the mouse study, DBAN was administered in drinking
water for 14 days to male and female B6C3F1 mice (5/sex/dose) at concentrations of 0, 12.5, 25,
50, 100, or 200 mg/L. The corresponding doses reported by the study authors were 0, 2.1, 4.3,
8.2, 14.7, and 21.4 mg/kg/day for males and 2.0, 3.3, 10.0, 13.9, and 21.6 mg/kg/day for females.
Animals were observed for clinical signs of toxicity, as well as body weight, and organ weight and
pathology. In addition, liver glutathione-S-transferase (GST) activity was measured. The only
treatment-related effect was a decrease in water consumption in both males and females. The
decrease in water consumption was concentration-related, decreasing to 58% of controls for
males and 54% controls for the 200 mg/L group mice. Since the only effect observed was a
concentration-related decrease in water intake, which could reflect poor water palatability, the

EPA/OW/OST/HECD

V - 7

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

high doses of 21.4 mg/kg/day for males and 21.6 mg/kg/day for females is considered a NOAEL.
No LOAEL is identified.

For the rat study, DBAN was administered in drinking water for 14 days to male and
female F344 rats (5/sex/dose) at concentrations of 0, 12.5, 25, 50, 100, or 200 mg/L. The
corresponding doses reported by the study authors were 0, 2, 3, 7, 12, and 18 mg/kg/day for
males and 2, 4, 7, 12, and 19 mg/kg/day for females. Animals were observed for clinical signs of
toxicity, as well as body weight, and organ weight and pathology. In addition, liver glutathione-
S-transferase (GST) activity was measured. A concentration-related decrease in water
consumption was observed for both males and females. Water consumption decreased to 60% of
controls for males and 61% controls for females in the 200 mg/L group mice. No effects other
than decreased water consumption were noted in females. However, in males DBAN exposure
caused a decrease in body weight gain and terminal body weight that was judged to be
toxicologically-significant only at the high dose. The reported body weight gains for males were
61.1, 66.3, 66.0, 65.2, 56.4, and 34.0 grams for the control and increasing dose-groups,
respectively. Terminal body weights as percent of controls were 100, 104.5, 104.3, 101.3, 98.7,
82.7%) for the control and increasing dose-groups, respectively. Significantly decreased testes
weights were observed in the high dose-group males. This finding was accompanied by testicular
atrophy in 2 of 5 males in this dose group. Elevated liver GST activity (126% of controls) was
also reported for high dose males. Based on decreased body weight and decreased testes weight

EPA/OW/OST/HECD

V - 8

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

and pathology in males, the NOAEL for this study is 12 mg/kg/day and the LOAEL is 18
mg/kg/day.

Hayes et al. (1986) investigated the subacute (14-day) toxicity of DBAN in adult CD rats.
The chemical was administered daily by gavage in corn oil at doses of 0, 23, 45, 90, or 180
mg/kg/day (10 animals/sex/dose). Endpoints assessed included mortality, body weight, organ
weight, serum chemistry, hematology, urinalysis, and gross necropsy, as shown in Tables V-2 and
V-3. Histopathological studies were not conducted. Also, food and water consumption rates
were not measured. There was 100% mortality at 180 mg/kg/day. At 90 mg/kg/day, 40% of the
males and 20% of the females were dead by day 14. No consistent, compound-related and dose-
dependent adverse effects were apparent in any of the serum chemistry, hematological, or urinary
parameters measured, although several serum chemistry parameters were significantly different
than controls, particularly at the high dose. The authors reported a trend toward higher values for
hemoglobin, total red blood cell and white blood cell counts, and fibrinogen in all treated animals
(no data reported). No remarkable findings were observed at necropsy (gross observation).
Sporadic relative or absolute organ weight changes were reported, mostly at the high dose, but
these were not considered as compound related. The only organ weight change that showed a
clear dose-dependence at multiple doses was for absolute liver weight in females, which was
significantly increased 12% over controls (p < 0.05) beginning at 23 mg/kg/day, up to 22%
above controls at 90 mg/kg/day. Liver weights relative to body weight and brain weight were
also affected, but only at higher doses. However, no corresponding significant elevation of serum

EPA/OW/OST/HECD

V - 9

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

levels of hepatic enzymes was observed in females, and in the absence of histopathology data, it is
unclear if the increased liver weight in females is an adverse response. The authors could not
identify specific target organs for DBAN toxicity, and concluded that decreased body weight was
the most sensitive indicator of toxicity. In males, no decrease in body weight was observed at
23 mg/kg/day; body weight was decreased by more than 20% in a dose-related manner at 45 and
90 mg/kg/day (p < 0.05). No effect on body weight was noted in females. The authors stated
that the NOAEL for DBAN was 45 mg/kg/day, but the decreased body weight in male rats
exposed to 45 mg/kg/day suggests that the NOAEL in this study is 23 mg/kg/day and the LOAEL
is 45 mg/kg/day. The data sets for body weight in males and relative liver weight in females were
further analyzed to determine benchmark doses (BMDs) according to draft EPA Guidance (U.S.
EPA, 2000c) to identify alternative critical effect levels. The results of the modeling are described
in detail in Appendix A. A BMDL of 16 mg/kg/day for decreased body weight in males was
selected as the most appropriate modeling result to serve as the basis for the quantitative dose-
response assessment.

EPA/OW/OST/HECD

V- 10

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Table V-2. Organ weights and ratios of CD rats exposed to dibromoacetonitrile by gavage for 14 days.3

Parameter15

Vehicle
(corn oil)
Male

23

mg/kg/day
Male

45

mg/kg/day
Male

90

mg/kg/day
Male

Vehicle
(corn oil)
Female

23

mg/kg/day
Female

45

mg/kg/day
Female

90

mg/kg/day
Female

Body Weight

275.5 ±3.2

274.6 ±5.5

243.7 ±4.6*

209.4 ± 10.1*

176.5 ±6.8

184.9 ±2.4

181.1 ±2.8

165.9 ±5.4

Brain

% body weight

1.81 ±0.05
0.66 ± 0.02

1.77 ±0.06
0.65 ±0.03

1.76 ±0.03
0.73 ±0.02*

1.66 ±0.05
0.80 ±0.06*

1.67 ±0.04
0.96 ±0.04

1.68 ±0.04
0.91 ±0.03

1.64 ±0.04
0.91 ±0.03

1.69 ±0.06
1.03 ±0.04

Liver

% body weight

12.91 ±
0.34

4.69 ±0.15

11.60 ± 1.24
4.25 ±0.45

11.47 ± 1.2
4.71 ±0.49

10.18 ± 1.89
4.72 ±0.83

8.59 ±0.26
4.89 ±0.12

9.65 ±
0.25*

5.22 ±0.11

9.73 ±
0.20*

5.39 ±0.15

10.50 ±0.44*
6.34 ±0.24*

Spleen

% body weight

0.64 ±0.03
0.23± 0.01

0.59 ±0.05
0.21 ±0.02

0.56 ± 0.05
0.23 ±0.02

0.42 ±0.05*
0.20 ± 0.02

0.48 ±0.03
0.27 ±0.01

0.45 ±0.02
0.24 ±0.01

0.52 ±0.03
0.29 ±0.02

0.42 ± 0.04
0.26 ± 0.02

Lungs

% body weight

1.69 ±0.10
0.61 ±0.03

1.49 ±0.09
0.54 ±0.04

1.44 ±0.07
0.59 ±0.03

1.30 ±0.14*
0.62 ± 0.06

1.58 ±0.19
0.88 ±0.08

1.42 ±0.09
0.77 ±0.05

1.25 ±0.06
0.69 ±0.04

1.09 ±0.06*
0.66 ±0.04*

Thymus
% body weight

0.51 ±0.03
0.19 ±0.01

0.52 ±0.04
0.19 ±0.01

0.46 ± 0.02
0.19 ±0.01

0.24 ±0.06*
0.11 ±0.03*

0.40 ±0.03
0.23 ±0.02

0.44 ±0.03
0.24 ± 0.02

0.40 ±0.01
0.22 ±0.01

0.21 ±0.02*
0.13 ±0.01*

Kidneys
% body weight

2.50 ±0.10
0.91 ±0.05

2.39 ±0.09
0.87 ±0.03

2.08 ±0.06*
0.85 ±0.02

2.13 ±0.15*
1.02 ±0.06

1.56 ±0.06
0.89 ±0.02

1.64 ±0.04
0.89 ±0.02

1.62 ±0.04
0.89 ±0.03

1.53 ±0.05
0.92 ±0.03

Testes/Ovaries
% body weight

2.70 ±0.09
0.99 ±0.04

2.71 ±0.07
0.99 ±0.03

2.67 ± 0.06
1.10 ±0.03

2.69 ±0.06
1.30 ±0.07*

0.11 ±0.01
0.06 ± 0.00

0.14 ±0.01
0.07 ±0.01

0.14 ±0.01
0.08 ±0.01

0.12 ±0.02
0.07 ±0.01

Adapted from Hayes et al. (1986).

" All data expressed as mean ± SEM.
b All absolute weights are presented in grams.
* Significantly different from vehicle control (p < 0.05).

EPA/OW/OST/HECD

V- 11

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Table V-3. Serum Chemistry Values for CD Rats Exposed to Dibromoacetonitrile (DBAN) for 14 days.3

Parameter

Vehicle
(corn oil)
Male

23

mg/kg/day
Male

45

mg/kg/day
Male

90

mg/kg/day
Male

Vehicle
(corn oil)
Female

23 mg/kg/day
Female

45mg/kg/
day

Female

90 mg/kg/day
Female

Serum Glutamate Pyruvate
Transaminase (IU/L)

168 ±55

75 ±9

76 ± 11

77 ± 19

74 ±6

62 ±8

74 ± 14

68 ± 11

Serum Glutamate Oxaloacetic
Transaminase (IU/L)

339± 105

142 ± 14

172 ± 26

273 ± 81

178 ±32

138 ±9

182 ±21

161 ± 15

Alkaline Phosphatase (IU/L)

360 ±41

466 ± 33

247 ± 29

187 ±8*

236 ± 26

299 ± 29

260 ± 24

179 ±34

5'-Nucleotidase (IU/L)

18 ±3

17 ± 1

13 ±2

14 ± 1

25 ±4

22 ±2

28 ±2

18 ±4

Protein (g/dL)

6.3 ±0.1

6.5 ±0.1

6.2 ±0.1

4.7 ±0.4*

6.7 ±0.1

6.2 ±0.3

6.2 ±0.1

5.3 ±0.2*

Albumin (g/dL)

4.0 ±0.1

4.0 ±0.0

4.0 ±0.1

2.9 ±0.3*

4.3 ±0.2

4.0 ±0.1

4.2 ±0.1

3.5 ± 0.1*

Globulin (g/dL)

2.3 ±0.1

2.5 ±0.1

2.2 ±0.1

1.8 ± 0.1*

2.5 ±0.1

2.2 ±0.2

2.1 ±0.1

1.8 ± 0.1*

Alb/globulin ratio

1.8 ±0.1

1.6 ±0.1

1.8 ±0.1

1.6 ±0.2

1.7 ±0.1

1.8 ±0.1

2.1 ±0.1

2.0 ±0.1

Glucose (mg/dL)

265 ± 47

212 ±41

217 ±40

130 ± 1

212 ± 17

171 ± 18

197 ± 22

138 ± 17

Cholesterol (mg/dL)

66 ±2

64 ±2

74 ±4

95 ±5*

66 ±4

58 ±6

66 ±6

71 ±3

Bilirubin (mg/dL)

0.3 ±0.0

0.4 ±0.0

0.2 ±0.0

0.3 ±0.1

0.3 ±0.0

0.2 ±0.0

0.3 ±0.0

0.2 ±0.0

BUN (mg/dL)

15 ± 1

13 ± 1

10 ± 1

18 ±6

15 ±2

13 ±2

13 ± 1

15 ±2

Creatinine (mg/dL)

1.3 ±0.2

0.8 ±0.1

0.9 ±0.1

0.7 ±0.0*

1.2 ±0.1

1.0 ±0.1

1.0 ±0.1

0.8 ±0.0*

BUN/creatinine ratio

13 ±2

16 ±2

12 ± 1*

24 ±8

12 ±2

15 ±4

14 ±2

19 ±3

Calcium (mg/dL)

11.3 ±0.4

12.5 ±0.6

11.8 ±0.4

9.6 ±0.6

11.9 ±0.4

11.1 ±0.3

12.1 ±0.5

11.1 ±0.3

Phosphorus (mg/dL)

12.0 ±0.4

11.2 ±0.4

11.3 ±0.5

7.6 ±0.7*

11.6 ±0.1

10.1 ±0.6

10.8 ±0.5

9.3 ±0.7*

Chloride (mEq/L)

99 ±2

99 ±2

99 ± 1

100 ±2

98 ± 1.4

102 ±0.7

100 ± 1.2

101 ±0.7

Adapted from Hayes et al. (1986).

a All data expressed as mean± SEM.

* Significantly different from vehicle control (p < 0.05).

EPA/OW/OST/HECD

V- 12

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Table V-4. Body Organ Weights and Ratios of CD rats Exposed to Dichloroacetonitrile by Gavage for 14 days.a

Parameterb

Vehicle
(corn oil)
Male

12

mg/kg/day
Male

23

mg/kg/day
Male

45

mg/kg/day
Male

90

mg/kg/day
Male

Vehicle
(corn oil)
Female

12

mg/kg/day
Female

23

mg/kg/day
Female

45

mg/kg/day
Female

90

mg/kg/day
Female

Body
Weight

157 ±5

170 ±7

147 ±5

137 ±5

115 ±4

148 ±6

147 ±6

143 ±4

146 ±5

113 ±5

Brain
% body
weight

1.68 ±0.09
1.06 ±0.03

1.63 ±0.12
0.99 ±0.09

1.66 ±0.03
1.14 ±0.04

1.56 ±0.04
1.15 ±0.04

1.52 ±
0.05*
1.34 ±
0.06*

1.61 ±0.04
1.09 ±0.04

1.52 ±0.08
1.04 ±0.04

1.47 ±0.06
1.03 ±0.03

1.61 ±0.05
1.11 ±0.07

1.45 ±0.04
1.30 ±0.06*

Liver
% body
weight

8.15 ±0.31
5.19 ±0.10

9.96 ±0.58
5.85 ±0.16*

9.71 ±0.64
6.57 ±
0.30*

10.11±
0.52*

7.37±0.18*

8.63 ±0.48
7.50 ±0.26*

8.06 ±0.53
5.40 ± .20

8.49 ±0.35
5.81 ±0.15

10.56 ±

0.50*

7.37 ±0.27*

11.14 ±0.57*
7.59 ±0.23*

7.78 ±0.92
7.07 ±0.82*

Spleen
% body
weight

0.54 ±0.04
0.34 ±0.02

0.62 ±0.07
0.36 ±0.04

0.46 ±0.03
0.31 ±0.01

0.47 ±0.05
0.34 ±0.03

0.35 ±
0.03*

0.30 ±0.02

0.44 ±0.03
0.30 ±0.02

0.43 ±0.04
0.30 ±0.03

0.40 ±0.02
0.28 ±0.02

0.43 ±0.02
0.29 ±0.02

0.29 ±0.02*
0.26 ±0.01

Lungs
% body
weight

1.27 ±0.07
0.81 ±0.04

1.37 ±0.11
0.82 ±0.07

1.26 ±0.08
0.85 ±0.05

1.17 ± 0.07
0.86 ±0.05

1.02 ±0.09
0.89 ±0.07

1.35 ±0.09
0.91 ±0.06

1.17 ±0.09
0.80 ±0.05

1.13 ±0.07
0.79 ±0.05

1.23 ±0.10
0.84 ±0.06

0.96 ±0.07*
0.86 ±0.07

Thymus
% body
weight

0.43 ±0.04
0.27 ±0.02

0.55 ±0.05
0.32 ±0.03

0.38 ±
0.02*

0.26 ±0.02

0.33 ±
0.02*

0.24 ±0.02

0.27 ±
0.02*
0.24 ±
0.02*

0.48 ±0.03
0.33 ±0.02

0.42 ±0.04
0.29 ±0.03

0.38 ±0.03
0.27 ±0.02

0.42 ±0.02
0.29 ±0.02

0.32 ±0.02*
0.29 ±0.02

Kidneys
% body
weight

1.53 ±0.05
0.98 ±0.03

1.85 ±0.08*
1.10 ±0.04*

1.52 ±0.05
1.03 ±0.02

1.48 ±0.06
1.08 ±0.02

1.38 ±0.07
1.20 ±
0.04*

1.52 ±0.05
1.03 ±0.03

1.42 ±0.06
0.97 ±0.03

1.47 ±0.04
1.03 ±0.03

1.51 ±0.06
1.03 ±0.02

1.28 ±0.05*
1.14 ±0.04

Testes/
Ovaries
% body
weight

1.54 ± 0.13
0.97 ±0.07

1.74 ±0.15
1.00 ±0.06

1.54 ±0.12
1.03 ±0.06

1.26 ±0.13
0.90 ±0.09

1.25 ±0.14
1.07 ±0.11

0.11 ±0.01
0.08 ±0.01

0.10 ±0.01
0.07 ±0.01

0.10 ±0.01
0.07 ±0.01

0.10 ±0.01
0.07 ±0.01

0.08 ±0.01
0.07 ±0.01

Adapted from Hayes et al. (1986).

a All data expressed as mean ± SEM.
b All absolute weights are presented in grams.
* Significantly different from vehicle control (p < 0.05).

EPA/OW/OST/HECD

V- 13

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Table V-5. Serum Chemistry Values for CD Rats Exposed to Dichloroacetonitrile (DCAN) for 14 days3.

Parameter

Vehicle
(corn oil)
Male

12

mg/kg/day
Male

23

mg/kg/day
Male

45

mg/kg/day
Male

90

mg/kg/day
Male

Vehicle
(corn oil)
Female

12

mg/kg/day
Female

23

mg/kg/day
Female

45

mg/kg/day
Female

90

mg/kg/day
Female

Serum Glutamate
Pyruvate

Transaminase (IU/L)

70 ±3

71 ±6

89 ±22

87 ± 15

88 ± 14

57 ±6

66 ±7

63 ±4

63 ±7

134 ±38*

Serum Glutamate
Oxaloacetic
Transaminase (IU/L)

167 ± 13

266 ± 22

213 ±53

276 ± 40

233 ±27

205 ±8

198 ±6

201 ±33

181 ±28

340 ± 96

Alkaline Phosphatase
(IU/L)

457 ±63

459 ±39

384 ±21

398 ±50

712 ±84*

261 ±24

381 ±48

352 ±45

384 ±35*

651±159*

5'-Nucleotidase
(IU/L)

14 ± 1

16 ± 1

14 ±2

17 ± 2

23 ±2*

23 ±3

20 ±2

16 ± 1

18 ± 1

24 ±2

Protein (g/dL)

5.8 ±0.2

6.0 ±0.2

5.7 ±0.2

5.5 ±0.2

6.0 ±0.1

6.4 ±0.3

6.2 ±0.2

6.4 ±0.3

5.9 ±0.2

6.1 ±0.4

Albumin(g/dL)

4.4 ±0.1

4.3 ±0.1

4.3 ±0.1

4.3 ±0.1

4.8 ±0.1

4.6 ±0.1

4.4 ±0.1

4.5 ±0.1

4.7 ±0.1

4.8 ±0.2

Globulin (g/dL)

1.4 ± 0.1

1.7 ± 0.1

1.4 ±0.2

1.2 ±0.2

1.2 ±0.2

1.8 ±0.2

1.9 ±0.2

1.9 ±0.2

1.2 ± 0.1

1.3 ± 0.3

Alb/globulin ratio

3.1 ±0.1

2.6 ±0.1*

3.4 ±0.5

3.8 ±0.5

4.5 ±0.7

2.8 ±0.4

2.4 ±0.2

2.4 ±0.2

3.9 ±0.3

4.0 ±0.7

Glucose (mg/dL)

154 ± 11

129 ± 16

138 ±6

134 ± 10

112 ± 16

191 ± 19

162 ± 30

142 ± 17

160 ±6

137 ± 15

Cholesterol (mg/dL)

90 ±5

93 ±2

91 ±6

80 ± 10

81 ± 11

84 ±3

86 ±4

99 ±6

82 ±8

73 ±8

Bilirubin (mg/dL)

0.2 ±0.0

0.2 ±0.0

0.3 ±0.0

0.4 ±0.0*

0.6 ±0.1

0.2 ±0.0

0.2 ±0.0

0.2 ±0.0

0.4 ±0.0*

0.7 ±0.1*

BUN (mg/dL)

17 ± 2

17 ± 2

15 ± 2

17 ± 2

15 ± 1

17 ± 1

17 ± 1

15 ± 2

20 ±3

23 ±2

Creatinine (mg/dL)

1.7 ±0.2

1.4 ±0.2

1.4 ± 0.1

1.5 ± 0.1

1.4 ± 0.1

1.6 ±0.2

1.3 ± 0.1

1.2 ± 0.1

1.6 ± 0.1

1.6 ±0.2

BUN/creatinine ratio

11 ±3

12 ± 1

11 ± 1

12 ±2

11 ± 1

11 ± 1

13 ± 1

13 ± 2

12 ± 1

14 ± 1

Calcium (mg/dL)

11.4 ± 0.1

11.0 ± 0.2

11.3 ± 0.2

9.8 ±1.5

11.7 ± 0.3

11.7 ± 0.5

11.1 ±0.3

11.3 ± 0.2

11.9 ± 0.3

11.8 ± 0.5

Phosphorus (mg/dL)

12.3 ±0.6

12.0 ±0.8

12.5 ±0.6

11.7 ± 0.5

12.9 ±0.4

11.4 ± 0.9

10.9 ±0.5

11.4 ± 0.9

12.7 ±0.1

9.9 ±0.4

Chloride( mEq/L)

100 ± 1

100 ± 1

101 ± 1

101 ±2

101 ± 1

100 ± 1

99 ± 1

100 ± 1

101 ± 1

101 ±2

Adapted from Hayes et al. (1986). aAll data expressed as mean ± SEM * Significantly different from vehicle control at (p<0.05)

EPA/OW/OST/HECD

V- 14

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

In a similar 14-day repeated dosing study with DCAN, Hayes el al. (1986) administered
gavage doses of 0, 12, 23, 45, or 90 mg/kg/day in corn oil to CD rats (10 animals/sex/dose).

Food and water consumption rates were not measured. No mortality was reported for any
treatment group. In males, depression in body weight to 94%, 87%, and 73% of body weight in
control animals was observed at 23, 45, and 90 mg/kg/day, respectively, as shown in Table V-4.
In females, body weight was decreased to 76% of controls at 90 mg/kg/day. Several serum
markers for organ toxicity were increased in treated animals as shown in Table V-5. Significantly
increased levels of serum glutamic pyruvate transaminase (SGPT) in females at 90 mg/kg/day, and
alkaline phosphatase (ALP) levels at 90 mg/kg/day in males and at 45 and 90 mg/kg/day in
females were reported. Although the authors did not consider these changes to be compound-
related adverse effects, they did not provide a reason. We considered these changes to be
adverse, based on the magnitude of the change, and the supporting data for DCAN in females in
the subchronic study. No other consistent dose-dependent adverse effects were observed in any
of the other serum chemistry, hematological, or urinary parameters measured. The authors
reported a trend toward elevated red blood cell and white blood cell counts in all treated animals,
but no data were provided. No remarkable findings were observed at necropsy (gross
observation). Relative liver weight was significantly increased (p < 0.05) in male and female rats.
The relative liver weights in males were 13%, 26%, 42%, and 45% greater than controls at doses
of 12, 23, 45, and 90 mg/kg/day, respectively. Absolute liver weight, in contrast, was
significantly increased (p < 0.05) only in the 45 mg/kg/day dose group. In female rats, both
relative and absolute liver weights were elevated beginning at 23 mg/kg/day, with relative liver
weights 36%), 40%, and 31% greater than controls at 23, 45, and 90 mg/kg/day, respectively.

EPA/OW/OST/HECD

V- 15

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Liver weight changes would be considered as potentially adaptive in the absence of other signs of
hepatic injury. However, the observed increase in serum levels of hepatic enzyme activity at
higher doses than those associated with liver weight gives greater weight to the potential
toxicological significance of the liver weight changes, even though the absence of histopathology
data makes it difficult to determine conclusively if the effects were adverse at low doses. Based
on this uncertainty, both decreased body weight and increased relative liver weight are considered
toxicologically-relevant responses. The more sensitive of these endpoints was selected as the
critical effect. Therefore, the lowest dose tested of 12 mg/kg/day is the study LOAEL for
increased relative liver weight in males, and no NOAEL is determined. The data sets for body
weight and relative liver weight in males and females were further analyzed to determine
benchmark doses (BMDs) according to draft EPA Guidance (U.S. EPA, 2000c) to identify
alternative critical effect levels. The results of the modeling are described in detail in Appendix A.
A BMDL of 5 mg/kg/day for relative liver weight in males was selected as the most appropriate
modeling result to serve as the basis for the quantitative dose-response assessment.

B. Long-Term Exposures

No studies were identified that evaluated the toxicity of long-term exposure to BCAN or
TCAN by any route. No studies were identified that evaluated the toxicity of long-term exposure
to DBAN or DCAN by the inhalation or dermal routes.

The subchronic toxicity of DBAN has been evaluated by the NTP (2002) in B6C3F1 mice
and F344 rats as part of initial dose-range finding studies for chronic exposure studies that are

EPA/OW/OST/HECD

V- 16

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

currently in progress. For the mouse study, DBAN was administered in drinking water for 13
weeks to male and female B6C3F1 mice (10/sex/dose) at concentrations of 0, 12.5, 25, 50, 100,
and 200 mg/L. The corresponding doses reported by the study authors were 0, 1.6, 3.2, 5.6,
10.7, and 17.9 mg/kg/day for males and 0, 1.6, 3.0, 6.1, 11.1, and 17.9 mg/kg/day for females.
Animals were observed for clinical signs of toxicity, as well as body weight, organ weight and
pathology, hematology, and clinical chemistry. A separate set of animals (10/sex/dose) were
exposed to the same concentrations as the main study groups for 26 days, but were co-exposed to
DBAN and 5-bromo-2-deoxyuridine during the last five days of this period. These animals were
used to collect tissue sample for analysis of induce cell proliferation. Decreased water
consumption and decreased body weight were the only effects related to DBAN treatment.
Decreased water consumption was observed in both males and females at DBAN concentrations
of 50 mg/L and higher. A slight and transient decrease in body weight gain was observed;
terminal body weights were 94% of controls in high-dose males and 96% of controls in high dose
females. These small changes are not judged as toxicologically-significant. Based on the minimal
effects observed for DBAN in this study, the NOAEL is 17.9 mg/kg/day for males and females.
No LOAEL was identified.

For the rat study, DBAN was administered in drinking water for 13 weeks to male and
female F344 rats (10/sex/dose) at concentrations of 0, 12.5, 25, 50, 100, and 200 mg/L. The
corresponding doses reported by the study authors were 0, 0.9, 1.8, 3.3, 6.2, and 11.3 mg/kg/day
for males and 0, 1.0, 1.9, 3.8, 6.8, and 12.6 mg/kg/day for females. Animals were observed for
clinical signs of toxicity, as well as body weight, organ weight and pathology, hematology, and

EPA/OW/OST/HECD

V- 17

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

clinical chemistry. A separate set of animals (10/sex/dose) were exposed to the same
concentrations as the main study groups for 26 days, but were co-exposed to DBAN and 5-
bromo-2-deoxyuridine during the last five days of this period. These animals were used to collect
tissue samples for analysis of cell proliferation. Decreased water consumption and decreased body
weight were the only effects related to DBAN treatment. Slight changes in clinical chemistry and
hematology findings were considered by the study authors to be related to decreased water
consumption. Decreased water consumption was observed in males at DBAN concentrations of
50 mg/L and higher and in females at the two highest concentrations. A slight decrease in body
weight gain was observed for high dose males and females. Terminal body weights were 94% of
controls in high-dose males and 95% of controls in high dose females. These small changes are
not judged as toxicologically-significant. Based on the minimal effects observed for DBAN in this
study, the NOAEL is 11.3 mg/kg/day for males and 12.6 mg/kg/day for females. No LOAEL was
identified.

In a 90-day study of DBAN toxicity, Hayes et al. (1986) administered doses of 0, 6, 23,
or 45 mg/kg/day in corn oil by gavage to groups of CD rats (20 animals/sex/dose). No
compound-related deaths occurred during this study. Body weights were depressed to 79% of
controls at the end of the study in males, but not in females, at the highest dose tested (45 mg/kg/
day). Food and water consumption rates were not measured. Observed changes in the serum
chemistry, hematological, urinary parameters, and organ weight were generally not dose-related
and were not considered to be compound-related (Tables V-6 and V-7). The only exceptions
were significantly increased ALP in females at 45 mg/kg/day and a significant increase in relative

EPA/OW/OST/HECD

V- 18

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

liver weight (but not absolute weight) in males at 45 mg/kg/day. Interim serum biochemistry
analyses at one and two months of exposure also revealed no treatment-related effects. No
remarkable findings were apparent at gross necropsy and the authors did not identify a specific
target tissue for DBAN, The finding that effects on the liver in this subchronic study were
comparable to the effects following the 14-day exposure at the same doses suggests that the liver
weight changes seen in the 14-day study were adaptive, or at least that tolerance developed to the
repeated dosing. In addition, no clear increase in serum enzyme markers of hepatic injury were
observed in males, even though they had a greater increase in relative liver weight than females.
Based on this discordance between serum biochemistry and liver weight findings, and in light of
the results from the 14-day study, the observed liver weight changes were not judged sufficiently
adverse to serve as the basis for the dose-response assessment. Based on decreased body weight
in males as the most sensitive endpoint, the NOAEL for this study is 23 mg/kg/day and the
LOAEL is 45 mg/kg/day.

The body weight data in males were further analyzed to determine benchmark doses
(BMDs) according to draft EPA Guidance (U.S. EPA, 2000c) to identify alternative critical effect
levels. The results of the modeling are described in detail in Appendix A. A BMDL of 20
mg/kg/day for decreased body weight in males was selected as the most appropriate modeling
result to serve as the basis for the quantitative dose-response assessment.

EPA/OW/OST/HECD

V- 19

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Table V-6. Body and Organ Weights for CD Rats Exposed to Dibromoacetonitrile

(DBAN) by Gavage for 90 Days3.

Parameterb

Vehicle
(corn oil)
Male

6

mg/kg/day
Male

23

mg/kg day
Male

45

mg/kg/day
Male

Vehicle
(corn oil)
Female

6

mg/kg/day
Female

23

mg/kg/day
Female

45

mg/kg/day
Female

Body
Weight

556.0±12.7

545.9±13.5

523.1±11.4

438.6±7.5*

292.9±5.8

279.6±7.3

291.7±9.5

274.9±7.2

Brain
% body
weight

1.96±0.04
0.36±0.01

1.98±0.05
0.37±0.01

1.95±0.04
0.38±0.01

1.93±0.4
0.44±0.01*

1.83±0.03
0.63±0.02

1.81±0.04
0.65±0.02

1.76±0.04
0.61±0.01

1.76±0.04
0.64±0.02

Liver
% body
weight

21.85±0.99
3.93±0.15

21.70±0.68
3.98±0.08

22.39±0.84
4.27±0.12

19.7±0.67
4.44±0.13*

11.62±0.37
3.98±0.12

11.1±0.36
3.98±0.10

12.31±0.33
4.25±0.11

11.95±0.39
4.36±0.12

Spleen
% body
weight

0.84±0.04
0.15±0.01

0.84±0.03
0.15±0.01

0.78±0.04
0.15±0.01

0.72±0.04
0.16±0.01

0.51±0.03
0.17±0.01

0.52±0.02
0.19±0.01

0.55±0.03
0.19±0.01

0.57±0.04
0.21±0.01

Lungs
% body
weight

3.22±0.14
0.58±0.03

2.95±0.09
0.54±0.02

3.09±0.11
0.59±0.02

2.73±0.09*
0.63±0.03

2.27±0.15
0.78±0.05

2.34±0.16
0.85±0.07

2.35±0.14
0.81±0.05

2.47±0.17
0.90±0.06

Thymus
% body
weight

0.55±0.03
0.10±0.01

0.50±0.03
0.09±0.01

0.53±0.02
0.10±0.00

0.43±0.03*
0.10±0.01

0.40±0.03
0.14±0.01

0.40±0.03
0.15±0.01

0.34±0.02
0.12±0.01

0.26±0.03*
1.10±0.01*

Kidneys
% body
weight

3.67±0.10
0.66±0.02

3.56±0.09
0.66±0.02

3.62±0.10
0.69±0.01

3.16±0.10*
0.72±0.02*

2.10±0.05
0.72±0.02

1.98±0.06
0.71±0.02

2.01±0.70
0.70±0.02

1.98±0.07
0.72±0.02

Testes/
Ovaries
% body
weight

3.58±0.05
0.65±0.01

3.76±0.07
0.69±0.02

3.49±0.60
0.67±0.02

3.51±0.15
0.80±0.03*

0.17±0.01
0.06±0.00

0.17±0.01
0.06±0.00

0.16±0.01
0.05±0.00

0.17±0.01
0.06±0.60

Adapted from Hayes et al. (1986).

aAll data expressed as mean ± SEM

bAll absolute weights are presented in grams.

* Significantly different from vehicle control (p < 0.05).

EPA/OW/OST/HECD

V - 20

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Table V-7. Serum Chemistry Values for CD Rats Exposed to Dibromoacetonitrile (DBAN) by Gavage for 90 days.3

Parameter

Vehicle
(corn oil)
Male

6

mg/kg/day
Male

23

mg/kg/day
Male

45

mg/kg/day
Male

Vehicle
(corn oil)
Female

6

mg/kg/day
Female

23

mg/kg/ day
Female

45

mg/kg/day
Female

Serum Glutamate Pyruvate
Transaminase (IU/L)

44 ±4

69 ±3

42 ±4

44 ±5

46 ±3

35 ±3

37 ±3

32 ±3*

Serum Glutamate Oxaloacetic
Transaminase (IU/L)

176 ± 12

245 ±52

190 ±30

181 ± 13

207 ±31

169 ±31

174 ± 10

161 ± 11

Alkaline Phosphatase (IU/L)

191 ± 16

171 ±28

192 ±23

157 ± 16

134 ± 11

130 ±9

138 ± 19

208 ± 32*

5'-Nucleotidase (IU/L)

15 ± 1

17 ±2

16 ± 1

13 ± 1

26 ±2

26 ±2

24 ±2

19 ± 1*

Protein (g/dL)

8.2 ±0.4

8.2 ±0.3

8.1 ±0.2

7.0 ±0.2*

7.3 ±0.2

7.5 ±0.2

7.1 ±0.2

6.9 ±0.2

Albumin(g/dL)

5.2 ±0.1

5.3 ±0.2

5.8 ± 0.1*

5.8 ± 0.1*

6.1 ±0.1

6.6 ±0.1*

6.1 ±0.1

5.6 ±0.1*

Globulin (g/dL)

2.9 ±0.5

2.9 ±0.4

2.3 ±0.2

1.2 ± 0.1*

1.2 ±0.1

0.9 ±0.2

1.1 ±0.1

1.3 ±0.2

Alb/globulin ratio

2.5 ±0.6

2.3 ±0.3

2.8 ±0.4

5.8 ±0.8

5.7 ±0.6

9.0 ± 1.3*

6.3 ±0.8

5.3 ±0.8

Glucose (mg/dL)

148 ±7

139 ±7

143 ±8

134 ±5

145 ±8

147 ±5

135 ± 10

130 ±7

Cholesterol (mg/dL)

73 ±5

70 ±5

69 ±4

77 ±5

80 ±4

80 ±6

78 ±4

64 ±3*

Bilirubin (mg/dL)

0.6 ±0.1

0.9 ±0.1

0.8 ±0.1

0.7 ±0.0

0.5 ±0.0

0.5 ±0.0

0.4 ±0.1

0.7 ±0.0*

BUN (mg/dL)

17 ± 1

15 ± 1

17 ± 1

18 ± 1

18 ± 1

18 ± 1

18 ± 1

15 ± 1

Creatinine (mg/dL)

1.0 ±0.1

0.9 ±0.1

1.3 ±0.0

1.0 ±0.0

1.2 ±0.0

1.3 ±0.1

1.0 ±0.1

1.0 ±0.0*

BUN/creatinine ratio

17 ± 1

17 ±2

14 ±0

19 ± 1

15 ± 1

14 ± 1

18 ±2

15 ± 1

Calcium (mg/dL)

11.3 ±0.5

11.7 ±0.2

12.8 ±0.2*

10.9 ±0.2

10.9 ±0.2

11.4 ±0.2

11.1 ±0.2

10.7 ±0.3

Phosphorus (mg/dL)

6.4 ±0.4

6.9 ±0.4

6.7 ±0.2

6.5 ±0.2

4.6 ±0.2

4.8 ±0.1

5.7 ±0.2*

6.2 ±0.3*

Chloride (mEq/L)

102 ± 1

101 ± 1

102 ± 1

102 ± 1

102 ± 1

101 ± 1

102 ± 1

101 ± 1

Adapted from Hayes et al. (1986).

" All data expressed as mean SEM
* Significantly different from vehicle control (p < 0.05).

EPA/OW/OST/HECD

V - 21

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

In a 90-day study of DCAN toxicity, Hayes el al. (1986) administered doses of 0, 8, 33,
or 65 mg/kg/day by gavage in corn oil to groups of CD rats (20 animals/sex/dose). Food and
water consumption data were not reported. At 65mg/kg/day, 50% of males and 25% of females
had died by the completion of the study; at 33 mg/kg/day, 10% of males and 5% of females had
died; and at 8 mg/kg/day, 5% of males had died. Supplementary information provided by the
authors indicated that there were one to three deaths (5% to 15%) in the control groups, and that
some of the deaths in the high-dose groups were due to gavage error. Most of the deaths that
were judged to be compound-related occurred in weeks 9 to 10. Body weight was significantly
depressed in male and female rats at 65 mg/kg/day (to 73% of controls) and in males at
33 mg/kg/day (to 81% of controls). Most of the observed changes in the serum chemistry,
hematological, and urinary parameters did not appear to be compound-related (Tables V-8 and V-
9). The exception was alkaline phosphatase, which was significantly increased in males and
females at the high dose, and in males also at 33 mg/kg/day. Sporadic organ weight changes were
observed, mostly at the high dose. Of these, a dose-dependent increase was seen only for relative
liver weights. Relative liver weight (relative to body weight) was significantly increased (p<0.05)
in males beginning at 33 mg/kg/day (60% increase), and in females beginning at 8 mg/kg/day
(17%) increase). The relative liver weight was also increased in males (by 12%) at 8 mg/kg/day.
Although the 12% increase in relative liver weight in males was not statistically significant, it is
judged to be biologically significant, based on the magnitude of the change and the observation in
the 14-day study that males were more sensitive to liver weight changes than females. Liver
weight changes would be considered as potentially adaptive in the absence of other signs of
hepatic injury. The observed increase in serum levels of ALP activity gives greater weight to the

EPA/OW/OST/HECD

V - 22

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

potential toxicological significance of the liver weight changes, although ALP is not a liver-
specific enzyme, and no corresponding increases in the liver-specific enzymes SGPT or serum
glutamate oxaloacetic transaminase (SGOT) were observed at study termination in the subchronic
study. However, the toxicological relevance of the ALP results following subchronic dosing is
supported by the increase in both ALP and SGPT observed in the 14-day study. The absence of
histopathology data makes it difficult to determine conclusively if the effects were adverse at low
doses. Based on this uncertainty, both decreased body weight and increased relative liver weight
are considered toxicologically-relevant responses. The more sensitive of these endpoints was
selected as the critical effect. Therefore, the lowest dose tested of 8 mg/kg/day is the study
LOAEL for increased relative liver weight in males and females, and no NOAEL is determined.

The body weight and relative liver weight data in males and females were further analyzed
to determine benchmark doses (BMDs) for these endpoints according to draft EPA Guidance
(U.S. EPA, 2000c) to identify alternative critical effect levels. The results of the modeling are
described in detail in Appendix A. A BMDL of 4 mg/kg/day for increased relative liver weight in
males was selected as the most appropriate modeling result to serve as the basis for the
quantitative dose-response assessment.

EPA/OW/OST/HECD

V - 23

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Table V-8. Body and Organ Weights for CD Rats Exposed to Dichloroacetonitrile (DCAN) by Gavage for 90

days3.

Parameterb

Vehicle
(corn oil)
Male

8

mg/kg/day
Male

33

mg/kg/ day
Male

65

mg/kg/day
Male

Vehicle
(corn oil)
Female

8

mg/kg/day
Female

33

mg/kg day
Female

65

mg/kg/day
Female

Body
Weight

541.4±13.1

572.1±14.5

438.0±16.6*

285.6±20.6*

309.0±6.91

282.5±17.4

280.6±10.7

225.5±8.9*

Brain
% body
weight

1.86±0.06
0.35±0.01

1.84±0.07
0.33±0.02

1.86±0.06
0.44±0.02*

1.77±0.06
0.63±0.03*

1.68±0.07
0.55±0.02

1.69±0.07
0.57±0.02

1.58±0.06
0.58±0.03

1.46±0.08
0.66±0.05

Liver
% body
weight

21.43±0.73
4.0±0.10

25.83±0.104
4.5±0.13

27.26±1.06
6.4±0.43*

17.91±1.60
6.4±0.92*

12.10±0.42
4.0±0.14

14.13±0.87
4.7±0.29*

16.83±0.58*
6.1±0.22*

13.88±0.705
6.1±0.15*

Spleen
% body
weight

0.72±0.03
0.13±0.03

0.73±0.03
0.13±0.02

0.63±0.03
0.15±0.04

0.50±0.06*
0.18±0.03

0.54±0.02
0.18±0.01

0.56±0.03
0.19±0.01

0.50±0.02
0.18±0.01

0.450±0.020
0.20±0.01

Lungs
% body
weight

2.98±0.13
0.56±0.02

2.77±0.13
0.49±0.02*

2.63±0.10
0.61±0.02

1.83±0.06*
0.65±0.04

2.04±0.10
0.67±0.03

2.20±0.15
0.74±0.05

2.22±0.16
0.79±0.03*

1.83±0.10
0.82±0.05*

Thymus
% body
weight

0.60±0.03
0.12±0.01

0.79±0.04*
0.14±0.01

0.62±0.03
0.15±0.01*

0.31±0.03*
0.11±0.02

0.50±0.03
0.16±0.01

0.54±0.03
0.18±0.01

0.42±0.02*
0.15±0.01

0.375±0.02
0.17±0.01

Kidneys
% body
weight

3.71±0.07
0.69±0.01

3.88±0.12
0.68±0.02

3.45±0.12
0.81±0.04*

2.77±0.19*
1.0±0.12*

2.23±0.06
0.72±0.01

2.38±0.06
0.80±0.02

2.20±0.08
0.80±0.03

2.25±0.12
1.00±0.02*

Testes/ Ovaries
% body
weight

3.42±0.08
0.63±0.02

3.66±0.06
0.65±0.02

3.40±0.07
0.80±0.04*

2.91±0.15*
1.0±0.12*

0.16±0.01
0.05±0.004

0.14±0.01
0.05±0.002

0.14±0.01
0.05±0.003

0.10±0.02*
0.05±0.01

Adapted from Hayes et al. (1986).

All data expressed as mean ± SEM.
b All absolute weights are presented in grams.
* Significantly different from vehicle control (p < 0.05).

EPA/OW/OST/HECD

V - 24

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Table V-9. Serum Chemistry Values for CD Rats Exposed to Dichloroacetonitrile (DCAN) for 90 days.3

Parameter

Vehicle
(corn oil)
Male

8

mg/kg/day
Male

33

mg/kg/day
Male

65

mg/kg/day
Male

Vehicle
(corn oil)
Female

8

mg/kg/ day
Female

33

mg/kg/day
Female

65

mg/kg/day
Female

Serum Glutamate Pyruvate
Transaminase (IU/L)

53 ± 1

45 ±2

116 ±60

53 ±4

51 ± 3

32 ±2*

27 ± 2*

35 ±3*

Serum Glutamate Oxaloacetic
Transaminase (IU/L)

195 ±28

140 ± 13

265 ± 96

163 ± 18

120 ± 14

127 ± 15

124 ± 14

126 ± 16

Alkaline Phosphatase (IU/L)

222 ± 20

320 ±31

471 ±46*

603 ± 79*

228 ± 29

226 ± 25

286 ± 30

499 ± 46*

5'-Nucleotidase (IU/L)

19 ±2

14 ± 1

22 ±3

21 ± 1

30 ±2

19 ± 1*

16 ± 1*

17 ± 1*

Protein (g/dL)

7.5 ±0.1

7.2 ±0.1

7.0 ±0.1*

6.5 ±0.1

7.1 ±0.1

7.3 ±0.1

7.1 ±0.1

6.6 ±0.1*

Albumin (g/dL)

5.6 ±0.1

6.0 ±0.1*

5.6 ±0.1*

5.4 ±0.1

6.2 ±0.2

6.0 ±0.1

6.0 ±0.1

5.5 ± 0.1*

Globulin (g/dL)

1.9 ±0.1

1.2 ±0.1

1.4 ±0.1

1.0 ±0.1

0.9 ±0.1

1.3 ±0.1

1.1 ±0.1

1.1 ±0.1

Alb/globulin ratio

3.1 ±0.4

5.3 ±0.4

4.0 ±0.4

5.0 ± 1

7.0 ± 1

5.1 ±0.4

6.0 ±0.4

5.0 ± 1

Glucose (mg/dL)

145 ±4

143 ±3

147 ±4

109 ±4*

150 ±6

127 ±5*

140 ±5

120 ± 6*

Cholesterol (mg/dL)

81 ±6

92 ±6

71 ±5

55 ±4*

73 ±3

96 ±6*

80 ±7

54 ±4*

Bilirubin (mg/dL)

0.5 ±0.0

0.5 ±0.0

0.5± 0.1

0.6 ±0.1

0.5 ±0.0

0.4 ±0.0

0.4 ±0.0

0.5 ±0.0

BUN (mg/dL)

14 ± 1

13 ± 1

14 ± 1

16 ±2

15 ± 1

16 ± 1

15 ± 1

16 ± 1

Creatinine (mg/dL)

1.1 ±0.0

1.1 ±0.0

1.2 ±0.0

1.2 ±0.0

1.1 ±0.0

1.2 ±0.1

1.2 ±0.1

1.4 ± 0.1*

BUN/creatinine ratio

13.8 ± 1

11.7 ±0.7

11.4 ±0.7

13.2 ± 1

14 ± 1

13 ± 1

12 ± 1

12 ± 1

Calcium (mg/dL)

10.1 ±0.2

9.4 ±0.1*

9.6 ±0.2

9.2 ±0.3*

9.7 ±0.1

10.1 ±0.2

9.8 ±0.2

9.8 ±0.2

Phosphorus (mg/dL)

5.7 ±0.3

5.1 ±0.2

6.1 ±0.4

5.9 ±0.3

5.0 ±0.2

5.1 ±0.2

5.4 ±0.2

5.0 ±0.2

Chloride (mEq/L)

105 ± 1

103 ± 1

103 ± 1

107 ± 1

107 ± 1

106 ±2

105 ± 1

106 ± 1

Adapted from Hayes et al. (1986).

". All data expressed as mean ± SEM.

* Significantly different from vehicle control at p < 0.05.

EPA/OW/OST/HECD

V - 25

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
C. Reproductive and Developmental Effects

No multigeneration studies were located on the reproductive effects of any of the
haloacetonitriles (HANs). No reproductive or developmental studies were identified for any of
the HANs following inhalation or dermal exposure.

Meier et al. (1985) evaluated the in vivo genotoxicity of several drinking water
disinfectants and their by-products in mice. As part of this study, the ability of a series of
disinfectants to induce sperm head shape abnormalities was examined as a measure of germ cell
mutagenicity. Of the disinfectants tested, only hypochlorite induced a dose-related increase in the
percent of abnormal sperm heads. To investigate whether this positive finding with hypochlorite
might be due to the in vivo formation of haloacetonitriles from hypochlorite, groups of 8 tol 1-
week old male B6C3F, mice (10/dose group) were administered 0, 12.5, 25, or 50 mg/kg/day of
BCAN, DBAN, DCAN, or TCAN in water by gavage for 5 days. The authors indicated that the
highest total dose of 250 mg/kg (5 days x 50 mg/kg/day) was selected to approximate the
reported LD50 values for these compounds. Positive controls received five daily doses of 200
mg/kg ethylmethanesulfonate administered intraperitoneally. Control and test animals were
sacrificed three or five weeks after the last treatment and sperm recovered from the caudae
epididymides were examined for abnormal sperm-head morphology. The study authors reported
no effects on sperm head shape abnormalities for any of the HANs at doses up to 50 mg/kg/day.
The positive control yielded an increase in sperm head abnormalities. The NOAEL for effects on
sperm head abnormalities in this study is 50 mg/kg/day for all of the HANs that were tested. The

EPA/OW/OST/HECD

V - 26

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
HANs were not tested for their potential to cause micronuclei or chromosome aberrations, since

no positive results were seen for these end points with the initial disinfectants.

R.O.W. Sciences (1997) conducted a reproductive and developmental toxicity screening
study for DBAN, Based on the results of the dose-range finding studies (described in detail in the
section on shorter-term toxicity studies), the main reproductive and developmental toxicity study
included concentrations of 0, 15, 50, or 150 ppm DBAN in the drinking water of Sprague-Dawley
rats (10 animals/dose). Male rats, 11 weeks of age on study day 1, were given treated water on
study days 6 through 34 or 35 (the day of necropsy). The estimated doses resulting from
exposure to 0, 15, 50, or 150 ppm DBAN were 0, 1.4, 3.3, and 8.2 mg/kg/day (calculated by the
study authors from body weight and water consumption data). Males were examined for clinical
signs of toxicity and body weight at various intervals during the study. At study termination,
clinical pathology (hematology and clinical chemistry), body and organ (liver, right kidney, spleen,
thymus, right testis, right epididymis, right cauda epididymis) weight measurements, and gross
and histopathology (of the same array of organs for which organ weight was determined) were
conducted. Sperm analyses for control and high-dose males included sperm motility, spermatid
head counts, sperm morphology, and chromatin structure. Sperm density was assessed for all
necropsied males. All parameters evaluated were within normal limits, unless described below.
The only concentration-related effects in males were a slight (4%) body weight decrease at 150
ppm that was not statistically significant, and statistically significant decreases in water
consumption at 50 ppm and 150 ppm. Decreased water consumption (reported on the basis of
grams/kg body weight/day) ranged from 68% to 78% for study days 8, 10, 21, and 33 in the 50

EPA/OW/OST/HECD

V - 27

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

ppm group, and from 53% to 67% in the 150 ppm exposure group. Decreased food consumption
(reported on the basis of grams/kg body weight/day) was observed at 150 ppm on study days 8
(85%) of controls) and 10 (89%> of controls), but not on study days 21 or 33. A 14% decrease in
the absolute weight of the right cauda epididymis was reported for the 50 ppm males. However,
the absence of a statistically significant effect at the highest concentration, coupled with the
absence of effects on any of the measured sperm parameters or on epididymal histopathology
suggests that this result is not biologically significant.

Female rats were divided into two groups to separately evaluate toxicity during
conception and early gestation (designated as Group A) versus toxicity during gestation through
parturition (designated as Group B). Female rats for both groups were approximately 11 weeks
of age on study day 1. Group A females (10 animals/dose) were exposed on study days 1 through
34, cohabitated with treated males on days 13 through 17, and examined on day 34 (the day of
necropsy). The estimated doses resulting from exposure to 0, 15, 50, or 150 ppm DBAN were 0,
1.8, 5.1, and 10.9 mg/kg/day (calculated by the study authors from body weight and water
consumption data). Group A females were removed from cohabitation upon detection of vaginal
sperm or a copulatory plug, or after five days in the absence of mating. Clinical signs of toxicity,
body weight, and feed and water consumption were determined at various intervals. At study
termination (corresponding to gestation day 16-20), gross necropsy was performed and the rats
were evaluated for pregnancy status, number and position of live and dead fetuses, number and
position of early and late resorptions, and number of corpora lutea. Mating, pregnancy, and
fertility indices, total number of implants, and pre- and post-implantation losses were calculated.

EPA/OW/OST/HECD

V - 28

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

No treatment-related changes were observed in any of the mating, fertility, pregnancy, or
developmental endpoints that were examined. All of the maternal parameters evaluated were
within normal limits, except for a 7% decrease in terminal body weight (non-statistically
significant) observed at 150 ppm. Feed and water consumption was also significantly decreased
in this concentration group. Feed consumption was statistically-significantly decreased only on
study day 3 (88% of controls) in the 150 ppm exposure group; slight decreases in food
consumption on other study days were not statistically significant. Water consumption was
statistically-significantly decreased in the 150 ppm exposure group on all four study days for
which the data were summarized. The decrease in water consumption relative to controls was
52% on day 3, 72% on day 5, 67% on day 21, and 71% on day 33. Slight decreases in water
consumption in the 50 ppm exposure group were not statistically significant.

Group B females (13 animals/dose) were cohabitated with males beginning on study day 1,
were separated as soon as they were sperm-positive, had a copulatory plug, or on study day 5,
and then were exposed on gestation day 6 through postnatal day (PND) 1. The estimated doses
resulting from exposure to 0, 15, 50, or 150 ppm DBAN were 0, 1.9, 5.3, and 10.8 mg/kg/day
(calculated by the study authors from body weight and water consumption data). The group B
females were examined on PND 5, and pups were examined on PND 1, 3, and 5. Clinical signs of
toxicity, body weight, and feed and water consumption were determined at various intervals
during the study. At study termination (PND 5), females were examined for terminal body
weight, gross necropsy was performed, and uterine evaluation was conducted with the number of
implantation sites and resorptions recorded. On PND 1, 3, and 5, pup weights, number of live

EPA/OW/OST/HECD

V - 29

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

and dead pups, and number of male and female pups were recorded. The anogenital distance of
pups was also measured on PND 1. All parameters were within normal control limits except for
decreased water consumption in the 50 and 150 ppm exposure groups on study days 8, 14, and
21, but not on study day 6. In the 50 ppm exposure group water consumption decreases ranged
from 72% to 83% of controls. In the 150 ppm exposure group water consumption decreases
ranged from 47% to 60%.

This study suggests that DBAN is not a reproductive or developmental toxicant at the
highest dose tested, since no treatment-related effects on reproductive or developmental
parameters were observed for males or Group A or Group B females. However, definitive
conclusions regarding the potential for DBAN to induce reproductive or developmental effects is
hampered by the fact that this was a screening study that was not designed to evaluate the full
spectrum of endpoints of interest. For example, males were not exposed during all stages of
spermatogenesis, and the pups were not evaluated for possible visceral or skeletal malformations.
The NOAEL for male reproductive effects in this study is 8.2 mg/kg/day. The NOAEL for female
reproductive and developmental effects is 10.8 mg/kg/day (the lower of the calculated doses for
Group A and Group B females exposed to the highest DBAN concentration). No LOAEL was
identified for male or female reproductive or developmental toxicity. Slight (non-statistically
significant) body weight decreases were observed in high-concentration group males and Group A
females. However, these results were not considered biologically significant because of their
limited magnitude, and the fact that they might be secondary to decreased water consumption,
rather than due to a toxicological effect. The systemic effects evaluated in males were more

EPA/OW/OST/HECD

V - 30

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
extensive than in females, and yield a critical effect level with the greatest degree of certainty.

Therefore, the NOAEL of 8.2 mg/kg/day for males is selected as the overall study NOAEL for

systemic toxicity. No LOAEL was identified.

Smith and colleagues have investigated the reproductive and developmental effects of
BCAN, DBAN, DCAN, and TCAN through a series of studies, first testing different HANs at the
maximum tolerated dose, and then conducting dose-response studies with the individual
compounds. These studies were followed by investigation of the potentially confounding role of
the tricaprylin solvent vehicle in the observed developmental toxicity. NOAELs and LOAELs are
reported for the initial studies, conducted in tricaprylin vehicle, for completeness. However, as
discussed below these NOAELs are not adequate to serve as the basis for the dose-response
assessment, since tricaprylin itself contributes to the developmental toxicity of the HANs, and the
effects due to tricaprylin cannot be separated from those that are due to the test chemical.

In the initial developmental toxicity screening studies (Smith et al., 1986; Smith el al.,
1987) sexually-mature Long-Evans rats (age not provided, 20 to 26 sperm positive rats per dose
group) were administered gavage doses on gestation days 7 to 21 at the maximum tolerated dose
for each chemical dissolved in tricaprylin: 55 mg/kg/day for BCAN, DCAN, and TCAN; 50
mg/kg/day for DBAN. Control rats received tricaprylin alone. No food and water consumption
data were reported. Duration of pregnancy, litter size, sex ratios, and litter weights were
determined at birth. Dams not delivering by day 23 of gestation were sacrificed, and the status of
their pregnancies was determined. Survival and weight gain of offspring to PND 4 were

EPA/OW/OST/HECD

V - 31

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

measured. At PND 6, litters were randomly culled to a consistent number. At weaning, the
number of pups per litter were further randomly reduced to a consistent number. The remaining
pups were monitored for growth and health until study day 41-42. At this time, all surviving pups
were sacrificed and subjected to gross necropsy, with livers, kidneys, spleens, lungs, and gonads
removed and weighed. The following parameters were evaluated: maternal toxicity (maternal
weight gain and mortality), reproductive success (percent pregnant, percent delivering viable
litters), and growth and viability of pups (live pups/litter; postnatal survival; mean birth weight;
survival, weight gain and development through necropsy; gross necropsy and organ weight
results). All parameters evaluated were within normal limits, unless described below.

Treatment-related increases in maternal deaths were observed for DBAN (15% of dams)
and for TCAN in one study group (20% of dams), although the result for TCAN was not
replicated in a second study group receiving the same dose. The study authors could not explain
the disparate results in duplicate dose groups for this endpoint, or other endpoints noted below.
Maternal deaths observed in the BCAN and DCAN groups were attributed to intubation error.
No maternal deaths were reported in the vehicle controls. All four HANs caused decreased
maternal weight gain during gestation, although the decrease was not statistically significant for
BCAN. Since the maternal weights were measured prior to delivery, decreased unadjusted
maternal weights are affected by litter resorptions and decreased pup weights. As a result, the
observed decreases in maternal weight gain cannot be attributed solely to maternal toxicity. Both
DCAN and TCAN decreased the percentage of sperm-positive females that became pregnant, but
the decrease in apparent pregnancy rate was observed only in one of two duplicate dose groups

EPA/OW/OST/HECD

V - 32

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

for each compound. According to the study authors, the apparent decrease in pregnancy rate
could reflect late preimplantation losses or very early resorptions. This conclusion was based on
the expected stage of pregnancy at the initiation of dosing on gestation day (GD) 7 and the
absence of detectable ammonium-sulfide staining decidua (which would indicate implantation
sites). DC AN and TCAN (in only one of two duplicate dose groups for TCAN) also decreased
the percentage of females delivering viable litters and increased the percentage of litters totally
resorbed (p < 0.05). Mean birth weights were reduced for all four compounds (p < 0.05), and
postnatal survival on day 4 was significantly reduced (p < 0.05) in pups from dams exposed to
DCAN (in only one of two duplicate dose groups) and TCAN. Cursory visual inspection of
nonviable pups did not reveal gross terata, although the authors noted that the current study
protocol was not expected to provide a sensitive assessment of malformations because of the
embryolethality of the single doses that were tested. No pups died after the culling of litters on
day 6 until weaning, but after weaning some pups (number not reported) from the DCAN
replicate died because they were too small to reach the water source. Weight gain to PND 4 was
significantly decreased for male and female pups for BCAN and DCAN (in one of two duplicate
dose groups), and in male pups for DBAN. Measurement of pup weights at weaning (days 21-
22) and adolescence (days 41-42) revealed significant decreases in pup body weight at both time
points for TCAN and at adolescence for DCAN (in one of two duplicate dose groups). Postnatal
weight gain from weaning until puberty and sacrifice was reduced in both male and female pups in
all groups administered HANs, but these effects were statistically significant only in males and
females receiving TCAN and females receiving BCAN. A scattering of effects in the relative
organ to body weight ratios obtained at the day 41-42 sacrifice was reported across all dose

EPA/OW/OST/HECD

V - 33

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

groups. However, these effects are judged to be without biological significance because no
uniform or consistent patterns of change were noted.

This study identifies a frank effect level (FEL) of 50 mg/kg/day for DBAN and
55 mg/kg/day for TCAN, based on increased maternal deaths. A maternal LOAEL of 55
mg/kg/day is assigned for BCAN and DCAN based on decreased maternal weight gain.

However, the contribution of developmental effects (e.g. decreased litter sizes, fetal weight) to
the observed decrease in weight gain cannot be ruled out, and therefore it is not known if the
observed effect was due to systemic toxicity, severe developmental toxicity, or both. The
LOAEL for developmental effects was 50 to 55 mg/kg for all four compounds. For DCAN and
TCAN, these LOAELs were based on severe effects (increased percent of early resorptions or
litters totally resorbed). Maternal and developmental NOAELs were not identified due to the
limited number of test doses used in this screening assay.

In a follow-up study evaluating the dose response for TCAN, Smith et al. (1988)
administered TCAN to sperm-positive Long-Evans rats aged 65-80 days (19 to 21 per dose group
of TCAN-treated animals, 30 rats in the vehicle control group, and 10 rats in water controls) by
gavage in tricaprylin at doses of 0, 1.0, 7.5, 15, 35, or 55 mg/kg/day on gestation days 6 to 18.
Dams were sacrificed on day 21 of gestation, their uterine horns were examined for number and
location of fetuses or resorption sites, and the fetuses were removed and examined. Two-thirds
of each litter was fixed for dissection and one-third stained for bone and cartilage examination.
The parameters evaluated in this study were maternal toxicity (maternal weight gain and

EPA/OW/OST/HECD

V - 34

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

mortality), embryolethality (number and location of fetuses or resorption sites, number of viable
litters, litter sizes), and the following fetal effects: weights, sex ratios, and structural abnormalities
including external, visceral, and skeletal effects. All parameters were within normal limits, unless
described below (statistical comparisons are made to the tricaprylin control unless noted
otherwise). The high dose was lethal in 4 out of 19 dams, and maternal weight gain was
decreased beginning at 15 mg/kg/day, but maternal weight gain adjusted for fetal weight and
excluding females with full-litter resorptions was significantly decreased only at 55 mg/kg/day.
The primary developmental effects were on fetal viability, malformations, and fetal body weight.
There was a dose-related increase in full-litter resorptions (compared to both water and tricaprylin
control groups) at 7.5 mg/kg/day and higher, affecting 2/3 of the surviving dams at the high dose.
Fetal weight was significantly decreased only at 35 mg/kg/day, while post-implantation loss (as
percent resorptions per litter) was significantly elevated at doses of 15 mg/kg/day and higher.
Post-implantation losses were significantly higher and male fetal body weight was significantly
lower (p < 0.05) in the tricaprylin controls compared to the water controls. The authors noted
that the value obtained for post-implantation loss for water controls in this study was lower than
the historical laboratory control levels for this endpoint, which might have enhanced the observed
effect of the tricaprylin control. While no malformations were observed in the water controls,
soft-tissue (fetal incidence of 3.8% and litter incidence of 6/30) and skeletal (fetal incidence of
13.3% and litter incidence of 7/30) malformations were observed in tricaprylin controls. The
percent of fetuses affected per litter with soft-tissue malformations was dose-dependent, ranging
from 18%) at 7.5 mg/kg/day to 35% at 35 mg/kg/day. This value decreased to 22% at the high
dose. The soft tissue malformation frequency at the low dose of 1.0 mg/kg/day (8.4% of fetuses

EPA/OW/OST/HECD

V - 35

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

affected per litter) was not statistically different from vehicle controls (3.8%), although the
authors expressed concern that this level of malformations could be of biological significance.
However, consideration of the data in terms of the more appropriate unit (U.S. EPA, 1991) of
percent of litters affected indicated no effect of TCAN compared to the tricaprylin control (4/20
litters affected at 1 mg/kg/day and 6/30 litters affected in tricaprylin controls). The incidence of
pups with cardiovascular malformations was increased at 15 mg/kg/day and urogenital
malformations were significantly increased at both 15 mg/kg/day and 35 mg/kg/day (p<0.05); the
percentage of litters with cardiovascular malformations was increased beginning at 7.5 mg/kg/day
(8/18 litters affected versus 6/30 litters in tricaprylin controls). No increase in the incidence of
external or skeletal malformations was reported, and the study authors do not appear to have
reported on the incidences of external, skeletal, or internal variations (which are more subtle
effects than malformations).

Since the apparent increase in soft tissue malformations at 1.0 mg/kg/day was not
statistically significant when compared to tricaprylin controls, and there was no effect on the
percentage of affected litters, a dose of 1.0 mg/kg/day of TCAN is the NOAEL for developmental
toxicity, and 7.5 mg/kg/day is the LOAEL, although this conclusion is limited by the absence of
reported data on the incidence of variations. The maternal NOAEL for this study is 35
mg/kg/day. The next higher dose of 55 mg/kg/day is a FEL, based on increased maternal deaths.
Although significant decreases in overall maternal body weight gain were observed beginning at
15 mg/kg/day, adjusted body weight gain was decreased only at 55 mg/kg/day. Adjusted body
weight gain is a more appropriate indicator of maternal toxicity than overall body weight gain,

EPA/OW/OST/HECD

V - 36

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

since the latter would be affected by developmental toxicity as well as maternal toxicity.
Regardless of the selection of the critical effect levels, the results of this study are not appropriate
for dose-response analysis due to the confounding effects of the tricaprylin solvent vehicle as
evidenced by the differences between water and tricaprylin controls in this study, and as described
in more detail in the study by Christ et al. (1996) for TCAN.

In a second follow-up study evaluating the dose-response for DCAN, Smith el al. (1989)
administered DCAN in tricaprylin to pregnant Long-Evans rats aged 65-80 days (22 to 24 rats per
dose group) by gavage at doses of 0, 5, 15, 25, or 45 mg/kg/day on gestation days 6 to 18.
Tricaprylin served as the vehicle control and distilled water served as an additional control group.
Dams were sacrificed on day 20 of gestation. Their livers, spleens, and kidneys were removed
and weighed, and their uterine horns were examined for number and location of fetuses or
resorption sites. Fetuses were removed and examined. Two-thirds of each litter was fixed for
dissection and one-third stained for bone and cartilage examination. The parameters evaluated in
this study were maternal toxicity (maternal body weight gain, liver, spleen, and kidney weight, and
mortality), embryolethality (number and location of fetuses or resorption sites, number of viable
litters, litter sizes), and the following fetal effects: weights, crown-rump lengths, sex ratios, and
structural abnormalities including external, visceral, and skeletal effects. All parameters were
within normal limits, unless described below (statistical comparisons are made to the tricaprylin
control unless noted otherwise). Two of 22 dams died in the high dose group (45 mg/kg/day).
No maternal deaths were observed in any other dose group, or in either of the control groups.
Maternal body weight gain and adjusted maternal body weight gain were significantly decreased

EPA/OW/OST/HECD

V - 37

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
(p<0.05) at 45 mg/kg/day. Surviving dams in the 45 mg/kg/day dose group showed elevated

spleen and kidney weights (p<0.05). Liver weight was significantly increased at 25 mg/kg/day,

but not at the high dose.

DCAN treatment affected a number of developmental toxicity parameters. Post-
implantation loss was significantly increased beginning at 25 mg/kg/day. At 45 mg/kg/day, the
following developmental effects were observed: increased number of totally resorbed litters,
decreased number of viable litters, decreased fetal weight in males and females, and decreased
crown-rump length in male and females. The incidence of malformations increased in a dose-
related manner. Significant increases (p<0.05) in total soft tissue, cardiovascular, urogenital
system, and skeletal malformations were observed in fetuses from dams exposed to 45 mg/kg/day.
Increases in the incidence of litters affected was most apparent for total soft tissue malformations:
water control 0/19; tricaprylin control 4/19; 5 mg/kg/day 5/22; 15 mg/kg/day 5/16; 25 mg/kg/day
9/16; 45 mg/kg/day 7/7. The authors did not present a statistical analysis of the malformation
data presented as the incidence of litters affected. Analysis of these data for preparation of this
Criteria Document (using Fischer's Exact Test) revealed that the litter incidence of malformations
is significantly greater (P<0.05) in the 25 mg/kg/day and 45 mg/kg/day dose groups than in
tricaprylin controls. The water controls were not significantly different than the tricaprylin
controls (P = 0.0525), based on the litter incidence, although the study authors reported a
significant difference between water and tricaprylin controls on the basis of fetuses affected per
litter.

EPA/OW/OST/HECD

V - 38

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

The NOAEL for maternal toxicity for this study is 15 mg/kg/day and the LOAEL is 25
mg/kg/day based on increased liver weight. No additional measures of liver toxicity were
included in this study, but the selection of increased liver weight as an adverse effect is supported
by ability of DCAN to induce liver toxicity at similar doses as reported in the subacute and
subchronic studies by Hayes et al. (1986). The NOAEL for developmental toxicity is
15 mg/kg/day and the LOAEL is 25 mg/kg/day, based on increased post-implantation loss and
malformations. No differences in fetal viability or size were noted between the water and
tricaprylin control groups. However, the incidence of total soft tissue malformations was
significantly lower in water (0/19 litters affected) versus tricaprylin controls (4/19 litters
affected), suggesting that tricaprylin induces developmental toxicity, and that this effect is further
potentiated following combined treatment with tricaprylin and DCAN (as high as 9/16 litters
affected). Such a finding is of considerable importance because, as mentioned in the paragraph
above introducing the Smith studies, it raises the possibility that the fetal malformations attributed
to the HANs in this study may result from the potentiation by the tricaprylin vehicle. Tricaprylin
effects and their implications for assessing HAN toxicity are discussed further in the studies
below.

The developmental toxicity of BCAN was evaluated in 120 to 150-day old Long-Evans
rats (Christ et al., 1995) in a third study to evaluate the dose-response for HANs. Pregnant rats
were administered BCAN by gavage in tricaprylin on gestation days 6 to 18 at doses of 0, 5, 25,
45, or 65 mg/kg/day. Dams treated with tricaprylin only served as the vehicle control and dams
treated with distilled water served as an additional control. The number of rats per treatment

EPA/OW/OST/HECD

V - 39

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

group is unclear, because while the methods section indicates that between 17 and 23 animals
were assigned per dose group, results for reproductive performance presented in the paper
suggest group sizes of 20 to 28 dams. Dams were sacrificed on day 20 of gestation. Their livers,
spleens, and kidneys were removed and weighed, and their uterine horns were examined for
number and location of fetuses or resorption sites. Fetuses were removed and examined. Two-
thirds of each litter was fixed for dissection and one-third stained for bone and cartilage
examination. The parameters evaluated in this study were maternal toxicity (maternal body
weight gain, liver, spleen, and kidney weight, and mortality), embryolethality (number and
location of fetuses or resorption sites, number of viable litters, litter sizes), and the following fetal
effects: weights, crown-rump lengths, sex ratios, and structural abnormalities, including external,
visceral, and skeletal effects. All parameters were within normal limits, unless described below
(statistical comparisons are made to the tricaprylin control unless noted otherwise).

Treatment with BCAN in tricaprylin resulted in both maternal and developmental toxicity.
Mortality was statistically significantly increased in the high-dose dams (3 of 26 treated dams)
compared with the water and tricaprylin control groups in which no treatment-related deaths were
observed. Dams in both the 45 and 65 mg/kg/day dose groups had significantly decreased
percentage of body weight gain compared with the tricaprylin control group, and dams in the 65
mg/kg/day group had decreased body weight gain after adjusting for gravid uterine weight.

Kidney weights were significantly increased in dams at doses >25 mg/kg/day compared with
tricaprylin controls. Liver and spleen weights were significantly increased only in the high-dose

EPA/OW/OST/HECD

V - 40

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
dams. There was no difference in any of these maternal toxicity parameters between tricaprylin

and water controls.

In terms of reproductive and developmental endpoints, an increase in the number of litters
totally resorbed and a decrease in the number of viable litters compared with tricaprylin controls
were observed beginning at 45 mg/kg/day. The percent post-implantation loss and the percent of
resorbed fetuses per litter increased at doses of >45 mg/kg/day compared with the tricaprylin
control group. Although the increase for post-implantation loss was not statistically significant in
the high-dose group, this parameter was statistically significant in the second-highest dose group,
and also substantially elevated, although not statistically, in the third-highest dose group. Fetal
crown-rump length was significantly decreased in all the treated groups, and fetal weights were
decreased beginning at 25 mg/kg/day. Statistical analysis of the malformation data was done in
terms of the percent of fetuses affected per litter. However, examination of the data on the basis
of the percent of litters affected appears to lead to similar conclusions with regard to selection of
effect levels. A significant increase in cardiovascular malformations compared with tricaprylin
controls was observed in all the dose groups. Urogenital malformations were significantly
increased only at 45 mg/kg/day, although the percentage of litters affected appeared to be
significantly greater at both 25 and 45 mg/kg/day. Total soft tissue malformations were increased
beginning at 25 mg/kg/day and skeletal malformations were significantly increased beginning at 45
mg/kg/day.

EPA/OW/OST/HECD

V - 41

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

In addition to the BCAN-treated groups, tricaprylin vehicle alone had significant effects on
embryotoxicity. Developmental endpoints affected by tricaprylin are noted as mean ± standard
error tricaprylin versus water controls. Animals in the tricaprylin control group had significantly
increased percent post-implantation loss (15.5 ± 19.0 versus 6.7 ± 9.8); decreased fetal body
weight (grams) in males (3.19 ± 0.3 versus 3.61 ± 0.2); decreased fetal body weight (grams) in
females (2.90 ± 0.3 versus 3.41 ± 0.2); decreased crown-rump length (cm) in males (3.4 ± 0.2
versus 3.6 ± 0.2); and decreased crown-rump length (cm) in females (3.4 ± 0.2 versus 3.5 ± 0.1).
An increased incidence of urogenital malformations (0/19 litters versus 4/23 litters) was also
induced by tricaprylin as compared to water controls.

The NOAEL for maternal toxicity in this study is 45 mg/kg/day and the high dose of 65
mg/kg/day is a FEL based on increased maternal deaths, and accompanied by decreased adjusted
maternal weight gain, and organ weight changes. The increase in kidney weight at the lower
doses was not used as the basis for assigning maternal toxicity effect levels because of the absence
of data to determine whether this effect was adverse. Developmental effects, including decreased
crown-rump length and increased cardiovascular malformations were observed at the the low
dose of 5 mg/kg/day. Therefore, 5 mg/kg/day is a developmental LOAEL for this study, and no
NOAEL is identified. However, based on the observation of embryotoxicity of the tricaprylin
vehicle in this study, and later work by this laboratory which suggests that tricaprylin may act
synergistically with TCAN to enhance developmental toxicity (Christ el al., 1996), use of this
study for dose-response assessment is not appropriate, because it may not accurately reflect the
toxicity of BCAN in drinking water in the absence of the tricaprylin vehicle.

EPA/OW/OST/HECD

V - 42

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Based on the observed increased embryotoxicity in tricaprylin versus water-treated
controls in earlier studies, Christ et al. (1996) investigated the effect of solvent vehicle on the
developmental toxicity of TCAN (Table V-10). Groups of approximately 20 sperm-positive
Long-Evans rats aged 65-80 days were treated with 15, 35, 55, or 75 mg/kg/day TCAN in corn
oil, or 15 mg/kg/day TCAN in tricaprylin. In addition, water, corn oil, and tricaprylin were used
as controls. Treatments were administered by oral gavage on gestation days 6 to 18 for all of the
treatment groups. Dams were sacrificed on day 20 of gestation. Their livers, spleens, and kidneys
were removed and weighed, and their uterine horns were examined for number and location of
fetuses or resorption sites. Fetuses were removed and examined. Two-thirds of each litter was
fixed for dissection and one-third stained for bone and cartilage examination. The parameters
evaluated in this study were maternal toxicity (maternal body weight gain, liver, spleen, and
kidney weight, and mortality), embryolethality (number and location of fetuses or resorption sites,
number of viable litters, litter sizes), and the following fetal effects: weights, crown-rump lengths,
sex ratios, and structural abnormalities, including external, visceral, and skeletal effects. All
parameters were within normal limits, unless described below. Of the 20 dams treated with 75
mg/kg/day TCAN in corn oil, five dams died, five were nonpregnant, and nine dams resorbed their
entire litter, so that only one viable litter was produced. For this reason, other data for the 75
mg/kg/day group were not reported. No maternal deaths were reported in any of the other
groups. Maternal weight gain was significantly decreased beginning at 15 mg/kg/day, and
maternal weight gain, after adjusting for gravid uterine weight, was significantly decreased at 35
mg/kg/day and higher doses. Relative maternal liver weight was increased at >35 mg/kg/day, and
liver, spleen, and kidney weights were significantly increased at 55 mg/kg/day as compared to

EPA/OW/OST/HECD

V - 43

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
corn oil vehicle controls. Similarly, increased relative liver weights were observed in the animals

given 15 mg/kg/day TCAN in tricaprylin as compared to 15 mg/kg/day TCAN administered in

corn oil.

The percent post-implantation loss was significantly increased, and the number of live
fetuses per litter, fetal body weight, and crown-rump length were all significantly decreased in the
group administered 55 mg/kg/day TCAN in corn-oil vehicle. Also in this dose group, the mean
percentage of fetuses per litter with external malformations, skeletal malformations, and soft-
tissue malformations was significantly increased. The incidence of cardiovascular and urogenital
malformations was not increased at any dose, but other soft tissue malformations (i.e. not
classified as either cardiovascular or urogenital) were significantly increased in the 55 mg/kg/day
dose group.

EPA/OW/OST/HECD

V - 44

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Table V-10. Reproductive and Developmental Toxicity of TCANa.







TCAN (mg/kg/d)

TCAN (mg/kg/d)



Water

Corn oil

15

35

55

75

Tricaprylin

15

No. Sperm-positive treated
females

20

20

17

21

21

20

20

21

Nonpregnant females

2

3

0

0

3

5

1

2

Deaths

0

0

0

0

0

5b

0

0

Litters totally resorbed

0

0

0

0

1

9b

0

3

Viable litters0

18

17

17

21

17

1

19

16

Percent maternal weight gaind

45.5 ±9.0

48.9 ±8.1

44.2 ± 7.8b

36.9 ±8.8b

23.7 ±7.7b

nde

41.6 ± 10.0

41.7 ± 10.5

Adjusted maternal weight
gaind

18.6 ±5.5

20.4 ±4.8

17.6 ±5.6

11.1 ± 5.7b

6.8 ± 5.4b

nde

18.1 ±5.8

17.3 ±6.7

Total implants per litter

13.6 ±1.6

14.2 ± 1.3

13.8 ± 1.3

13.4 ±2.4

13.8 ± 1.7

12.3 ±4.3

12.7 ± 2.5f

13.4 ±3.0

Percent preimplantation lossg

3.3 ±5.6

4.6 ±7.5

3.5 ±6.6

5.0 ± 13.8

3.8 ±8.3

17.7 ±
28.1

14.0 ± 13.9h

11.6 ±20.0

Percent post-implantation loss1

12.1 ±
12.4

7.0 ±7.6

6.3 ±7.0

8.4 ± 11.1

29.7 ± 25.2b

98.8 ±
4.0b

15.9 ± 17.8

25.4 ±34.8

Live fetuses per litter

11.9 ±2.1

13.2 ± 1.6

12.9 ± 1.4

12.3 ±2.8

10.4 ± 2.6b

nd

10.7 ± 3.2f

12.4 ± 1.8

Fetal body weight (g) (male)

3.58 ±
0.25

3.42 ±
0.17

3.52 ±0.20

3.38 ±0.30

2.54 ± 0.46b

nd

3.22 ± 0.28f

2.93 ±
0.37bj

Fetal body weight (g) (female)

3.41 ±
0.26

3.28 ±
0.19

3.26 ±0.22

3.25 ±0.30

2.39 ± 0.41b

nd

3.01 ±0.25h

2.76 ±
0.38bj

Crown-rump length (cm)
(male)

3.58 ±
0.08

3.57 ±
0.11

3.60 ±0.13

3.54 ±0.14

3.28 ± 0.23b

nd

3.45 ± 0.13h

3 .46 ± 0.161

EPA/OW/OST/HECD

V - 45

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Table V-10. Reproductive and Developmental Toxicity of TCANa.







TCAN (mg/kg/d)

TCAN (mg/kg/d)



Water

Corn oil

15

35

55

75

Tricaprylin

15

Crown-rump length (cm)
(female)

3.51 ±
0.09

3.52 ±
0.09

3.51 ±0.17

3.45 ±0.13

3.16 ± 0.19b

nd

3.36 ± 0.13h

3.36 ± 0.18j

Fetal malformations

















External

0

0

0

1.4 ±6.2
(1/21)k

5.3 ± 8.8b
(6/17)

nd

0

0.5 ±1.9
(1/16)

Total soft tissue (visceral)

0

1.9 ±4.3
(3/17)

0.6 ±2.4
(1/17)

0

15.0 ± 17. lb
(3/17)

nd

2.4 ±6.3
(3/19)

15.0 ±
21.2bj
(9/16)

Cardiovascular

0

0.6 ±2.4
(1/17)

0.6 ±2.4
(1/17)

0

4.5 ± 11.2
(3/17)

nd

1.3 ±5.7
(1/19)

12.0 ±
17.3bj
(7/16)

Urogenital

0

1.3 ±3.8
(2/17)

0

0

0

nd

2.4 ±6.3
(3/19)

3.0 ±7.1
(4/16)

Other soft tissues

0

0

0

0

10.4 ± 16.3b
(3/17)

nd

0

0

Skeletal

0

0

0

1.6 ±7.3
(1/21)

7.1 ± 14.4b
(2/17)

nd

0

0

a.	Adapted from Christ et. al. (1996)

b.	Significantly different from vehicle control, p < 0.05.

c.	Viable litters were those containing at least one live pup.

d.	Weight gain analysis included only females with viable litters (mean ± SD reported); adjusted
maternal weight gain controls for the effect of gravid uterine weight.

e.	Values for 75 mg/kg/day are not reported since there was only one dam with a viable litter.

f.	Significantly different from corn oil control, p < 0.05.

g.	Percent preimplantation loss = (number of copora lutea - number of implants) / number of
corpora lutea x 100.

h.	Significantly different from water or corn oil control, p < 0.05.

i.	Percent post-implantation loss = (number of implants - number of live fetuses) / number of
implants x 100.

j. Significantly different from TCAN 15/mg/kg/day in corn oil.
k. (number litters examined / number of litters affected)

EPA/OW/OST/HECD

V - 46

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

No developmental effects were observed at the doses below 55 mg/kg/day of TCAN in
corn oil. By contrast, fetal body weight was significantly decreased in the group dosed with 15
mg/kg/day TCAN in the tricaprylin vehicle and the percentage of fetuses per litter with total soft
tissue and cardiovascular malformations also was significantly increased in this group (both
comparisons relative to both the the tricaprylin control and the same dose of TCAN in corn oil).
The authors noted a shift in the spectrum of soft tissue malformations from cardiovascular
(communication or vascular defects) and urogenital effects in the groups treated with TCAN in
tricaprylin to external cranio-facial malformations and positional cardiovascular malformations
(e.g., levocardia) in the groups treated with TCAN in corn oil. Comparing the two vehicles,
TCAN administered at 15 mg/kg/day in tricaprylin produced effects, including increased liver and
kidney weights, decreased fetal weight, decreased crown-rump length, and increased percent of
fetuses with soft-tissue malformations that were not observed when TCAN was administered at
15 mg/kg/day in corn oil. When the water, corn oil, and tricaprylin control groups were
compared, no differences were observed between the water and corn oil groups. However, the
tricaprylin group had the following statistically significant changes compared to water or corn oil
groups: increased maternal kidney weight, decreased total implants per litter, increased
preimplantation loss, decreased live fetuses/litter, and decreased fetal weight and crown-rump
length.

The comparison of solvent vehicle effects in this study clearly shows that when TCAN is
administered in tricaprylin, maternal and developmental toxicity are observed at lower doses and
the spectrum of effects is changed as compared to TCAN administered in corn oil. Since corn oil

EPA/OW/OST/HECD

V - 47

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

and water control responses were not different in this study, but tricaprylin controls showed
increased toxicity for several endpoints, the data for TCAN dissolved in corn oil are judged to be
most relevant for dose-response assessment, and the data for TCAN in tricaprylin are judged as
inadequate for use in dose-response assessment. Based on this rationale, the NOAEL for
maternal toxicity in this study is 15 mg/kg/day, and the LOAEL is 35 mg/kg/day for decreased
maternal body weight after adjusting for gravid uterine weight. No developmental effects were
observed below 55 mg/kg/day TCAN in corn oil. Therefore, for developmental effects, the
NOAEL of TCAN in corn oil is 35 mg/kg/day and the LOAEL is 55 mg/kg/day. The data sets for
maternal and developmental endpoints were further analyzed to determine benchmark doses
(BMDs) according to draft EPA Guidance (U.S. EPA, 2000c) to identify alternative critical effect
levels. The results of the modeling are described in detail in Appendix A. A BMDL of 17
mg/kg/day for decreased adjusted maternal body weight gain was selected as the most appropriate
modeling result to serve as the basis for the quantitative dose-response assessment.

Based on the effects of tricaprylin alone and its ability to potentiate the toxicity of TCAN,
the use of the data from the developmental toxicity studies using this vehicle is not appropriate for
risk assessment. The mechanism responsible for the greater sensitivity and different pattern of
malformations produced by TCAN when it is administered in tricaprylin instead of water or corn
oil is not understood. However, an earlier abstract by this same laboratory (Gordon et al., 1991)
suggested that multiple exposures to TCAN in tricaprylin results in different distribution of TCAN
(or the TCAN/tricaprylin combination) to maternal tissues and embryos than occurs when TCAN
is administered in corn oil. However, as described in more detail in Chapter III (Toxicokinetics)

EPA/OW/OST/HECD

V - 48

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

the data are not sufficient to determine whether the observed differences in response when
tricaprylin was used as the solvent vehicle were due to a toxicokinetic or a toxicodynamic
interaction.

D. Mutagenicity and Genotoxicity

The genotoxicity of HANs has been tested in a variety of diverse assays. In this section
we present the study results for the HANs by study type. The results of mutagenicity assays are
described first, followed by measures of chromosome effects, and then assays of DNA damage.
Table V-12 provided at the end of this section summarizes the overall genotoxicity study results
on a chemical-by-chemical basis.

The mutagenicity of HANs has been studied by several investigators using standard
protocols or variations of the Salmonella typhimurium/mammaWan microsome mutagenesis assay.
In a report summarizing results of the USEPA Gene-Tox program, Kier et ol. (1986) presented
the results of testing DCAN in strains TA1535, TA1537, TA1538, TA100, and TA98. These
varying tester strains are used to identify a range of mutagenic target sites, where strains TA100
and TA1535 detect transitions and transversions, while strains TA98 and TA1538 detect
frameshifts and small deletions/insertions. (The study results were initially reported by Simmon et
al., 1977; and Simmon and Kauhanen, 1978). A positive result was defined as the generation of
greater revertant counts than controls at two doses and at one dose the number of the revertants
had to be 3-times greater than control for strains TA1535, TA1537, or TA1538, or twice the
control value for TA100 or TA98. Positive mutagenic activity with or without addition of S9

EPA/OW/OST/HECD

V - 49

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

activation was reported in strains TA100 and TA1535, while findings were negative regardless of
S9 activation in strains TA98, TA1537, and TA1538. Overall, DCAN was positive in this assay.

Bull et al. (1985) studied the mutagenic effects of BCAN, DBAN, DCAN, and TCAN in
the Salmonella!microsome assay using tester strains TA98, TA100, TA1535, TA1537, and
TA1538. Cells were dosed with up to 5.44 |imol/plate BCAN, up to 0.58 |imol/plate DBAN, up
to 12.4 |imol/plate of DCAN, and up to 11.7 |imol/plate TCAN. Wide ranges of doses were used
and the highest concentration equaled or exceeded the LC50 for each compound. In this assay,
BCAN was positive in strain TA100 (+S9) and in TA1535 (+S9 or - S9), while DCAN was
positive in TA98, TA100, and TA1535 regardless of S9 activation. The positive results for BCAN
and DCAN in strain TA1535 led the authors to conclude that both of these compounds can
induce base-pair substitutions. None of the HANs increased the number of revertants in
strains TA1537 or TA1538. In addition, DBAN and TCAN failed to produce dose-related
increases in the frequency of histidine revertants in any strain.

The Ames-fluctuation test was performed by exposing S. typhimurium strain TA100 to the
HANs in liquid culture with and without the addition of S9 (Le Curieux el al., 1995). The range
of doses tested, in |ig/mL, were as follows: BCAN, 0.03-10 (-S9), 0.3-100 (+S9); DBAN, 0.03-
10 (-S9), 0.1-30 (+S9); DCAN, 0.3-300 (-S9), 0.3-1000 (+S9); TCAN, 0.1-1000 (+/- S9). In all
cases the tested dose range included cytotoxic concentrations (defined as sufficient to reduce the
number of positive wells compared to controls). All of the compounds except DBAN generated
positive results (i.e. generated a statistically significant increase in the number of positive wells

EPA/OW/OST/HECD

V - 50

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

compared to controls). The effective doses ranged widely and no clear pattern of dependence on
S9 activation was apparent. BCAN yielded a positive result at a concentration as low as 0.6
|ig/mL (-S9), DCAN yielded a positive result at 10 |ig/mL (-S9) and 300 |ig/mL (+S9), and
TCAN at 30 |ig/mL (-S9).

Gee el al. (1998) conducted a validation analysis for the Salmonella!microsome assay by
comparing the effects of a group of substances in base-specific tester strains to mutagenic activity
in traditional strains. Mutagenic activity of TCAN in the base-specific strains TA7001, TA7002,
TA7003, TA7004, TA7005, and TA7006 and a mix of these six strains was contrasted to the
mutagenic activity of TCAN in the traditional frameshift strains TA98 and TA1537. The assay
was done using a liquid fluctuation exposure protocol in the presence and absence of S9
activation. For each combination of tester strains, four doses of TCAN ranging from 50 to 1000
|ig/mL in dimethyl sulfoxide were used, plus solvent control and positive control groups. The test
agent was considered mutagenic if any of the test doses were found to generate revertants at
statistically significant levels compared to the control. The only positive result reported was for
the base-specific strain mix with S9 activation. This result must be interpreted with caution,
however, because the authors did not explain why a positive result was identified in the base-
specific strain mix, when none of the base-specific strains was positive when tested individually.
In addition, inspection of the raw data (available from an online site provided in the paper) did not
reveal any clear increase in the frequency of revertants in the mix versus the individual test strains.

EPA/OW/OST/HECD

V - 51

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Taken together, these mutagenicity assays indicate that BCAN and DCAN are mutagenic
in S. typhimurium (Bull et al., 1985; Kier et al., 1986; Le Curieux et al., 1995). TCAN has
produced mixed results, with negative results reported in the standard assay protocol (Bull et al.,
1985; Le Curieux et al., 1995; Gee et al., 1998). DBAN has yielded negative results in all of the
assays reported (Bull et al., 1985; Le Curieux et al., 1995). No studies testing any of these
compounds in mammalian gene mutation assays were located.

Several investigators have tested the ability of HANs to induce chromosome damage, with
mixed results. Bull et al. (1985) studied the ability of HANs to produce chromosomal damage or
loss by examining micronuclei production in polychromatic erythrocytes in an in vivo assay in CD-
1 mice (5 animals/sex/group). Animals were dosed by gavage with BCAN, DBAN, DCAN, or
TCAN dissolved in 10% Emulphor at 0, 12.5, 25, or 50 mg/kg/day for five consecutive days and
sacrificed 6 hours after the last dose. The highest dose was selected to generate a cumulative
dose (5 doses x 50 mg/kg/day = 250 mg/kg/day) approximating the oral LD50. No significant
increases in micronuclei frequency were observed for any of the HANs tested. It is unclear,
however, whether sufficiently high doses were tested in this study. The study authors did not
present the supporting data and did not report whether cytotoxicity of the target tissue occurred
(as evidenced by a change in the ratio of polychromatic to normochromatic erythrocytes). In
addition, the shallow duration-response curve seen for the 14-day versus 90-day toxicity (Hayes
et al., 1986) suggests that the cumulative dose is not an appropriate surrogate for a single dose at
the LD50.

EPA/OW/OST/HECD

V - 52

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

In contrast to this result, other investigators have reported positive evidence for
chromosome damage using less standard assay systems. Le Curieux et al. (1995) treated
Pleurodeles waltl larvae (newt) with HANs dissolved in the container water. Following 12 days
of treatment, blood was collected and micronucleated erythrocytes were counted from a
population of 1,000 erythrocytes. The range of concentrations tested, in |ig/mL, were as follows:
BCAN, 0.0312-0.125; DBAN, 0.12-0.5; DCAN, 0.25-1; TCAN, 0.025-0.1. All the test
compounds generated a positive result in this assay, with the lowest effective concentration for
TCAN beginning at 0.1 |ig/mL, for DBAN, and BCAN at 0.12 |ig/mL, and for DCAN at 0.25
|ig/mL, Although all of the compounds significantly increased micronuclei formation, the
magnitude of this effect was characterized as relatively weak for DCAN and BCAN, since median
increases in the number of micronucleated erythrocytes was on the order of 2-fold compared to
controls. Based on results provided in a summary table, the maximum increase in the number of
micronucleated erythrocytes for TCAN was also near 2-fold, while for DBAN increases were
larger (maximum of 6.17-fold).

In another assay for chromosome damage, DCAN at an inhalation concentration of 8.6
ppm induced aneuploidy in the offspring of female Drosophila melanogaster exposed for up to
45 minutes (Osgood and Sterling, 1991). DBAN was highly toxic; a concentration of only 0.3
ppm killed 30-40% of flies after a 45-minute exposure, compared to 8.6 ppm for DCAN.
However, at a dose of 0.3 ppm, DBAN did not induce aneuploidy.

EPA/OW/OST/HECD

V - 53

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

A variety of other studies have been conducted to test whether HANs can result in DNA
damage. Bull et al. (1985) studied the ability of HANs to induce sister chromatid exchanges
(SCE), which measures a repair response to DNA damage. Chinese hamster ovary (CHO) cells
were treated with BCAN, DBAN, DCAN, or TCAN (concentrations indicated in Table V-l 1)
without exogenous metabolic activation for 2 hours, at which time 5-bromo-2-deoxyuridine
(BrdU) was added to the medium and the incubation was continued for 26 to 32 hours. The cells
with metabolic activation were treated with HANs for 2 hours in the presence of S9, at which
time the cells were rinsed and fresh medium containing BrdU was added and the incubation was
continued for 26 hours. The average SCE frequency was significantly elevated in the presence of
nonactivated or S9-activated for BCAN, DBAN, DCAN, and TCAN. Comparisons of the
potency of the HANs established the following: DBAN > BCAN > TCAN > DCAN.

Zimmermann et al. (1984) compiled and reviewed published results of genetic damage
assays in Saccharomyces cerevisiae (yeast). The only HAN compound for which data were
presented was DCAN (based on the study of Simmon et al., 1977; and Simmon and Kauhanen,
1978). Based on their analysis of the original report, Zimmerman and colleagues summarized the
results of a homozygosity assay, which serves to indicate gene recombination and gene conversion
events. The assay was conducted in a stationary culture of S. cerevesiae strain D3 with and
without metabolic activation with liver microsomes. DCAN yielded a positive result in this assay

EPA/OW/OST/HECD

V - 54

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Table V-ll. Induction of Sister Chromatid Exchange in Chinese Hamster Ovary Cells by

BCAN, DBAN, DCAN, and TCAN.

.	CHO assay without CHO assay with

. . S-9 activation	S-9 activation

concentration		

Compound	(yM)	Trial SCEs/cell3 P value SCEs/cell3 P value

DCAN 0 1	9.0±0.A	—	8.9±0.4

0.2	8.9±0.4	NS	NTC

0.6	8.9+.04	NS	NT

2.1	8.1±0.4	NS	8.1±0.4 NS

6.2	9.0+0.4	NS	8.8+0.4 NS
20.8	10.9±0.5	NS	9.4+0.4 NS
62.2	TR	—	13.2+0.5 0.01

207.5	NT	—	15.7+0.6 0.01

0

124.5

155.6

186.8

217.9
249.0

NT
NT
NT
NT
NT
NT

7.5±0.4
9.4+0.4
9.8+0.4
11. 7±0.5
11.7+0.5
12.5+0.5

NS
NS
0.01
0.01
0.01

DBAN 0 1	7. 7±0.4 —	8.0+0.4

0.06	8.8+0.4 NS	NT

0.17	10.6±0.5	0.01	8.8+0.4	NS

0.58	9.7±0.4	0.01	8.8+.04	NS

1.73	10.2±0.5	0.01	8.9+0.4	NS

5.78	NT —	9.1+0.4	NS

17.33	NT --	11.6+0.5	0.01

0

2 8.6+0.4

—

8.9+0.4

—

1.7

8.8±0.4

NS

NT

—

2.3

10.0±0.4

NS

NT

—

2.9

10.3+0.5

NS

NT

—

3.5

10.5±0.5

0.01

NT

—

4.0

10.9+0.5

0.01

NT

—

11.5

NT

—

9.1±0.4

NS

17.3

NT

—

9.3±0.4

NS

23.1

NT

—

11.7+0.5

0.01

28.9

NT

—

13.5±0.7

0.01

34.7

NT

—

18.9±0.8

0.01

continued-

^Mean±standard error.

cNS = Not significantly greater than concurrent solvent control at P=0.01.
jNT = Dose level not tested in specific assay.

TR = Insufficient number of second metaphase cells for scoring due to
toxic response.

A-V-17

EPA/OW/OST/HECD

V - 55

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

regardless of metabolic activation. In a more complex assay system in yeast, Zimmermann and
Mohr (1992) studied the effects of several agents, including DBAN, on mitotic chromosome loss
and mitotic recombination. Diploid S. cerevisiae strain D61.M, heterozygous for three recessive
alleles (cyh2, leu I, and ade6) on chromosome VII, was used to test for effects of DBAN on
chromosomal malsegregation or mitotic recombination. Chromosome loss was scored based on
the number of colonies expressing all three recessive markers, and mitotic recombination was
evaluated based on the expression of the chy2 and ade6 (but not leu I which was located between
these two markers on the chromosome). Gene expression was identified by colony formation on
the appropriate selective plates. The yeast cultures were treated with DBAN at concentrations
ranging from 0 to 18.2 mg/mL. Mitotic recombination was induced in a dose-dependent fashion.
Chromosome loss was not the reason for the expression of the recessive markers. In contrast to
this result, when yeast were treated with DBAN in combination with propionitrile, which is
known to induce chromosome loss and enhances the sensitivity of the malsegregation analysis, the
expected loss of chromosomes was observed. The authors speculated that failure to induce
malsegregation with DBAN treatment alone reflects high toxicity at doses that would induce
malsegregation.

The studies that have evaluated DNA damage at the chromosome level have resulted in
inconsistent results. No increase in micronuclei was reported for any of the HANs in a standard
assay for this end point (Bull et al., 1985), but it is unclear whether high enough doses were
tested. Positive results were reported for all four compounds in newt larvae, but this is not a well-
characterized assay system. Positive results have generally been observed in assays that measure

EPA/OW/OST/HECD

V - 56

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

responses to DNA damage, including sister chromatid exchange in CHO cells (Bull et al., 1985)
and recombination studies in yeast (Zimmermann et al., 1984; Zimmermann and Mohr, 1992).

Several studies have been conducted to evaluate DNA damage. Le Curieux et al. (1995)
studied the genotoxicity of BCAN, DBAN, DCAN, and TCAN in the SOS chromotest. This
assay measures genotoxic activity, based on induction of the SOS DNA repair system (measured
by increased P-galactosidase activity) in the Escherichia coli strain PQ37. The test was
conducted with and without S9 activation. The range of doses tested, in |ig/mL, was: BCAN, 3-
3000 (+/-S9); DBAN, 3-1000 (+/-S9); DCAN 3-1000 (+/-S9); TCAN, 3-1000 (+S9), 0.01-1000
(-S9). TCAN was negative up to cytotoxic concentrations. BCAN and DBAN were active
beginning at 5 |ig/mL and 10 |ig/mL, respectively, in the absence of S9, but were inactive in the
presence of S9 activation. DCAN generated a positive result in the presence of S9 beginning at
50 |ig/mL, and was negative in the absence of S9. The results were dose-dependent at the lower
doses, but decreased at the higher concentrations, as the doses exceeded the threshold
concentrations for cytotoxicity. The responses seen with BCAN, DBAN, and DCAN were
considered weak, based on limited induction of P-galactosidase activity.

Lin and colleagues in a series of papers have reported on the ability of HANs to induce
direct DNA damage by measuring DNA strand breaks, the ability to bind to the nucleophilic agent
(4-p-nitrobenzyl-pyridine), and formation of covalent DNA adducts (reviewed in Lin el al., 1986).
Daniel et al. (1986) reported that HANs produced DNA strand breaks in cultured human
lymphoblastic cells. The most potent initiator of DNA strand breaks compared to control cells

EPA/OW/OST/HECD

V - 57

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

was TCAN which induced 2-fold more strand breaks than the positive controls. BCAN and
DBAN had intermediate activity, while DCAN was described as having marginal activity. The
protocol description did not clearly state the concentration of the HANs at which strand breaks
were observed, but exposure to 50 |iM for 1 hour resulted in cell survival decreases ranging from
40% to 80% of control cells. The HANs also showed highly variable alkylation potential as
measured by the potential to bind to 4-p-nitrobenzyl-pyridine (Daniel el al., 1986). The relative
alkylation potential of the HANs was DBAN>BCAN»DCAN>TCAN. The range in reactivity
toward 4-p-nitrobenzyl-pyridine varied by 627-fold among the HANs.

Daniel et al. (1986) also reported the results of DNA adduct analysis. In a cell-free
system, [14C]-DCAN was incubated with calf thymus DNA, the DNA was precipitated,
hydrolyzed and separated by HPLC. Similar elution peaks were observed between the hydrolyzed
calf thymus DNA and incubations of 14[C]-DCAN with polyadenylic acid or polyguanylic acid,
suggesting that DCAN forms an adduct with these nucleotides. Oral administration of DCAN or
DBAN to rats did not result in detectable adduct formation in liver DNA (no supporting data
were presented) (Lin et al., 1986). In a subsequent study, Lin el al. (1992) reported the
formation of DNA adducts following gavage dosing of radiolabeled TCAN to male F344 rats. A
single oral gavage dose of either [1-14C]- or [2-14C]-TCAN (in tricaprylin, 7.2-69.3 mg/kg) was
administered to male F344 rats and tissues were analyzed from 2 to 48 hours following dosing.
TCAN bound to both DNA and proteins and DNA binding was highest in the stomach, followed
by liver and kidney.

EPA/OW/OST/HECD

V - 58

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Overall, the data suggest that HANs can directly damage DNA as evaluated by a wide
array of assays (summarized in Table V-12). The weight of the evidence varies for each
compound. BCAN has yielded positive results in all assays tested. DBAN yielded negative
results in S. typhimurium mutation assays and failed to form DNA adducts in vivo. DBAN
appears to induce DNA strand breaks and yield positive results in assays that reflect responses to
DNA damage (i.e. SCE, gene conversion, and SOS assays). The results for DCAN and TCAN
are less consistent. DCAN yielded positive results in S. typhimurium mutation assays and assays
reflecting DNA recombination, but the reason for absence of significant effects in the DNA strand
break assay is not clear. For TCAN, the weak responses in S. typhimurium mutation assays did
not correspond well with the observed in vivo formation of DNA adducts, although the positive
results in the DNA strand break assay and SCE assay were consistent.

Table V-12. Summary of Results of Genotoxicity and Tumor
Screening Assays for Haloacetonitriles.

BCAN

DBAN

DCAN

TCAN

Mutation assays (S. typhimurium)

Micronuclei

Aneuploidy (I). Melanogaster)

Sister Chromatid Exchange

Gene Conversion/Recombination

(S. cerevisiae)

SOS chromotest

DNA Strand Breaks

DNA adducts (in vivo)

Lung tumors (A/J mice)

Skin tumors (Senear mice)

nt = not tested

EPA/OW/OST/HECD

V - 59

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

The evidence for the induction of chromosome damage by HANs is less compelling, due
to the limited number of studies available for evaluation and the inconsistent results. In the single
study that used a standard assay protocol to evaluate induction of micronuclei, no effect was
observed for any of the HANs, although is was not clear that sufficiently high doses were used.
In contrast, positive results for micronuclei formation were reported for all four compounds in a
less well characterized newt larvae system. DCAN, but not DBAN, induced aneuploidy in the
Drosophila melanogaster assay system.

E. Carcinogenicity

No 2-year carcinogenicity bioassays have been conducted for any of the HANs by any
route of exposure. No alternative carcinogenicity studies were identified for any the HANs by the
inhalation route. There are, however, several short-term assays that can aid in hazard
identification. In addition, DBAN is currently under test in a full cancer bioassay by NTP (2002).

In a published review paper, Bull and Robinson (1985) reported studies on the incidence
of lung tumors in groups of 40 female A/J mice (10 weeks of age) that were administered a single
oral dose of 10 mg/kg of BCAN, DBAN, DCAN, or TCAN, three times weekly for 8 weeks
(Table V-13). Control groups received the vehicle only (10% emulphor) or ethyl carbamate
(positive control). As discussed in Chapter VII, emulphor solutions have generally been deemed
as more appropriate solvent vehicles than corn oil for disinfectant byproducts. All animals were
sacrificed at 9 months of age, allowing for an approximately 6 months post-exposure observation
period. The incidence of lung tumors (adenomas) was significantly increased in groups given

EPA/OW/OST/HECD

V - 60

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

BCAN and TCAN (p < 0.05). DBAN and DCAN produced marginal, but nonsignificant (p >
0.05) increases in lung tumors. The authors stated that the results should be interpreted with
caution, since there is a relatively large variation in the background incidence of lung tumors in
this strain of mice and the 10 mg/kg dose level was considerably below the maximum tolerated
dose, decreasing the reliability of the negative findings with DBAN and DCAN.

Table V-13. Effects of Orally Administered Haloacetonitriles on the Development of

Lung Adenomas in Female A/J Mice.

Number of %Animals Tumors/
Chemical Dosea animals necropsied w/tumors animal

Vehicle (emulphor 10%)

0.2 mL/mouse x24

0.10

BCAN

10 mg/kg x24

0.34b

DBAN

10 mg/kg x 24

0.19

DCAN

10 mg/kg x 24

0.23

TCAN

10 mg/kg x 24

0.38b

Ethyl carbamate
(positive control)

42 mg/kg x 24

100

9.00

a. Forty female strain A/J mice were administered the indicated doses of each chemical three times weekly for a
period of 8 weeks. Treatment was begun at 10 weeks of age. Animals were sacrificed at nine months of age.

b. Significantly increased above controls at P<0.05.

Adapted from Bull and Robinson (1985).

Bull et al. (1985) studied the ability of BCAN, DBAN, DCAN, and TCAN to induce
tumors in mouse skin the ability of dermally-applied HANs to act as tumor initiators was studied
using a tumor initiation/promotion protocol. Six topical doses of 0, 200, 400, or 800 mg/kg HAN
dissolved in acetone were applied to the shaved backs of female Senear mice (40 animals/dose
group) over a two-week period, for total doses of 0, 1200, 2400 and 4800 mg/kg. Beginning two

EPA/OW/OST/HECD

V - 61

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

weeks after the last HAN dose, 1.0 |ig of 12-O-tetradecanoylphorbol-13-acetate (TPA) was
applied three times per week for 20 weeks. Papilloma incidence and regression were recorded on
a weekly basis. Animals were then maintained for 1 year and sacrificed to determine the incidence
of squamous cell carcinomas. The results from this initiation/promotion study are presented in
Table V-14. These data were compiled by the study author from three independent experiments
as indicated in the column of the table labeled "Experiment No".

Table V-14. Histopathological Diagnosis of Tumors Resulting from Topical Treatment
with Halogenated Haloacetonitriles in the Senear Mouse









Number of animals

Squamous cell tumors





Experiment

Total Dose1

TV*

with squamous cell

diagnosed

% Animals

Treatment

No.

(mg/kg)



tumors





bearing









Number

%

Papilloma

Carcinoma

carcinomas

Acetone

1

0.2 ml X 6

34

1

3

0

1

2.9



2

0.2 ml X 6

37

3

8

3

0

0



3

0.2 ml X 6

34

5

15

1

4

11.7

BCAN

2

1200

35

1

3

0

1

2.9



2

2400

37

7

19

0

7

18.9*



2

4800

37

8

22*

3

6

16.2*

DBAN

1

1200

36

8

22*

6

2

5.6



2

2400

35

17

49**

9

8

22.9**



3

2400

35

16

45**

7

9

25.7**



2

4800

37

7

19

5

2

5.4



3

4800

37

3

11

1

2

7.4

DCAN

1

1200

39

4

10

4

0

0



2

2400

35

4

11

2

3

8.6



2

4800

35

1

3

1

0

0

TCAN

1

1200

34

2

6

1

1

2.9



2

2400

36

11

31**

5

6

16.7*



3

2400

38

1

3

0

1

2.6



2

4800

36

3

8

2

1

2.8



3

4800

29

2

7

1

1

3.4

EPA/OW/OST/HECD

V - 62

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

" Total doses of haloacetonitrile shown were delivered topically in six equal doses over a 2-week period. Then 2
weeks later animals were treated with 1.0 |ig 12-O-tctradccanoylphorbol-13-acetate (TPA) in 1.2 mL acetone
topically 3 times weekly for 20 weeks.

b There were 40 animals initiated in each group, N = number available for histological examination.
* Significantly different from the control, p < 0.05, Fisher exact test.

** Significantly different from control, p < 0.01, Fisher exact test.

Adapted from Bull et al. (1985).

Both BCAN and DBAN induced dose-related increases in the percent of animals with
squamous cell tumors (i.e., combined papillomas and carcinomas) (during the first 24 weeks after
TP A application was started) and the percent of animals with carcinomas (at 1.5 years)(Table V-
14). The study authors indicated that the decreased tumor response for DBAN at the high dose
as compared to the low and mid-doses may have resulted from the severe irritation and
ulcerations induced by this treatment. For TCAN, an increase in the percent of animals with
squamous cell tumors and in carcinomas was observed at the mid-dose (in experiment 2), but this
increase was not replicated in experiment 3 at this dose. The combined data for TCAN did not
yield a significant increase in tumor response. No significant increase in papillomas or carcinomas
was observed with DCAN. A similar pattern of results for each of the HANs was obtained when
the number of tumors/animal was evaluated as the response metric. Based on these results, the
authors concluded that DBAN is the most potent mouse skin tumor initiator of the HANs tested.
DBAN is followed in potency by BCAN, while TCAN and DCAN were judged as ineffective
inducers of skin tumors in this screening bioassay.

Bull et al. (1985) conducted a second series of studies, designed to assess the ability of
orally-administered HANs to act as tumor initiators. In this study, total oral doses of 50 mg/kg
were administered to female Senear mice six times over a 2-week period. The promotion phase

EPA/OW/OST/HECD

V - 63

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

of the study was conducted using the same procedures as for the dermal-dosing
initiation/promotion study described above. No statistically-significant increase in tumor yield or
decrease in the time-to-tumor was observed for any of the HANs when the data across all oral
experiments were combined. Sporadic increases in tumor yield at various times for individual
HANs were noted (data not shown by the authors), but were not judged by the study authors to
be biologically significant. The positive control, 300 mg/kg urethane, increased the yield of
papillomas.

In a third cancer screening study presented in the paper by Bull et al. (1985), the ability of
dermally-applied HANs to act as complete carcinogens was assessed. For this study, BCAN,
DCAN, or TCAN (800 mg/kg), or DBAN (400 mg/kg) was applied to the skin of female Senear
mice 3 times per week for 24 weeks, and the number of squamous cell tumors recorded. None of
the HANs induced skin tumors in this assay (data not shown by the authors).

DBAN, DCAN, and TCAN were tested for initiating activity using the rat liver
gamma-glutamyltranspeptidase foci (GGT-foci) assay in F344 rats as an indicator of
carcinogenicity (Herren-Freund and Pereira, 1986; Lin el al., 1986). The protocol for this assay
consists of a two-thirds partial hepatectomy followed 18 or 24 hours later by the administration of
the initiator. The doses employed were 2.0 mmol/kg (398 mg/kg) for DBAN, 2.0 mmol/kg (220
mg/kg) for DCAN, and 1.0 mmol/kg (144 mg/kg) for TCAN. Seven days after initiation, the rats
received the promoter (500 ppm sodium phenobarbital in their drinking water) for at least ten
weeks. The halogenated acetonitriles were inactive as initiators in the GGT-foci assay.

EPA/OW/OST/HECD

V - 64

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

The existing data provide at best only marginal support for the conclusion that HANs are
carcinogenic. The evidence is stronger for BCAN, which increased tumor yields in both lung
tumor and dermal screening assays. DBAN was positive at non-ulcerative doses in the dermal
screening assay. TCAN was positive only in the lung assay, and DCAN treatment did not
increase either lung or skin tumors. Opposing these positive findings are the negative results for
DBAN, DCAN, and TCAN in the GGT foci assay. Overall, the data are insufficient to
qualitatively or quantitatively assess the carcinogenic potential of any of the HANs. The positive
results in two tumor screening assays, together with positive bacterial gene mutation results,
suggest that it would be worthwhile to conduct a full 2-year bioassay for BCAN. Results for the
other HANs are mixed, with inconsistencies between the genotoxicity and tumor screening data.

F. Summary

The toxicity data on the HANs are summarized in Tables V-15 through V-18. Overall,
very little data are available evaluating the non-cancer effects of the HANs. Acute oral LD50
values for DBAN, DCAN, and TCAN in rodents have been reported to range from 50 to 361
mg/kg. DBAN and TCAN have been reported to be irritants. DBAN causes eye, nasal, and
respiratory tract irritation following inhalation, and skin irritation following dermal exposure.
TCAN also causes skin irritation following dermal exposure. No data on the acute toxicity of
BCAN are available, and no subacute or subchronic studies of either BCAN or TCAN are
available. No chronic studies have been conducted on any of the HANs.

EPA/OW/OST/HECD

V - 65

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

No target organ has been clearly established for HANs following oral exposure, although
absolute and relative organ weight changes, including decreased testes weight (NTP, 2002) and
increased liver weight (Hayes et al., 1986; Christ et al., 1996) have been reported. Fourteen-day
or longer systemic toxicity studies have been conducted in mice and rats. In 14-day and 90-day
studies of DBAN (NTP, 2002; Hayes et al., 1986), consistent, compound-related, dose-
dependent effects were limited to decreased water consumption, decreased body weight, and
decreased testes weight and pathology. However, effects on the testes reported in the NTP
(2002) study were observed only in rats in the 14-day study. No effects on the testes were
observed in rats in the 13-week study, or in mice (NTP, 2002). In addition, no effects were
observed on the testes in rats in a 14-day or 90-day gavage study in rats, even at much higher
doses (Hayes et al., 1986). For DBAN, the observed liver weight increases reported in Hayes et
al. (1986) were not supported by other measures of liver toxicity in the same study or observed in
the more recent NTP study (NTP, 2002), and therefore, this endpoint was not selected as the
basis for the quantitative dose-response assessment. Taken together, the data suggest that
decreased body weight appears to be the primary indicator of toxicity for DBAN. Overall, male
rats appear to be more sensitive than female rats for DBAN. For DC AN, consistent, compound-
related, dose-dependent effects were limited to decreased body weight and increased liver weight
(Hayes et al., 1986). In this case, the observed liver weight increases were supported by changes
in serum biochemistry parameters suggestive of liver damage. No histopathological evaluation
was done in the key study for DCAN (Hayes et al., 1986), so the degree, if any, of liver damage
can not be confirmed. The data for TCAN and BCAN are too limited to identify with confidence
any potential target organs.

EPA/OW/OST/HECD

V - 66

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

The data are inadequate to determine whether HANs are reproductive toxicants. No
multigeneration reproductive toxicity study has been conducted. BCAN, DBAN, DCAN, and
TCAN at doses of up to 50 mg/kg/day had no effect on sperm morphology (Meier el al., 1985),
but the data on testes weight changes are mixed (NTP, 2002; Hayes et al., 1986). DBAN at
doses up to approximately 10 mg/kg/day had no effect on any male or female reproductive
parameter evaluated in a screening assay (R.O.W Sciences, 1997). A series of developmental
toxicity studies in rats has also been conducted. Exposure to BCAN and DBAN on gestation
days 7 to 21 resulted in reduced mean birth weight (Smith et al., 1986; Smith el al., 1987). It
addition to this effect, DCAN and TCAN decreased the percentage of females delivering viable
litters and increased fetal resorptions (Smith et al., 1986; Smith et al., 1987; Smith et al., 1988;
Smith et al., 1989; Christ et al., 1996). DCAN and TCAN also significantly increase the
frequency of malformations in fetuses (Smith et al., 1988; Smith et al., 1989; Christ et al., 1996).
These studies by Smith and colleagues on the developmental toxicity of HANs in rats were
conducted using tricaprylin as a vehicle, because these compounds are very miscible in this
vehicle. However, in these studies, comparison of tricaprylin versus water-treated controls
revealed increased embryotoxicity due to tricaprylin. A recent study by Christ et al. (1996)
indicates that tricaprylin also influences the pattern of malformations observed in fetuses caused
by TCAN. For TCAN in corn oil, the malformations were primarily cranio-facial in nature while
for TCAN in tricaprylin the malformations were primarily cardiovascular and urogenital in nature.
Therefore, the use of data from studies in which tricaprylin was used as the vehicle is not
appropriate for risk assessment purposes. In the one developmental toxicity study that used a

EPA/OW/OST/HECD

V - 67

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

vehicle other than tricaprylin (Christ el al., 1996), maternal toxicity was observed at lower doses
than developmental effects.

EPA/OW/OST/HECD

V - 68

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

The existing data provide at best only marginal support for the conclusion that HANs are
carcinogenic. The evidence is stronger for BCAN, which increased tumor yields in both lung
tumor and dermal screening assays (Bull and Robinson, 1985; Bull et al., 1985). DBAN was
positive at non-ulcerative doses in the dermal screening assay. TCAN was positive only in the
lung assay, and DCAN treatment did not increase either lung or skin tumors. Opposing these
positive findings are the negative results for DBAN, DCAN, and TCAN in the GGT foci assay
(Herren-Freund and Pereira, 1986). Overall, the data are insufficient to qualitatively or
quantitatively assess the carcinogenic potential of any of the HANs. The positive results in two
tumor screening assays, together with positive bacterial gene mutation results, suggest that it
would be worthwhile to conduct a full 2-year bioassay for BCAN. DBAN is currently on test for
a full cancer bioassay (NTP, 2002). Results for the other HANs are more mixed, with
inconsistencies between the genotoxicity and tumor screening data.

EPA/OW/OST/HECD

V - 69

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Table V-15. Summary of Oral Studies of BCAN Toxicity.

Reference

Species/
Strain

Route

Exposure
Duration

Endpoints
Evaluated

NOAEL
(mg/kg/day)

LOAEL (mg/kg/day)

Meier et
al. (1985)

Mouse-
B6C3F1

Gavage in
water

0, 12.5,25,
or 50

mg/kg/day

5 Days

Sperm head
abnormalities

50 (Free-
standing
NOAEL)

NDa

Smith et
al. (1987)

Rat-

Long-
Evans
Hooded

Gavage in
tricaprylinb

55

mg/kg/day

Days 7 to 21
of gestation

Maternal weight,
reproductive
success, pup
viability and
growth

Maternal: ND

Developmental:
ND

Maternal: ND
(Nonsignificant
decrease maternal
weight gain)

Development: 55
(Decreased birth
weight, decreased
postnatal weight gain)

Christ et
al. (1995)

Rat-

Long-
Evans

Gavage in
tricaprylinb

0, 5,25,

45,65

mg/kg/day

Days 6 to 18
of gestation

Maternal body and
organ weight,
reproductive
success, pup
viability and
growth,
malformations

Maternal: 45

Developmental:
ND

Maternal: 65 (FEL for
maternal death;
decrease maternal
weight gain)

Development: 5
(Decreased crown-
rump length,
increased
cardiovascular
malformations)

a.	ND = not determined.

b.	Due to the demonstrated ability of tricaprylin to enhance developmental toxicity of TCAN, this study was not considered
in derivation of the Health Advisories.

EPA/OW/OST/HECD

V - 70

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Table V-16. Summary of Oral Studies of DBAN Toxicity.

Reference

Species/
Strain

Route

Exposure
Duration

Endpoints Evaluated

NOAEL
(mg/kg/day)

LOAEL
(mg/kg/day)

Hayes et al.
(1986)

Mouse-
B6C3F1

Gavage in
corn oil

25 - 3,200
mg/kg/day

Acute

Lethality

NDa

LD50 = 289 (M)
303 (F)

Rat-
CD

Gavage in
corn oil

25 - 1,600
mg/kg/day

Acute

Lethality

ND

LD50 = 245 (M)
361 (F)

Eastman
Kodak Co.
(1992)

Mouse
Unspecified

Gavage

25 - 1,600
mg/kg/day

Acute

Lethality

ND

LD50 = 50

Rat

Unspecified

Gavage

25 - 3200
mg/kg/day

Acute

Lethality

ND

LD50 = 50 - 100

Meier et al.
(1985)

Mouse-
B6C3F1

Gavage in
water

0, 12.5,25,
or 50

mg/kg/day

5 Days

Sperm head
abnormalities

50 (Free-
standing
NOAEL)

ND

R.O.W

Sciences

(1997)

Rat-

Sprague-
Dawley

Drinking
Water

0,0.7,2.2,
5.8, 13.2
mg/kg/day
(males)

0,0.8,2.4,
6.8, 17.9
mg/kg/day
(females)

14 Days

Clinical signs, body
weight, food
consumption

13.2 (m);
17.9(f)
(Free-
standing
NOAEL)

ND

Hayes et al.
(1986)

Rat-
CD

Gavage in
corn oil

0, 23,45,
90, 180
mg/kg/day

14 Days

Body weight, organ
weight, serum
chemistry, hematology,
urinalysis, gross
necropsy

23

45 (Decreased
body weight in
males)

EPA/OW/OST/HECD

V - 71

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Reference

Species/

Route

Exposure

Endpoints Evaluated

NOAEL

LOAEL



Strain



Duration



(mg/kg/day)

(mg/kg/day)



Rat-

Gavage in

90 Days

Body weight, organ

23

45 (Decreased





corn oil



weight, serum



body weight in



CD





chemistry, hematology,



males)





0, 6, 23,45



urinalysis, gross









mg/kg/day



necropsy





NTP (2002)

Mice-

Drinking

14 Days

Clinical signs, body

21 (Free-







Water



weight, water

standing





B6C3F1





consumption, organ

NOAEL)







0,2.1,4.3,



weight and pathology,









8.2, 14.7,



liver GST activity









21.4













mg/kg/day













(Males)













0,2.0, 3.3,













10.0, 13.9,













21.6













mg/kg/day













(Females)











Rat-

Drinking

14 Days

Clinical signs, body

12 (m)

18 (Decreased





Water



weight, water



body weight,



Fischer-344





consumption, organ



decreased testes





0, 2, 3, 7,



weight and pathology,



weight and





12, 18



liver GST activity



pathology in





mg/kg/day







males)





(Males)













0, 2, 4, 7,













12, 19













mg/kg/day













(Females)











Mice-

Drinking

13 Weeks

Clinical signs, body

17.9 (Free-







Water



weight, water

standing





B6C3F1





consumption, organ

NOAEL)







0, 1.6, 3.2,



weight and pathology,









5.6, 10.7,



hematology and clinical









17.9



chemistry









mg/kg/day













(Males)













0, 1.6, 3,













6.1, 11.1,













17.9













mg/kg/day













(Females)









EPA/OW/OST/HECD

V-72

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Reference

Species/

Route

Exposure

Endpoints Evaluated

NOAEL

LOAEL



Strain



Duration



(mg/kg/day)

(mg/kg/day)



Rat-

Drinking

13 Weeks

Clinical signs, body

11.3 (m);







Water



weight, water

12.6(f)





Fischer- 344





consumption, organ

(Free-







0,0.9, 1.8,



weight and pathology,

standing







3.3,6.2,



hematology and clinical

NOAEL)







11.3



chemistry









mg/kg/day













(Males)













0, 1, 1.9,













3.8,6.8,













12.6













mg/kg/day













(Females)









R.O.W

Rat-

Drinking

(M) 30

(M) Clinical pathology,

Paternal: 8.2

ND

Sciences



Water

Days, (F)

organ weight, sperm

(M); 10.8 (F)



(1997)

Sprague-



35 days

analysis, histopathology:







Dawley

0, 1.4, 3.3,

periconcep

(F) maternal weight,

Reproductive







8.2

tion or 35

reproductive success,

/development







mg/kg/day

days

pup viability and growth

al: 8.2 (M);









gestation



10.8(F)









day 5 to













PND 1



(Free-













standing













NOAEL)



Smith et al.

Rat-

Gavage in

Gestation

Maternal weight,

Maternal: ND

Maternal: 50

(1987)



tricapyrlinb

days 7 to

reproductive success,



(FEL for



Long- Evans



21

pup viability and growth



maternal death;



Hooded

50







decrease





mg/kg/day







maternal weight











Development

gain)











al: ND















Development:













50 (Decreased













litter size,













decreased fetal













weight)

a.	ND = not determined.

b.	Due to the demonstrated ability of tricaprylin to enhance developmental toxicity of TCAN, this study was not considered in derivation of the
Health Advisories.

EPA/OW/OST/HECD

V - 73

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Table V-17. Summary of Oral Studies of DCAN Toxicity.

Reference

Species/
Strain

Route

Exposure
Duration

Endpoints Evaluated

NOAEL
(mg/kg/day)

LOAEL (mg/kg/day)

Hayes et
al. (1986)

Mouse-
B6C3F1

Gavage in
corn oil

25 - 3,200
mg/kg/day

Acute

Lethality

NDa

LD50 = 270 (M)
279 (F)



Rat-
CD

Gavage in
corn oil

25 - 1,600
mg/kg/day

Acute

Lethality

ND

LD50 =339 (M)
330 (F)



Rat-
CD

Gavage in
corn oil

0, 12, 23,
45, 90
mg/kg/day

14 Days

Body weight, organ
weight, serum
chemistry,
hematology,
urinalysis, gross
necropsy

ND

12 (Increased liver
weight)



Rat-
CD

Gavage in
corn oil

0, 8,33,65
mg/kg/day

90 Days

Body weight, organ
weight, serum
chemistry,
hematology,
urinalysis, gross
necropsy

ND

8 (Increased liver
weight)

Meier et
al. (1985)

Mouse-
B6C3F1

Gavage in
water

0, 12.5,25,
or 50

mg/kg/day

5 Days

Sperm head
abnormalities

50 mg/kg

(Free-standing
NOAEL)

ND

Smith et
al. (1987)

Rat-

Long-
Evans
Hooded

Gavage in
tricapyrlinb

55

mg/kg/day

Gestation
days 7 to
21

Maternal weight,
reproductive success,
pup viability and
growth

ND

Maternal: 55
(Decreased maternal
weight)

Development: 55
(Decreased pregnancy
rate; decreased viable
litters; increased
litters resorbed;
decreased fetal
weight)

EPA/OW/OST/HECD

V - 74

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Reference

Species/
Strain

Route

Exposure
Duration

Endpoints Evaluated

NOAEL
(mg/kg/day)

LOAEL (mg/kg/day)

Smith et

Rat-

Gavage in

Gestation

Maternal weight,

Maternal: 15

Maternal: 25

al. (1989)



tricapyrlinb

days 6 to

reproductive success,



(increased liver



Long-



18

pup viability and



weight)



Evans

0, 5, 15,



growth







Hooded

25,45
mg/kg/day





Developmental:
15

Development: 25
(Increased post-
implantation loss,
increased soft-tissue
malformations)

a.	ND = not determined.

b.	Due to the demonstrated ability of tricaprylin to enhance developmental toxicity of TCAN, this study was considered in
derivation of the Health Advisories.

EPA/OW/OST/HECD

V - 75

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Table V-18. Summary of Oral Studies of TCAN Toxicity.

Reference

Species/
Strain

Route

Exposure
Duration

Endpoints
Evaluated

NOAEL
(mg/kg/day)

LOAEL (mg/kg/day)

Smyth et
al. (1962)

Rat-
Wistar

Gavage

0.19-0.32
mg/kg/day

Acute

Lethality

NDa

LD50 = 360

Meier et
al. (1985)

Mouse-
B6C3F1

Gavage in
water

0, 12.5,25,
or 50

mg/kg/day

5 Days

Sperm head
abnormalities

50 mg/kg (Free-
standing
NOAEL)

ND

Smith et
al. (1987)

Rat
Long-
Evans
Hooded

Gavage in
tricaprylinb

55

mg/kg/day

Days 7 to
21 of
gestation

Maternal weight,
reproductive
success, pup
viability and
growth

Maternal: ND

Developmental:
ND

Maternal: 55 (FEL for
maternal death; decrease
maternal weight gain)

Development: 55
(Decreased pregnancy
rate; decreased viable
litters; increased litters
resorbed; decreased fetal
weight)

Smith et
al. (1988)

Rat
Long-
Evans
Hooded

Gavage in
tricaprylinb

0, 1,7.5,
15, 35, 55
mg/kg/day

Days 6 to
18 of
gestation

Maternal weight,
reproductive
success, pup
viability and
growth,
malformations

Maternal: 35
Developmental: 1

Maternal: 55 (FEL for
maternal death; decrease
maternal weight gain)

Developmental: 7.5
(Increased full-liter
resorptions; increased
cardiovascular
malformations)

Christ et
al. (1996)

Rat

Long-

Evans

Gavage in
corn oilc

0, 15, 35,

55,75

mg/kg/day

Days 6 to
18 of
gestation

Maternal body and
organ weight,
reproductive
success, pup
viability and
growth,
malformations

Maternal: 15

Developmental:
35

Maternal: 35 (Decreased
maternal weight gain;
organ weight changes)

Development: 55
(increased post-
implantation loss,
cardiovascular and
cranio-facial
malformations;
decreased live fetuses
per litter, fetal body
weight, crown-rump
length.

EPA/OW/OST/HECD

V - 76

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

a.	ND = not determined

b.	Due to the demonstrated ability of tricaprylin to enhance developmental toxicity of TCAN, this study was not considered
in derivation of the Health Advisories.

c.	Only data relating to the corn oil control are reported in the table, since the developmental toxicity reported in the groups
administered tricaprylin were not considered in derivation of the Health Advisories.

EPA/OW/OST/HECD

V - 77

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
Chapter VI. Health Effects in Humans

Although most studies of human health effects following exposure to water disinfectant
byproducts have used total trihalomethanes as the exposure metric, Klotz and Pyrch (1999),
conducted a case-control study on the relationship between neural tube defects and drinking water
exposure to trihalomethanes, HANs, and haloacetic acids. The study included 112 eligible cases
of neural tube defects in 1993 and 1994 that were identified through the New Jersey Birth Defect
and Fetal Death Registries. A total of 248 controls were selected randomly from all New Jersey
births with approximately ten controls selected for each month over 24 months. While a
statistically significant prevalence odds ratio (POR) was reported for the highest tertile (third) of
trihalomethane exposure, only a slight non-statistically significant excess risk (POR 1.3: 95%
confidence interval 0.6-2.8 for the mid tertile, and POR 1.3: 95% confidence interval 0.6-2.5 for
the upper tertile) was found for cases when analyzed based on total HAN tertiles. The specific
compounds that were measured as part of the total HAN exposure estimate were not identified.
Based on the results of the study, the authors concluded that the HANs did not exhibit a clear
association with neural tube defects.

EPA/OW/OST/HECD

VI-1

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

No epidemiological studies have evaluated directly the carcinogenic potential of HANs in
humans. Rather, studies have evaluated the carcinogenic potential of chlorinated versus
unchlorinated drinking water or the presence of trihalomethanes as a marker of chlorination by-
products (IARC, 1999; Mills el al., 1998). Many of these studies have shown an association
between chronic exposure to chlorinated water and increased risks of bladder, rectal, or colon
cancers (Mills et al., 1998; WHO, 2000).

EPA/OW/OST/HECD

VT-2

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
Chapter VII. Mechanism of Toxicity and Sensitive Subpopulations

A. Biochemical Basis of Toxicity

Both DBAN and DCAN are general systemic toxicants, with depressed body weight
identified as a common adverse effect (NTP, 2002; Hayes et al., 1996). Although other organ
weight changes were observed at doses associated with decreased body weight, the liver is an
organ for which dose-dependent effects on organ weight were supported by other measures of
toxicity. Increased enzyme levels (e.g., ALP) in the serum were observed at doses higher than
those that induced liver weight changes (Hayes et al., 1986), suggesting that DBAN and DCAN
might induce hepatocellular necrosis. However, ALP is not a liver specific enzyme, and changes
in the liver specific enzymes SGPT and SGOT were not consistently dose-related. In addition,
Hayes et al. (1986) did not perform histopathological examinations, and therefore, cellular
changes resulting in increased liver weight could not be determined. The data suggest that DCAN
may be the more potent liver toxicant compared to DBAN, since the observed effects in the
gavage study (Hayes et al., 1986) for DCAN were more severe than for DBAN. Furthermore,
DBAN administered in drinking water did not induce liver toxicity, other than a treatment-related
increase in liver GST activity (NTP, 2002). However, the highest doses in the drinking water
study (NTP, 2002) were similar to the NOAEL in the Hayes et al. (1986) study. None of the
available data are adequate to determine mechanisms of liver toxicity for the HANs.

One postulated mode of action for the toxicity of haloacetonitriles (HANs) is through
direct interactions with cellular macromolecules (Pereira et al., 1984; Daniel et al., 1986; Lin and
Guion, 1989; Lin et al., 1992). Depletion of reduced glutathione (GSH) could also play a role.

EPA/OW/OST/HECD

VII-1

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
HANs have been shown to induce transient decreases in rat liver GSH levels and inhibit
glutathione-S-transferase activity in vivo (Lin and Guion, 1989; Ahmed et al., 1991). Ahmed et
al. (1989) noted that the relative degree of inhibition of rat liver glutathione-S-transferase (GST)
activity in vitro, reported as TCAN>DBAN>DCAN, is consistent with the relative toxicity of
these compounds reported in the literature, suggesting perturbation of GSH protection as an
important mechanism of toxicity. As further support for the relatedness of GSH depletion and
HAN-induced toxicity, Ahmed et al. (1991) noted that the sustained depletion of cellular GSH
levels in stomach tissues by DBAN was consistent with their earlier preliminary finding that acute
oral doses of HANs can damage the gastric tissues. However, the effect of HANs on gastric
tissue might simply reflect the direct irritancy of these compounds, rather than GSH depletion.
The initial finding of HANs damaging gastric tissues cannot be further investigated, because the
subacute and subchronic studies (Hayes et al., 1986) did not include a histopathological
examination of the stomach.

GSH depletion could enhance cytotoxicity by allowing damage to cellular macromolecules
by HANs, their metabolites, or other reactive species accumulated in the cell. Since GSH is an
important cellular antioxidant, its depletion might induce cellular oxidative stress. In support of
this idea, Ahmed et al. (1999) reported that orally-administered monochloroacetonitrile (MCAN)
induced a dose-dependent decrease in GSH levels and increased levels of oxidative DNA damage
in the stomach mucosa of rats. Alternative mechanisms for MCAN-induced oxidative stress were
discussed by the study authors. One proposed mechanism involves GSH depletion, resulting in
decreased ability of the cell to detoxify endogenous reactive oxygen species. In an alternative
mechanism, cyanide derived from MCAN metabolism might alter cellular oxygen utilization and

EPA/OW/OST/HECD

VTI-2

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
increase the formation of reactive oxygen species. In agreement with the proposed role of
oxidative stress, Mohamadin and Abdel-Naim (1999) reported that MCAN decreased cellular
GSH content and increased lipid peroxidation, as measured by the concentration of thiobarbituric
acid reactive substances (TBARS), in rat gastric epithelial cells in culture. Cell viability, measured
by the release of lactate dehydrogenase activity, was correlated with the depletion of cellular GSH
levels ® =0.96). Supplementing the culture medium with treatments that protect against cellular
oxidative stress (e.g. antioxidants or iron chelators), decreased the cytotoxicity and lipid
peroxidation induced by MCAN. MCAN is not a compound under review for this document.
However, since it shares the ability to deplete GSH and form cyanide with other HANs, these
data are appropriate for discussion here, although in generalizing about the HANs as a class of
compounds, differences in their ability to deplete GSH should be considered in evaluating the
likely mechanisms for any individual compound.

In addition to the induction of oxidative stress secondary to GSH depletion or cyanide
activity on cell respiration, an alternative pathway might include activation of macrophages to
release reactive oxygen species. Ahmed et al. (2000) reported that DCAN induced oxidative
stress responses, including increased oxidation of glutathione, increased formation of reactive
oxygen intermediates, and increased tumor necrosis factor alpha (TNF-a) secretion (a cellular
response during macrophage activation) in a mouse macrophage cell line in culture. To explain
these data, the authors proposed that DCAN treatment activates macrophages, with subsequent
increases in reactive oxygen intermediate production and in TNF-a secretion. The increased
production of reactive oxygen species induces oxidative stress that reduces cell viability through
apoptosis and necrosis.

EPA/OW/OST/HECD

VTI-3

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
The identification of thiocyanate as a urinary metabolite in animals orally-dosed with
HANs, and the hypothesized metabolism to cyanide (Pereira el al., 1984), suggest additional
possible effects of HANs that should be investigated. Longer-term exposure to thiocyanate
(either from thiocyanate dosing or as a metabolite of cyanide) causes thyroid effects. Central
nervous system effects (e.g., myelin degeneration) and male reproductive effects (decreased
epididymal weight and sperm motility) have also been observed following long-term exposure to
cyanide (U.S. EPA, 2002c). Effects on the central nervous system or thyroid were not observed
in a recent 14-day or 13-week NTP (2002) study for DBAN, Although decreased testes weight
and testes atrophy were reported in the 14-day (NTP, 2002) study for DBAN, this effect has not
been observed in subchronic studies (Hayes et al., 1986; NTP, 2002), or in other studies that
evaluated male reproductive tract parameters (R.O.W. Sciences, 1997; Meier et al., 1985).
However, only limited conclusions can drawn regarding the potential for HANs to induce a
similar array of effects as cyanide and thiocyanate, based on limitations in the overall database.

A comparison of relative potency among HANs in thiocyanate excretion, alkylation
potential (p-nitrobenzopyridine binding), protein binding (inhibition of dinitrosamine demethylase
activity), and production of DNA strand breaks, suggested that relative thiocyanate excretion did
not correspond well to some of these markers of potential macromolecule interaction (Pereira el
al., 1984). For example, monochloroacetonitrile (MCAN) was the most potent thiocyanate
former, but TCAN was more potent than MCAN in protein binding and inducing DNA strand
breaks. The authors suggested that this discordance between propensity for cyanide formation
and induction of toxic outcomes might reflect the formation of reactive intermediates other than
cyanide from TCAN, such as phosgene and cyanoformyl chloride. The results of the comparative

EPA/OW/OST/HECD

VTI-4

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
analysis by Pereira et al. (1984) suggest that intermediate metabolites, other than cyanide, may be
important in producing some of the toxic effects observed for HANs. In further support of this
conclusion, the systemic toxicity induced by cyanide and thiocyanate does not closely parallel the
range of effects observed for HANs. For example, cyanide and thiocyanate are not potent liver
toxicants, a target organ for the effects of HANs (U.S. EPA, 2002c).

BCAN, DBAN, DCAN, and TCAN all cause developmental toxicity in vivo (Smith et al.,
1986; Smithed al., 1987; Smithed al., 1988; Smithed al., 1989; Christen al., 1995; Christen al.,
1996), but the relative potency of these compounds is unclear, due to the likely potentiating
effects of the solvent vehicle (tricaprylin) used in these studies (Christ el al., 1996).

As discussed in Chapter III (Toxicokinetics), it remains unclear whether the potentiating
effect of tricaprylin on the developmental toxicity of HANs represents a toxicokinetic or
toxicodynamic interaction, or whether both of these mechanisms play a role. Since the data on
tricaprylin effects for HANs were limited to two published abstracts (Roth et al., 1990; Gordon et
al., 1991), a review of the literature on interactions between solvent vehicles and other
disinfectant by-products or unsaturated nitriles was done to determine whether a consistent
relationship could be found. Solvent vehicles are needed for studies with these compounds, due
to the minimal water solubility. No data were found for interactions between tricaprylin and the
comparison compounds. Therefore, data on observed solvent vehicle interactions are too limited
for trihalomethanes and unsaturated nitriles to be useful in reaching a general conclusion about the
mechanisms involved in the ability of tricaprylin to potentiate the developmental toxicity of
HANs.

EPA/OW/OST/HECD

VTI-5

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
Developmental toxicity has been observed even in the absence of confounding by
tricaprylin, at least for TCAN (Christ el al., 1996). Maternal toxicity was observed at lower
doses than developmental effects in this study, and therefore, it is possible that the observed
developmental effects were secondary to maternal effects. Mechanistic explanations for the
developmental toxicity of HANs have been explored, including cyanide formation and glutathione
depletion. Early evidence suggested that the developmental toxicity of HANs might not be
secondary to cyanide formation. When a maximally tolerated dose of a series of HANs in a
tricaprylin vehicle was administered to Long-Evans rats, developmental toxicity was greatest for
the highly chlorine-substituted acetonitriles (Smith et al., 1986). These results were contrasted
with the work of Pereira et al. (1984), who found that thiocyanate excretion (and hence cyanide
production) is inversely related to chlorine substitution. Based on this comparison, in which
greater chlorine substitution increases developmental toxicity and decreases thiocyanate excretion,
the authors concluded that the degree of developmental toxicity was not due to cyanide
formation. However, conclusions drawn from the HANs studies must be tempered by uncertainty
regarding their true relative developmental toxicity potencies. The tricaprylin vehicle used in the
developmental toxicity studies causes developmental effects by itself (Smith et al., 1989; Christ et
al., 1995), and interacts with TCAN to cause both qualitative changes in the spectrum of
developmental effects, as well as more-than-additive quantitative changes. Because of these
effects of the vehicle used in the relevant developmental toxicity studies, and because the relative
potentiating ability of tricaprylin may differ for each of the HANs tested, the true relative
developmental toxicity potencies are unknown. Another potential avenue for determining the
contribution of cyanide to the developmental toxicity of HANs would be to compare the
developmental effects observed in animals exposed to cyanide versus those observed for HANs.

EPA/OW/OST/HECD

VTI-6

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
However, the available developmental toxicity studies for both HANs (see Chapter V) and
cyanide (U.S. EPA, 2002c) are too limited to conduct a meaningful comparison of their relative
potencies as developmental toxicants.

Christ et al. (1995), in a study on the developmental toxicity of BCAN, reiterated the
discordance between developmental toxicity and degree of metabolism to cyanide, and introduced
GSH depletion as another possible mechanism for developmental toxicity. To address the role of
GSH depletion in the developmental toxicity of HANs, Christ et al. (1995) cited the study of
Abdel-Aziz et al. (1993), who compared maternal and fetal uptake of [2-uC]-MCAN in CD-I
mice. Abdel-Aziz et al. (1993) treated dams with diethylmaleate (DEM) by intraperitoneal
injection to induce an oxidative stress response on gestation day 13. One hour later, control mice
and DEM-pretreated mice were given a 77 mg/kg dose of radiolabeled MCAN by i.v. injection.
MCAN treatment significantly decreased maternal liver, uterus, and fetal tissue levels of GSH,
and the degree of GSH depletion was further increased in the DEM-pretreated mice. Urinary
excretion of thiocyanate, a measure of MCAN metabolism, was five times higher in the DEM-
pretreated mice given MCAN as compared to control mice treated with MCAN only. MCAN
equivalents determined from the yield of radioactivity in maternal uterine tissues, amniotic fluid,
and in fetuses rose rapidly to similar levels at 1 hour for both DEM-pretreated mice and mice not
given DEM. In the absence of DEM pretreatment, MCAN equivalents declined rapidly for all
three tissues examined. However, in DEM-pretreated animals, the removal of tissue radioactivity
was significantly slower. At 24 hours, the level of MCAN equivalents was two-fold higher in fetal
DNA than in maternal uterine DNA. This effect was further enhanced by DEM pretreatment.
The total DNA-bound MCAN equivalent level in DEM-pretreated mice was four-fold higher in

EPA/OW/OST/HECD

VTI-7

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

fetal DNA than in maternal uterine DNA, suggesting that fetuses might be particularly sensitive to
the effects of GSH depletion. Based on the increases in thiocyanate excretion and DNA binding
under oxidative stress conditions that deplete GSH, the authors suggested that GSH depletion
increases the metabolism of the remaining unconjugated MCAN and/or increases the availability
of MCAN to react directly with cellular macromolecules such as DNA. These data suggest that
the developmental toxicity of HANs may be directly related to the ability of these compounds to
deplete GSH levels.

Saillenfait and Sabate (2000) tested the developmental toxicity of aliphatic nitriles (sodium
cyanide, acetonitrile, propionitrile, n-butyronitrile, acrylonitrile, methacrylonitrile, allylnitrile, cis-
2-pentenenitrile, and 2-chloroacrylonitrile) in a rat whole embryo assay and in vivo. Although no
HANs were included in this study, the results may have implications for the developmental
toxicity of HANs based on similar chemical reactivity. In the whole embryo testing experiment, a
wide range of embryotoxicity was observed. In addition, no common pattern of developmental
effects was observed among all the compounds tested. Enhancement of metabolism by
supplementation of the cultures with microsomes increased the severity of embryolethality and
developmental toxicity for the unsaturated aliphatic nitriles (e.g., acrylonitrile), but not the
saturated aliphatic nitriles (e.g., acetonitrile). This result, coupled with the difference between the
spectrum of dysmorphogenesis observed for the unsaturated nitriles and for embryos treated with
sodium cyanide, led the authors to suggest that microsomal metabolism of unsaturated nitriles
generates toxic metabolites in addition to cyanide. In further support of a mechanism distinct
from cyanide release for the in vitro developmental toxicity of the tested compounds, their relative

EPA/OW/OST/HECD

VTI-8

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
potency in the whole-embryo culture assay did not directly correspond with increasing cyanide
release kinetics as determined from in vitro metabolism studies.

No haloacetontriles were included in the study by Saillenfait and Sabate (2000), but based
on their shared ability to deplete cellular GSH (as described in the rest of this paragraph), HANs
may act chemically more like unsaturated nitriles than saturated ones, and thus may induce
developmental toxicity through mechanisms other than cyanide release. Ahmed et al. (1982)
reported that unsaturated but not saturated aliphatic nitriles decrease liver glutathione levels.
Clinical signs of toxicity were different for unsaturated nitriles than for potassium cyanide, while
animals administered saturated nitriles and potassium cyanide showed a similar spectrum of
symptoms. Taken together, the results of Saillenfait and Sabate (2000) and Ahmed et al. (1982)
suggest that reactive metabolites in addition to cyanide might be important for the induction of
developmental toxicity. In further support for a common toxic mechanism for unsaturated nitriles
and HANs, Smith et al. (1989) noted that a similar spectrum of soft-tissue malformations is
observed for TCAN, DCAN, and the unsaturated nitrile, acrylonitrile. However, later evidence
presented by Christ et al. (1996), demonstrated that the use of tricaprylin in these earlier studies
might have caused a shift in the type of soft-tissue malformations, placing the earlier conclusion in
doubt.

Some results, however, do suggest an involvement of cyanide in the developmental
toxicity of HANs. In contrast to their whole-embryo assay results, Saillenfait and Sabate (2000)
reported similarities in the spectrum of developmental effects induced by all of the nitriles when
tested in vivo. In this experiment, CD-I mice dams received a single dose of aliphatic nitrile on

EPA/OW/OST/HECD

VTI-9

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
gestation day 10, and evaluation of embryos for defects was conducted on gestation day 12. The
dysmorphogenic effects characteristic of sodium cyanide, including misdirected allantois (a tubular
structure of the embryonic hindgut), trunk, or caudal extremity were induced by saturated as well
as unsaturated aliphatic nitriles. The concordance in these results in vivo is consistent with
cyanide being an active moiety for both saturated and unsaturated aliphatic nitriles.

Moudgal et al. (2000) evaluated relationships between the structure of 244 disinfectant
byproducts, including 21 nitriles, and potential developmental toxicity using a rat oral
developmental toxicity submodel of TOPKAT®, a quantitative structure toxicity relationship
(QSTR) prediction tool. Based on individual structural descriptors, model probabilities were used
to derive qualitative estimates as follows: 0.0 to 0.3 negative, 0.3 to 0.7 indeterminate, 0.7 to 1.0
positive. The probability estimate is independent of the potency or severity of developmental
effects that could be induced, and would be interpreted as the likelihood that the chemical can
cause developmental toxicity in rats following oral dosing. As a group, the nitrile disinfectant
byproducts were characterized as having a high probability of developmental toxicity. Of the 21
individual nitrile compounds, 13 were positive, 5 were negative, and for 3 the model did not make
a prediction. Only three of the four specific HANs under consideration in this document were
assessed. DBAN and TCAN were predicted as positive and BCAN was predicted as negative.
TCAN was one of the chemicals in the model training set. Four other HANs were predicted as
positive. A cursory evaluation of these data do not reveal a clear reason for the negative
prediction for BCAN, since positive results were obtained for bromoacetonitrile,
bromodichloroacetonitrile, and dibromochloroacetonitrile. The effects of individual structural
moieties were also examined by the software for various structural classes of compounds. The

EPA/OW/OST/HECD

VII-10

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
nitrile moiety, chlorine atom, and bromine atom were identified as contributing most significantly
to the developmental toxicity predictions for the group of 20 nitriles. The importance of the
nitrile group in the developmental toxicity predictions is consistent with cyanide as a causal factor
in the developmental toxicity of these compounds in vivo, since this functional group gives rise to
cyanide during HAN metabolism.

In summary, mechanisms of toxicity for noncarcinogenic effects (decreased body weight,
gastric and liver toxicity, and developmental effects) have been hypothesized to be related to
direct interaction with cellular proteins (disrupting critical enzyme functions), depletion of cellular
antioxidant defenses (i.e, depletion of GSH and inhibition of GST), or effects secondary to
cyanide formation. The data are not adequate to rule out any of these possibilities, and therefore,
any or all of these mechanisms could be involved.

B. Mechanism of Carcinogenesis

The carcinogenic potential of the HANs is unknown. No epidemiological studies have
evaluated directly the carcinogenic potential of HANs in humans. Rather, studies have evaluated
the carcinogenic potential of chlorinated versus unchlorinated drinking water or the presence of
trihalomethanes as a marker of chlorination by-products (IARC, 1999; Mills et al., 1998). Many
of these studies have shown an association between chronic exposure to chlorinated water and
increased risks of bladder, rectal, or colon cancers (Mills et al., 1998; WHO, 2000). No standard
cancer bioassays of HANs have been done in animals. Limited short-term exposure data from the
mouse skin assay (Bull et al., 1985) and the mouse lung assay (Bull and Robinson, 1985) indicate
that BCAN, DBAN, and TCAN may be tumorigenic, although DBAN, DCAN, and TCAN were

EPA/OW/OST/HECD

VII-11

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
reported to be negative in the rat liver GGT-foci assay (Herren-Freund and Pereira, 1986). In a
qualitative review of structure activity relationships, Bull and Robinson (1985) discussed potential
structural relationships among the results for the cancer screening assays and genotoxicity results
for various HANs. They commented that there is no apparent consistent pattern in the potency of
the individual compounds across the various assays.

Quantitative analysis of structural relationships of HANs with carcinogenic outcomes have
also been investigated. Moudgal et al. (2000) evaluated relationships between the structure of
244 disinfectant byproducts, including 21 HANs, and potential carcinogenicity using mouse and
rat oral submodels of TOPKAT®, a quantitative structure toxicity relationship (QSTR) prediction
tool. As a group, the nitrile disinfectant byproducts were characterized as having a low
probability of carcinogenicity. However, for the subset of six HANs in the total group of 20
nitrile compounds for which individual data were presented (carcinogenicity predictions for
TCAN were not included), four were predicted as positive in at least one sex in mice or rats. The
results were largely mixed across species or sex for each test compound. For example, BCAN
was predicted as positive in male and female mice, but negative in both sexes of rats. DBAN was
negative in both sexes of mice, and female rats, but indeterminate in male rats. These QSTR
results are consistent with the screening bioassays in indicating at least limited carcinogenic
potential of the HANs.

The HANs or their metabolites are reactive compounds that can bind macromolecules
including DNA (Daniel et al., 1986; Lin et al., 1992). Nouraldeen and Ahmed (1996)
demonstrated in vitro that the degree of DNA binding was much lower for DC AN and TCAN

EPA/OW/OST/HECD

VII-12

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
than it was for bromoacetonitrile or MCAN, Based on changes in fluorescence as a measure of
adduct formation, relative DNA adduct formation as compared to bromoacetonitrile was 8.6%,
1.0%, and 0.2% for MCAN, DCAN, and TCAN respectively. The chemical nature of the
adducts that were formed was not identified for each compound. However, a 7-
(cyanomethyl)guanine adduct was the single major adduct identified from the reaction of
bromoacetonitrile with calf-thymus DNA. Since the more fully halogenated acetonitriles BCAN,
DBAN, DCAN, and TCAN, may be metabolized to monohaloacetonitriles in vivo (Pereira el al.,
1984), they might induce the formation of similar adducts. On the other hand, the relative DNA
reactivity observed in vitro may not directly translate to mutagenic or carcinogenic potential in
vivo, since the metabolism of the compound and the mutagenicity of the adducts formed may
differ for each HAN.

Glutathione conjugation may be an important cellular protection against these reactive
HANs or their metabolites (Lin and Guion, 1989; Ahmed et al., 1991). The genotoxicity of each
of the HAN compounds BCAN, DBAN, DCAN, and TCAN has been evaluated in at least one
assay, and in most cases a variety of different assays. Although some of the data have provided
contradictory results, all of the tested compounds appear to have some capacity to induce
genotoxic effects. For example, BCAN, DCAN, and TCAN (but not DBAN) have been found to
generate a positive result in at least one reported Salmonella!microsome assay. In addition, while
not uniformly consistent, a variety of other assays, including those for chromosome effects, have
yielded positive results for some of the HANs. Overall, these data suggest that HANs induce
genotoxicity through direct interactions with DNA.

EPA/OW/OST/HECD

VII-13

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
C. Interactions and Susceptibilities

Potential Interactions

No studies on interactions of HANs with other classes of compounds were identified,
except as noted above for solvent vehicle effects.

Childhood Susceptibility

As discussed above, developmental toxicity has been associated with exposure to BCAN,
DBAN, DC AN, and TCAN, although the findings for all these compounds, except TCAN, are
confounded by the developmental toxicity of the vehicle. The developmental toxicity of the
HANs is supported by the reports of Roth et al. (1990) and Gordon el al. (1991) in published
abstracts that radioactivity was detected in embryos of dams given [14C]TCAN, suggesting that
fetuses may be targets for HAN toxicity. In addition, Abdel-Aziz et al. (1993) found that fetuses
may be more susceptible than adults to direct DNA damage induced by haloacetontriles.
Although the relationship between metabolism of HANs and the onset of toxicity has not been
thoroughly determined (as discussed above), acute cyanide intoxication from accidental pediatric
acetonitrile exposures (Caravati et al., 1988; Geller et al., 1991; Kurt et al., 1991) indicates that
HAN metabolism to cyanide occurs in children.

There are limited data available for assessing directly the susceptibility of fetuses and
children to HANs. No systemic toxicity studies have been identified that evaluated age-related

EPA/OW/OST/HECD

VII-14

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
differences in sensitivity and no multigeneration reproductive studies have been reported. In
addition, age-related differences in the metabolism of the HANs could not be investigated,
because the enzyme(s) responsible for HAN metabolism are not known. Pereira et al. (1984)
hypothesized that mixed function oxidases are involved, but the isozyme involved has not been
identified, and the age-dependent expression differs among the different cytochrome P450
isozymes. Although the developmental toxicity of the HANs has been evaluated, most of the
available literature cannot be used to determine relative maternal and developmental toxicity of
these compounds, due to potential interactions between the test compound and the dosing vehicle
(tricaprylin). Excluding these studies from this evaluation leaves only limited data. No
developmental toxicity was observed following administration of up to 10.9 mg/kg/day DBAN in
drinking water (R.O.W. Sciences, 1997), but this was only a screening study that did not include
evaluation of pups for malformations. The only maternal effect at this dose was a decrease in
drinking water consumption. In a study using corn oil as the solvent vehicle for TCAN, the
maternal NOAEL was 15 mg/kg/day and the NOAEL for developmental toxicity was 35
mg/kg/day (Christ et al., 1996). Thus, these two studies do not provide evidence that fetuses are
more susceptible than adults, although the data are too limited to make a definitive conclusion.

Other potential susceptibilities

Although a role of oxidative metabolism and glutathione conjugation have been
hypothesized to be involved in HAN metabolism, the identity of enzymes involved in these
pathways has not been determined. Therefore, the potential role of interindividual differences in
metabolism due to genetic polymorphism, age, or other factors cannot be determined.

EPA/OW/OST/HECD

VII-15

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

D. Summary

The HANs induce general systemic toxicity. Decreased body weight and a variety of
organ weight changes occur following oral dosing, and the testes (NTP, 2002) and liver might be
particularly sensitive (Hayes et al., 1986), although the reported effects in these organs in
available studies are fairly limited. The HANs also induce developmental effects (Smith et al.,
1986; Smithed al., 1987; Smithed al., 1988; Smithed al., 1989; Christen al., 1995; Christen al.,
1996). The mechanism(s) of toxicity are not known, but several possibilities have been described.
HANs may act through direct interactions with cellular macromolecules such as DNA (Daniel el
al., 1986; Lin et al., 1992; Nouraldeen and Ahmed, 1996). HAN toxicity might be secondary to
GSH depletion (Ahmed et al., 1991) or oxidative stress (Ahmed et al., 1999; Mohamadin and
Abdel-Naim, 1999). Formation of cyanide from HAN might be another important mechanism of
toxicity, although important systemic effects that are sensitive indicators of cyanide toxicity have
not been fully examined.

The role of cyanide in the developmental toxicity of HANs has received much attention.
Some studies suggest that metabolites other than cyanide play a critical role (Smith et al., 1986),
and implicated glutathione depletion as an important factor (Christ et al., 1995; Abdel-Aziz et al.,
1993). Although some indirect data supports a role of cyanide (Moudgal et al., 2000; Saillenfait
and Sabate, 2000), evaluation of the available developmental toxicity studies of cyanide itself do
not support this hypothesis (U.S. EPA, 2002c).

The ability of the HANs to bind to cellular macromolecules (Daniel et al., 1986; Lin et al.,
1992; Nouraldeen and Ahmed, 1996), as well as generally positive results in genotoxicity assays,

EPA/OW/OST/HECD

VII-16

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

supports direct DNA damage as the mode of action for the tumorigenicity observed in cancer
screening studies (Bull et al., 1985; Bull and Robinson, 1985). However, the carcinogenic
potential of the HANs is unknown, since epidemiology studies are not available and standard
cancer animal bioassays of HANs have not been conducted.

Identification of potential susceptible subpopulations is hampered by the incomplete
characterization of HAN metabolism or identification of the toxic moiety. Although a metabolic
pathway for the HANs has been proposed (Pereira et al., 1984), the enzymes important for
catalyzing HAN metabolism are unknown. In addition, no studies on age-dependent differences
in metabolism or toxicity were identified, although one study demonstrated that HANs may bind
more greatly to fetal DNA than to DNA in maternal tissues (Abdel-Aziz et al., 1993). Analysis of
the developmental toxicity studies for TCAN revealed a lower maternal than developmental
NOAEL, which does not suggest that fetuses are more susceptible than adults.

EPA/OW/OST/HECD

VII-17

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
Chapter VIII.	Quantification of Toxicological Effects

The quantification of toxicological effects of a chemical consists of separate assessments
of noncarcinogenic and carcinogenic health effects. Unless otherwise specified, chemicals which
do not produce carcinogenic effects are believed to have a threshold dose below which no
adverse, noncarcinogenic health effects occur, while carcinogens are assumed to act without a
threshold.

A. Introduction to Methods

A.l Quantification of Noncarcinogenic Effects

In quantification of noncarcinogenic effects, a Reference Dose (RfD) (formerly called the
Acceptable Daily Intake (ADI)) is calculated. The RfD is "an estimate (with uncertainty spanning
approximately an order-of-magnitude) of a daily exposure to the human population (including
sensitive subgroups) that is likely to be without appreciable risk of deleterious effects over a
lifetime" (U.S. EPA, 1993). The RfD is derived from a no observed adverse effect level
(NOAEL), lowest observed adverse effect level (LOAEL), or a NOAEL surrogate such as a
benchmark dose identified from a subchronic or chronic study, and divided by a composite
uncertainty factor(s). The RfD is calculated as follows:

RfD = NOAEL (LOAEL)

UF x MF

where:

NOAEL = No-observed-adverse-effect level from a high-quality toxicological

study of an appropriate duration

EPA/OW/OST/HECD

VIII-1

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

LOAEL = Lowest-observed-adverse-effect level from a high-quality

toxicological study of an appropriate duration. In situations where
there is no NOAEL for a contaminant but there is a LOAEL, the
LOAEL can be used for the RfD calculation with the inclusion of
an additional uncertainty factor.

UF = Uncertainty factor chosen according to EPA/NAS guidelines

MF = Modifying factor

Selection of the uncertainty factor to be employed in calculation of the RfD is based on
professional judgment, while considering the entire database of toxicological effects for the
chemical. To ensure that uncertainty factors are selected and applied in a consistent manner, the
Office of Water (OW) employs a modification to the guidelines proposed by the National
Academy of Sciences (NAS, 1977, 1980). According to the EPA approach (U.S. EPA, 1993),
uncertainty is broken down into its components, and each dimension of uncertainty is given a
quantitative rating. The total uncertainty factor is the product of the component uncertainties.
The individual components of the uncertainty are as follows:

UFh A factor of 1, 3, or 10-fold used when extrapolating from valid data in

studies using long-term exposure to average healthy humans. This factor is
intended to account for the variation in sensitivity (intraspecies variation)
among the members of the human population.

UFa An additional factor of 1, 3, or 10 used when extrapolating from valid

results of long-term studies on experimental animals when results of studies
of human exposure are not available or are inadequate. This factor is
intended to account for the uncertainty involved in extrapolating from
animal data to humans (interspecies variation).

UFS An additional factor of 1, 3, or 10 used when extrapolating from less-than-

chronic results on experimental animals when there are no useful long-term
human data. This factor is intended to account for the uncertainty involved
in extrapolating from less-than-chronic NOAELs to chronic NOAELs.

EPA/OW/OST/HECD

VTII-2

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

UFl An additional factor of 1, 3, or 10 used when deriving an RfD from a

LOAEL, instead of a NOAEL. This factor is intended to account for the
uncertainty involved in extrapolating from LOAELs to NOAELs.

UFd An additional factor of 1, 3- or 10 used when deriving an RfD from an

"incomplete" database. This factor is meant to account for the inability of
any single type of study to consider all toxic endpoints. The intermediate
factor of 3 (approximately V2 log10 unit, i.e., the square root of 10) is often
used when there is a single data gap exclusive of chronic data. It is often
designated as UFD.

On occasion, EPA also uses a modifying factor in the determination of the RfD. A
modifying factor is an additional uncertainty factor that is greater than zero and less than or equal
to 10. The magnitude of the MF depends upon the professional assessment of scientific
uncertainties of the study and database not explicitly treated above (e.g., the number of species
tested). The default value for the MF is 1.

In establishing the UF or MF, it is recognized that professional scientific judgment must be
used. The total product of the uncertainty factors and modifying factor should not exceed 3000.
If the assignment of uncertainty results in a UF/MF product that exceeds 3000, then the database
does not support development of an RfD. The quantification of toxicological effects of a
chemical consists of separate assessments of noncarcinogenic and carcinogenic health effects.
Unless otherwise specified, chemicals which do not produce carcinogenic effects are believed to
have a threshold dose below which no adverse, noncarcinogenic health effects occur, while
carcinogens are assumed to act without a threshold.

EPA/OW/OST/HECD

VTII-3

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
A. 1.1. Drinking Water Equivalent Level

The drinking water equivalent (DWEL) is calculated from the RfD. The DWEL
represents a drinking-water-specific lifetime exposure at which adverse, noncarcinogenic health
effects are not anticipated to occur. The DWEL assumes 100% exposure from drinking water.
The DWEL provides the noncarcinogenic health-effects basis for establishing a drinking-water
standard. For ingestion data, the DWEL is derived as follows:

DWEL	= (RfD) x BW

WI

where:

BW	= 70-kg adult body weight

WI	= Drinking water intake (2 L/day)

A.1.2. Health Advisory Values

In addition to the RfD and the DWEL, EPA calculates Health Advisory (HA) values for
noncancer effects. HAs are determined for lifetime exposures as well as for exposures of shorter
duration (1-day, 10-day, and longer-term). The shorter duration HA values are used as informal
guidance to municipalities and other organizations when emergency spills or contamination
situations occur. The lifetime HA becomes the MCLG for a chemical that is not a carcinogen.

The shorter-term HAs are calculated using an equation similar to the ones for RfD and
DWEL; however, the NOAELs or LOAELs are derived from acute or subchronic studies and
identify a sensitive noncarcinogenic endpoint of toxicity. The HAs are derived as follows:

EPA/OW/OST/HECD

VTII-4

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

HA = NOAEL or LOAEL x BW
UF x WI

where:

NOAEL or LOAEL =

BW

UF

WI

No- or lowest-observed-adverse-effect-level in mg/kg
bw/day

Assumed body weight of a child (10 kg) or an adult (70 kg)

Uncertainty factor, in accordance with EPA or NAS/OW
guidelines

Assumed daily water consumption of a child (1 L/day) or an
adult (2 L/day)

Using the above equation, the following drinking water HAs are developed for noncarcinogenic
effects:

1-day HA for a 10-kg child ingesting 1 L water per day.
10-day HA for a 10-kg child ingesting 1 L water per day.
Longer-term HA for a 10-kg child ingesting 1 L water per day.
Longer-term HA for a 70-kg adult ingesting 2 L water per day.

Each of these shorter-term HA values assumes that the total exposure to the contaminant comes
from drinking water.

The lifetime HA is calculated from the DWEL and takes into account exposure from
sources other than drinking water. It is calculated using the following equation:

EPA/OW/OST/HECD

VTII-5

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
Lifetime HA = DWEL x RSC

where:

DWEL	= Drinking water equivalent level

RSC	= Relative source contribution. The fraction of the total exposure

allocated to drinking water following EPA guidance (U.S. EPA,
2000b).

A.2 Quantification of Carcinogenic Effects

Under the 1986 guidelines, the EPA categorizes the carcinogenic potential of a chemical
based on the overall weight-of-evidence according to the following scheme:

Group A: Human Carcinogen. Sufficient evidence exists from epidemiology studies to
support a causal association between exposure to the chemical and human cancer.

Group B: Probable Human Carcinogen. Sufficient evidence of carcinogenicity in animals
with limited (Group Bl) or inadequate (Group B2) evidence in humans.

Group C: Possible Human Carcinogen. Limited evidence of carcinogenicity in animals in
the absence of human data.

Group D: Not classified as to Human Carcinogenicity. Inadequate human and animal
evidence of carcinogenicity or for which no data are available.

Group E: Evidence of Noncarcinogenicity for Humans. No evidence of carcinogenicity in
at least two adequate animal tests in different species or in both adequate epidemiologic
and animal studies.

EPA/OW/OST/HECD

VTII-6

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

If toxicological evidence leads to the classification of the contaminant as a genotoxic,
probable or possible human carcinogen, mathematical models are used to calculate the estimated
excess cancer risk associated with ingestion of the contaminant in drinking water. The data used
in these estimates usually come from lifetime-exposure studies in animals. In order to predict the
risk for humans from animal data, animal doses must be converted to equivalent human doses.

This conversion includes correction for noncontinuous exposure, less-than-lifetime studies and
differences in size. It is assumed that the average adult human-body weight is 70 kg and that the
average water consumption of an adult human is two liters of water per day.

For contaminants with a carcinogenic potential, chemical levels are correlated with a
carcinogenic-risk estimate by employing a cancer potency (unit risk) value together with the
assumption for lifetime exposure via ingestion of water. Under the 1986 Carcinogen Risk
Assessment Guidelines, the cancer unit risk is usually derived from a linearized multistage model
with a 95% upper confidence limit providing a low-dose estimate; that is, the true risk to humans,
while not identifiable, is not likely to exceed the upper-limit estimate and, in fact, may be lower.
Excess cancer-risk estimates may also be calculated using other models such as the one-hit,
Weibull, logit and probit models. There is little basis in the current understanding of the biological
mechanisms involved in cancer to suggest that any one of these models is able to predict risk more
accurately than any of the others. Because each model is based upon differing assumptions, the
estimates that are derived for each model can differ by several orders of magnitude.

The scientific data base used to calculate and support the setting of cancer-risk rates has
an inherent uncertainty due to the systematic and random errors in scientific measurement. In

EPA/OW/OST/HECD

VTII-7

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
most cases, only studies using experimental animals have been performed. Thus, there is

uncertainty when the data are extrapolated to humans. When developing cancer-risk rates, several

other areas of uncertainty exist, such as the incomplete knowledge concerning the health effects of

contaminants in drinking water, the impact of the experimental animal's age, sex and species, the

nature of the target organ system(s) examined and the actual rate of exposure of the internal

targets in experimental animals or humans. Dose-response data usually are available only for high

levels of exposure, not for the lower levels of exposure at which a standard may be set. When

there is exposure to more than one contaminant, additional uncertainty results from a lack of

information about possible synergistic or antagonistic effects.

The quantification of toxicological effects of a chemical consists of separate assessments
of noncarcinogenic and carcinogenic health effects. Chemicals that do not produce carcinogenic
effects are believed to have a threshold dose below which no adverse, noncarcinogenic health
effects occur, while carcinogens are assumed to act without a threshold.

B. Noncarcinogenic Effects

Analysis of the dose-response data for noncarcinogenic effects for each of the four HANs
and the derivation of Health Advisories is described below and summarized in Tables VIII-1
through VIII-7.

B.l BCAN

The oral toxicity data for BCAN are summarized in Table VIII-1. No systemic toxicity
studies of BCAN are available. The most comprehensive studies of BCAN toxicity were two

EPA/OW/OST/HECD

VTII-8

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

developmental studies of BCAN. Smith et al. (1987) reported that the single dose tested of 55
mg/kg/day of BCAN, administered by gavage to pregnant rats on days 7 to 21 of gestation,
resulted in decreased maternal weight gain and reduced pup birth weights. In a dose-response
study (Christ et al., 1995), sperm-positive female rats were administered BCAN by gavage in
tricaprylin on gestation days 6 to 18 at doses of 0, 5, 25, 45, and 65 mg/kg/day. Treatment with
BCAN in tricaprylin resulted in both maternal and embryotoxicity. The LOAEL for
developmental effects was 5 mg/kg/day compared to tricaprylin treated controls. No maternal
effects were observed at this dose. However, use of this study for dose-response assessment is
not appropriate, because it may not accurately reflect the toxicity of BCAN in drinking water.
Tricaprylin vehicle alone produced embryotoxicity in this study, and later work by this laboratory
(Christ et al., 1996) suggests that tricaprylin may act synergistically with TCAN to enhance
developmental toxicity. In the absence of a more complete data base, the data are inadequate for
derivation of any Health Advisory values for BCAN.

EPA/OW/OST/HECD

VTII-9

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Table VIII-1 Summary of Oral Studies of BCAN Toxicity

Reference

Species/
Strain

Route/
Dose

Exposure
Duration

Endpoints
Evaluated

NOAEL
(mg/kg/day)

LOAEL (mg/kg/day)

Meier et
al. (1985)

Mouse-
B6C3F1

Gavage in
water

0, 12.5,25,
or 50

mg/kg/day

5 Days

Sperm head
abnormalities

50 (Free-
standing
NOAEL)

NDa

Smith et
al. (1987)

Rat-

Long-
Evans
Hooded

Gavage in
tricaprylinb

55

mg/kg/day

Days 7 to 21
of gestation

Maternal weight,
reproductive
success, pup
viability and
growth

Maternal: ND

Developmental:
ND

Maternal: ND
(Nonsignificant
decrease maternal
weight gain)

Development: 55
(Decreased birth
weight, decreased
postnatal weight gain)

Christ et
al. (1995)

Rat-

Long-
Evans

Gavage in
tricaprylinb

0, 5,25,

45,65

mg/kg/day

Days 6 to 18
of gestation

Maternal body and
organ weight,
reproductive
success, pup
viability and
growth,
malformations

Maternal: 45

Developmental:
ND

Maternal: 65 (FEL for
maternal death;
decrease maternal
weight gain)

Development: 5
(Decreased crown-
rump length,
increased
cardiovascular
malformations)

a.	ND = not determined.

b.	Due to the demonstrated ability of tricaprylin to enhance developmental toxicity of TCAN, this study was not considered
in derivation of the Health Advisories.

B.2 DBAN

B.2.1 One-Day Health Advisory for DBAN

The oral toxicity data for DBAN are summarized in Table VIII-2. Two studies (Hayes el
al., 1986; Eastman Kodak Co., 1992) identify acute oral LD50 values for DBAN of 50 to 361
mg/kg. Clinical signs observed include convulsions, ataxia, depressed respiration and activity, and
coma. However, LD50 studies are not suitable for the development of one-day health advisories

EPA/OW/OST/HECD

VIII-10

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
and no other suitable acute studies were located for DBAN, In the absence of such data, the

Ten-day HA value is recommended as a conservative estimate of an appropriate One-day HA

value.

B.2.2 Ten-Day Health Advisory for DBAN

Two reports of adequate general toxicity studies (NTP, 2002, which tested both mice and
rats, and Hayes et al., 1986, which tested rats only) of a suitable duration for the Ten-day HA
were located. NTP (2002) conducted a 14-day drinking water toxicity study in B6C3F1 mice and
F344 rats. Concentration-related decreases in water consumption were noted in males and
females of both species, but this effect was not considered to be toxicologically significant. The
only toxicologically-significant effects in either species were observed in male rats at the high dose
of 18 mg/kg/day, and included decreased body weight, decreased testes weight and testes
atrophy. The NOAEL for this study was 12 mg/kg/day with a LOAEL of 18 mg/kg/day. BMD
modeling was not performed for this study, since the full NTP study reports were not available at
the time of preparation of this document. However, preliminary modeling based on the body
weight gain data provided in the study summaries suggests that the BMDL would not differ
significantly from the study NOAEL.

Hayes et al. (1986) conducted a 14-day study of DBAN toxicity in rats. No significant
effects on serum chemistry, hematological or urinary parameters or remarkable findings at
necropsy were observed. The only organ weight change that showed a clear dose dependence
was relative liver weight in females, which was increased 12% over controls (p < 0.05) at 23
mg/kg/day, and was increased by as much as 22% above controls at 90 mg/kg/day. The increase

EPA/OW/OST/HECD

VIII-11

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
in relative liver weight was statistically significant only at the high dose. In the absence of

histopathology data or clinical chemistry findings, however, it is unclear if this is an adverse

response. In males, body weight was decreased at the 45 mg/kg/day dose level, but not at the 23

mg/kg/day dose level. Therefore, decreased body weight in males is considered the critical effect

and 23 mg/kg/day is considered the NOAEL. BMD modeling was conducted to identify

alternative critical effect levels for this study. A BMDL of 16 mg/kg/day for decreased body

weight in males was selected as the most appropriate modeling result for this endpoint (see

Appendix A).

As part of a reproductive toxicity study, R.O.W. Sciences (1997) conducted range-finding
studies of DBAN that evaluated its short-term effects. Among rats exposed to drinking water
concentrations of DBAN up to 200 ppm for 2 weeks, the only consistent effect was a decrease in
water consumption at the high concentration. The absence of clinical signs of toxicity or body
weight changes indicates that the highest concentration tested of 200 ppm in the second range-
finding study (equivalent to doses of 13.2 mg/kg/day in males; 17.9 mg/kg/day in females) is a
study NOAEL. These same rats had previously been exposed for 4 days to higher concentrations
that caused significant decreases in body weight, and decreased food and water consumption; the
rats were allowed to recover to control body weights before being exposed to the lower
concentrations in the second range-finding study. In the main reproductive and developmental
study the exposure duration was 30 days for males and 35 days for females - longer than is
suitable for a Ten-day HA. In addition, there were no adverse effects observed at the highest
dose tested, 8.2 mg/kg/day in males and 10.8 mg/kg/day in females. Smith et al. (1987) reported
on the developmental toxicity of DBAN. However, use of this study for dose-response

EPA/OW/OST/HECD

VIII-12

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
assessment is not appropriate, because it may not accurately reflect the toxicity of DBAN in

drinking water, due to interactions between tricaprylin vehicle and HANs.

Decreased body weight, decreased testes weight, and testes atrophy in male F344 rats
reported in NTP (2002) at doses greater thanl2 mg/kg/day is considered the most appropriate
basis for deriving the Ten-day HA for DBAN. Decreased body weight was clearly an appropriate
endpoint to serve as the critical effect for deriving the Ten-day HA. Although the reported effects
of DBAN on the testes were considered to be adverse, their appropriateness to serve as the basis
for the HA was not clear. The Ten-day HA is based on water consumption by children directly.
Therefore, male reproductive effects are not an appropriate endpoint for this HA value unless the
observed effects are likely to persist to a reproductive age. The data were not adequate to make
this determination, since none of the shorter-term studies tracked the recovery of this endpoint
after cessation of exposure. It is noteworthy that no effects on the testes were observed in the 13-
week NTP (2002) study in the same strain of rats, suggesting that the testes effects might be
transient. However, the highest dose in the subchronic study NTP (2002) study was 11.3
mg/kg/day, which is essentially the same as the NOAEL for testes effects in the 14-day NTP
study. Therefore, comparison across the 14-day and 13-week NTP studies cannot answer
whether the testes effects are likely to be persistent. Hayes et al. (1986) also evaluated testes
weight in CD rats after 14-day and subchronic gavage dosing with DBAN. This study did not
identify decreased testes weight at either time point. This could be due to rat strain differences in
sensitivity to male reproductive tract toxicity, or reflect differences in the route of DBAN
administration. DBAN administered in drinking water caused a decrease in water consumption in
the NTP (2002) study. If the effects on the testes were secondary to decreased water

EPA/OW/OST/HECD

VIII-13

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
consumption, then they would not have been observed in the Hayes et al. (1986) study, which

used gavage dosing. DBAN did not affect male reproductive parameters in Sprague-Dawley rats

in a reproductive and developmental screening study (R.O.W. Sciences, 1997) in which males

were exposed for 30 days to DBAN in drinking water. However, in this study, the highest dose

tested was 8.2 mg/kg/day (the drinking water concentration was 150 mg/L), which was below the

NOAEL of 12 mg/kg/day for F344 rats in the 14-day NTP study. None of the available studies

allow for a determination of the degree to which the testes effects are likely to persist. Therefore,

since the potential of the observed testicular effects to persist to a reproductive age cannot be

excluded, they are considered to be appropriate co-critical effects for derivation of the Ten-day

HA.

Based on the decreased body weight, decreased testes weight, and testes atrophy in male
rats reported in NTP (2002) at doses greater than 12 mg/kg/day, the Ten-day HA for DBAN may
be calculated as shown below and summarized in Table VIII-3. An uncertainty factor of 10 is
used to account for extrapolation from a NOAEL in an animal study and an uncertainty factor of
10 is used to account for inter-individual variability in human sensitivity. The composite
uncertainty factor used is 100.

„ , (12 mg/kg/day) (10 kg) „ . , .

Ten-day HA = ^ (i £/day) = m§ (rounded t0 1 mS/L)

where:

12 mg/kg/day = NOAEL, based on decreased body weight, decreased testes weight, and

testes atrophy in male rats exposed to a LOAEL of 18 mg/kg/day DBAN in drinking

water for 14 days (NTP, 2002).

10kg= assumed body weight of a child.

EPA/OW/OST/HECD

VIII-14

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

100 = composite uncertainty factor, chosen to account for extrapolation from a NOAEL

in animals, and inter-individual variability in humans.

1 L/day = assumed daily water consumption by a 10-kg child.

B.2.3 Longer-Term Health Advisory for DBAN

Only two reports of suitable studies for deriving a longer-term HA were located. NTP
(2002) evaluated the subchronic toxicity of DBAN in B6C3F1 mice and F344 rats exposed to
DBAN in their drinking water. In both species the only effects were decreased water
consumption and slight decreases in body weight. The highest dose tested was the study NOAEL
of 17.9 mg/kg/day for both male and female mice, 12.6 mg/kg/day for female rats, and 11.3
mg/kg/day for male rats. No NOAEL was identified.

Hayes et al. (1986) conducted a 90-day gavage study of DBAN toxicity in CD rats. At
the high dose of 45 mg/kg/day, males, but not females, had decreased body weight. The next
lower dose of 23 mg/kg/day was the NOAEL for decreased body weight. The only other
noteworthy effects observed in the study were significantly increased ALP in females at 45
mg/kg/day and a significant increase in relative (but not absolute) liver weight in males at 45
mg/kg/day. With the exception of elevated ALP levels at the high dose, there were no significant
treatment-related effects on serum chemistry, hematological, or urinary parameters or remarkable
findings at necropsy at any dose level. The observed liver weight changes were not judged as
adverse since no clinical chemistry signs of liver toxicity were observed in males. Females had an
increase in ALP at the high dose, but did not have a corresponding increase in liver weight. No
histopathology examination was performed to clarify if the liver weight changes in males was

EPA/OW/OST/HECD

VIII-15

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
adverse or adaptive. Based on these considerations, decreased body weight in males was selected

as the critical effect for this study. The NOAEL was 23 mg/kg/day and the LOAEL was 45

mg/kg/day. BMD modeling was conducted for decreased body weight in males to identify

alternative critical effect levels for this study. A BMDL of 20 mg/kg/day was selected as the most

appropriate modeling result for this endpoint (see Appendix A).

Both the NTP (2002) subchronic drinking water study in male rats and the subchronic
gavage study by Hayes et al. (1986) were considered in the selection of the critical study for
derivation of the Longer-term HA. The NOAEL of 11.3 mg/kg/day for male rats in the 13-week
NTP study was selected as the most appropriate basis for derivation of the Longer-term HA. This
value was judged to be more appropriate for deriving the HA than the NOAEL of 23 mg/kg/day
for decreased body weight observed in male rats reported in Hayes et al. (1986) for several
reasons. First, in the NTP study DBAN was administered in drinking water, a dose route more
relevant to environmental exposure than the corn-oil gavage dosing employed by Hayes et al.
(1986). Second, although the NTP 13-week study did not identify a LOAEL, the NOAELs for
decreased body weight were the same for the 14-day and 13-week NTP studies, and the LOAEL
was 18 mg/kg/day in the 14-day study. Since slight body weight decreases were also observed in
the 13-week study at 11.3 mg/kg/day, this suggests that the LOAEL for the 13-week study might
approximate the LOAEL of 18 mg/kg/day for the 14-day study, which is significantly lower than
the LOAEL of 45 mg/kg/day reported in Hayes et al. (1986). This argues that the
NOAEL/LOAEL boundary would be lower in the NTP (2002) study than in Hayes et al. (1986).
Third, since the NTP (2002) and Hayes et al. (1986) studies are not directly comparable, due to
differences in the methods of dose administration and rats strains employed, and both studies were

EPA/OW/OST/HECD

VIII-16

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
of adequate quality to derive the Longer-term HA, selection of the lower study NOAEL would be

most appropriate, even in the absence of a LOAEL.

The Longer-term HA value for a 10-kg child for DBAN may be calculated as shown
below and summarized in Table VIII-3. Derivation of the health advisories is shown using the
study NOAEL of 11.3 mg/kg/day from the NTP (2002) 13-week study as the point of departure.
An uncertainty factor of 10 is used to account for extrapolation from an animal study and an
uncertainty factor of 10 is used to account for inter-individual variability in human sensitivity, in
the absence of sufficient data to depart from these defaults. An additional uncertainty factor of 3
is used to account for database insufficiencies. This factor is selected since none of the available
reproductive or developmental studies were adequate to use in the quantitative dose-response
assessment. The data gap may be particularly relevant since cyanide, a metabolite of DBAN,
induces male reproductive system toxicity (U.S. EPA, 2002c), and due to uncertainty regarding
the significance of the testes effects observed in the 14-day NTP (2002) study for DBAN. The
reproductive and developmental toxicity study by R.O.W. Sciences (1997) was limited by the fact
that this was a screening study that was not designed to evaluate the full spectrum of endpoints of
interest. The developmental toxicity study by Smith et al. (1987) is of limited use, because it was
a single-dose study, because an insufficient array of endpoints was evaluated, and because the
observed toxicity was confounded by the use of tricaprylin as the solvent vehicle. A full factor of
10 was not used for the database uncertainty factor since the systemic toxicity of DBAN has been
tested in two species in subchronic studies (NTP, 2002; Hayes et al., 1986). The composite
uncertainty factor used is 300.

EPA/OW/OST/HECD

VIII-17

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Derivation of the Longer-term HA based on the study NOAEL

Longer-Term HA

(11.3 mg/kg/dav) (10 kg)
(300) (1 L/day)

0.38 mg/L (rounded to 0.4 mg/L)

where:

11.3 mg/kg/day = NOAEL in male F344 rats exposed to DBAN in drinking water for 13-
weeks (NTP, 2002).

10 kg = assumed body weight of a child.

300 = composite uncertainty factor, chosen to account for extrapolation from a NOAEL
in animals, inter-individual variability in humans, and insufficiencies in the database.

1 L/day = assumed daily water consumption by a 10-kg child.

The Longer-term HA for a 70-kg adult consuming 2 L/day of water is calculated as
follows:

where:

11.3 mg/kg/day = NOAEL in male F344 rats exposed to DBAN in drinking water for 13-
weeks (NTP, 2002).

70 kg = assumed body weight of an adult.

300 = composite uncertainty factor, chosen to account for extrapolation from a NOAEL
in animals, inter-individual variability in humans, and insufficiencies in the database.

2 L/day = assumed daily water consumption by a 70-kg adult.

B.2.4 Lifetime Health Advisory for DBAN

No chronic studies of DBAN toxicity were located, although DBAN is currently under
test for chronic toxicity in mice and rats (NTP, 2002). In the absence of such data, the available

Longer-Term HA

(11.3 mg/kg/dav) (70 kg)
(300)(2 L/day)

1.3 mg/L (rounded to 1 mg/L)

EPA/OW/OST/HECD

VIII-18

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
subchronic studies (NTP, 2002; Hayes et al., 1986) may be used to derive the Lifetime HA. As

described for the Longer-term HA, the NOAEL of 11.3 mg/kg/day for male rats identified in the

NTP (2002) study is the most appropriate basis for deriving the RfD. The derivation of the RfD

is shown below and summarized in Table VIII-3. An uncertainty factor of 10 is used to account

for extrapolation from an animal study and an uncertainty factor of 10 is used to account for inter-

individual variability in human sensitivity, in the absence of sufficient data to depart from these

defaults. An uncertainty factor of 3, instead of the default value of 10, was chosen to account for

less-than-lifetime exposure, based on the absence of progression of toxicological effects (or even

regression) from 14 days to 90 days (NTP, 2002; Hayes et al., 1986). An uncertainty factor of 3

is used to account for insufficiencies in the database. This factor was chosen to replace the default

factor of 10 because the subchronic toxicity of DBAN has been evaluated in two species (NTP,

2002; Hayes et al., 1986). Furthermore, decreased body weight was the identified as the most

sensitive effect in both studies, even though the NTP study included a thorough examination of

tissue histopathology, hematology, and clinical chemistry. These results suggest that no new

systemic target organs for DBAN are likely to be identified. However, none of the available

reproductive or developmental studies were adequate to use in the quantitative dose-response

assessment. The data gap may be particularly relevant since cyanide, a metabolite of DBAN,

induces male reproductive system toxicity (U.S. EPA, 2002c), and due to uncertainty regarding

the significance of the testes effects observed in the 14-day NTP (2002) study for DBAN. The

reproductive and developmental toxicity study by R.O.W. Sciences (1997) was limited by the fact

that this was a screening study that was not designed to evaluate the full spectrum of endpoints of

interest. The developmental toxicity study by Smith et al. (1987) is of limited use, because it was

a single-dose study, because an insufficient array of endpoints was evaluated, and because the

EPA/OW/OST/HECD

VIII-19

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

observed toxicity was confounded by the use of tricaprylin as the solvent vehicle. Therefore,
based on default factors of 10 each for interspecies extrapolation and inter-individual variability,
and partial factors of 3 each for subchronic to chronic extrapolation and for database
insufficiencies (lack of adequate developmental and reproductive toxicity studies), the composite
uncertainty factor used is 1000.

Derivation of the Lifetime HA based on the study NOAEL.

Step 1: Determination of the Reference Dose (RfD) for DBAN

RfD = ^' ' (1	= 0.011 mg/kg/day (rounded to 0.01 mg/kg/day)

where:

11.3 mg/kg/day = NOAEL in male F344 rats exposed to DBAN in drinking water for 13-
weeks (NTP, 2002).

1000 = composite uncertainty factor chosen to account for extrapolation from a NOAEL
in animals, inter-individual variability in humans, less-than-lifetime exposure, and
insufficiencies in the database.

Step 2: Determination of the Drinking Water Equivalent Level (DWEL) for DBAN
tyyxttjt (0.011 mg/kg/day) (70 kg)

DWEL =	(2 L/day)	=0.39 mg/L (rounded to 0.4 mg/L)

where:

0.011 mg/kg/day = RfD (before rounding)

70 kg = assumed body weight of an adult.

2 L/day = assumed daily water consumption by a 70-kg adult.

Step 3: Determination of the Lifetime Health Advisory for DBAN

Lifetime HA = (0.39 mg/L) (20%) = 0.078 mg/L (rounded to 0.08 mg/L)

EPA/OW/OST/HECD	VIII-20	Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

where:

0.39 mg/L = DWEL

20% = assumed relative source contribution from water

Table VIII-2 Summary of Oral Studies of DBAN Toxicity

Reference

Species/
Strain

Route/
Dose

Exposure
Duration

Endpoints Evaluated

NOAEL
(mg/kg/day)

LOAEL
(mg/kg/day)

Hayes et al.
(1986)

Mouse-
B6C3F1

Gavage in corn
oil

25 - 3,200
mg/kg/day

Acute

Lethality

NDa

LD50 = 289 (M)
303 (F)

Rat-
CD

Gavage in corn
oil

25 - 1,600
mg/kg/day

Acute

Lethality

ND

LD50 = 245 (M)
361 (F)

Eastman
Kodak Co.
(1992)

Mouse
Not

specified

Gavage

25 - 1,600
mg/kg/day

Acute

Lethality

ND

LD50 = 50

Rat
Not

specified

Gavage

25 - 3200
mg/kg/day

Acute

Lethality

ND

LD50 = 50 - 100

Meier et al.
(1985)

Mouse-
B6C3F1

Gavage in water

0, 12.5,25, or
50 mg/kg/day

5 Days

Sperm head
abnormalities

50 (Free-
standing
NOAEL)

ND

R.O.W

Sciences

(1997)

Rat-

Sprague-
Dawley

Drinking Water

0,0.7,2.2,5.8,
13.2 mg/kg/day
(males)

0,0.8,2.4,6.8,
17.9 mg/kg/day
(females)

14 Days

Clinical signs, body
weight, food
consumption

13.2 (m); 17.9
(f) (Free-
standing
NOAEL)

ND

Hayes et al.
(1986)

Rat-
CD

Gavage in corn
oil

0, 23,45, 90,
180 mg/kg/day

14 Days

Body weight, organ
weight, serum
chemistry, hematology,
urinalysis, gross
necropsy

23

45 (Decreased
body weight in
males)

EPA/OW/OST/HECD

VIII-21

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Reference

Species/
Strain

Route/
Dose

Exposure
Duration

Endpoints Evaluated

NOAEL
(mg/kg/day)

LOAEL
(mg/kg/day)



Rat-
CD

Gavage in corn
oil

0, 6, 23,45
mg/kg/day

90 Days

Body weight, organ
weight, serum
chemistry, hematology,
urinalysis, gross
necropsy

23

45 (Decreased
body weight in
males)

NTP (2002)

Mice-
B6C3F1

Drinking Water

0,2.1,4.3, 8.2,
14.7,21.4
mg/kg/day
(Males)

0,2.0, 3.3,
10.0, 13.9,21.6
mg/kg/day
(Females)

14 Days

Clinical signs, body
weight, water
consumption, organ
weight and pathology,
liver GST activity

21 (Free-
standing
NOAEL)





Rat-

Fischer-
344

Drinking Water

0, 2,3,7, 12,

18	mg/kg/day
(Males)

0, 2, 4, 7, 12,

19	mg/kg/day
(Females)

14 Days

Clinical signs, body
weight, water
consumption, organ
weight and pathology,
liver GST activity

12 (m)

18 (Decreased
body weight,
decreased testes
weight and
pathology in
males)



Mice-
B6C3F1

Drinking Water

0, 1.6, 3.2, 5.6,
10.7, 17.9
mg/kg/day
(Males)

0, 1.6,3,6.1,
11.1, 17.9
mg/kg/day
(Females)

13 Weeks

Clinical signs, body
weight, water
consumption, organ
weight and pathology,
hematology and clinical
chemistry

17.9 (m) (Free-
standing
NOAEL)





Rat-

Fischer-
344

Drinking Water

0,0.9, 1.8,3.3,
6.2, 11.3
mg/kg/day
(Males)

0, 1, 1.9, 3.8,
6.8, 12.6
mg/kg/day
(Females)

13 Weeks

Clinical signs, body
weight, water
consumption, organ
weight and pathology,
hematology and clinical
chemistry

11.3 (m); 12.6
(!) (Free-
standing
NOAEL)



EPA/OW/OST/HECD

VTII-22

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Reference

Species/

Route/

Exposure

Endpoints Evaluated

NOAEL

LOAEL



Strain

Dose

Duration



(mg/kg/day)

(mg/kg/day)

R.O.W

Rat-

Drinking Water

(M) 30

(M) Clinical pathology,

Paternal: 8.2

ND

Sciences





Days, (F)

organ weight, sperm

(M); 10.8 (F)



(1997)

Sprague-



35 days

analysis,







Dawley

0, 1.4, 3.3, 8.2

penconce

histopathology: (F)

Reproductive/de







mg/kg/day

ption or
35 days
gestation
day 5 to
PND 1

maternal weight,
reproductive success,
pup viability and
growth

velopmental:
8.2 (M); 10.8
(F)

(Free-standing
NOAEL)



Smith et al.

Rat-

Gavage in

Gestation

Maternal weight,

Maternal: ND

Maternal: 50

(1987)



tricapyrlinb

days 7 to

reproductive success,



(FEL for



Long-



21

pup viability and



maternal death;



Evans

50 mg/kg/day



growth



decrease



Hooded







Developmental:
ND

maternal weight
gain)

Development:
50 (Decreased
litter size,
decreased fetal
weight)

a.	ND = not determined.

b.	Due to the demonstrated ability of tricaprylin to enhance developmental toxicity of TCAN, this study was not considered in derivation of the
Health Advisories.

EPA/OW/OST/HECD

VTII-23

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Table VTTT-3 Summary of Development of the Health Advisories for DBAN

Study

Critical Effect

Critical
Effect
Level

Uncertainty
Factors3

RfD

(mg/kg/day)

Health
Advisory
(mg/L)

Ten-day

NTP (2002)

Decreased body
weight, decreased
testes weight, and
testes atrophy

12

mg/kg/day
(NOAEL)

100 (10H, 10A)



1

Longer-term

NTP (2002)

Decreased body
weight

11.3

mg/kg/day
(NOAEL)

300 (10H, 10A,
3d)



Child 0.4
Adult 1

Lifetime

NTP (2002)

Decreased body
weight

11.3

mg/kg/day
(NOAEL)

1000 (10H, ioA,

3S, 3d)

0.01

0.08

a. Areas of uncertainty addressed by uncertainty factors are: animal to human extrapolation (A); intrahuman variability and
protection of sensitive subpopulations (H); extrapolation from a LOAEL to a NOAEL(L); extrapolation from a subchronic
to chronic exposure (S); and lack of a complete database (D)

B.3 DCAN

B.3.1 One-Day Health Advisory for DCAN

The oral toxicity data for DCAN are summarized in Table VIII-4. Hayes et al. (1986)
identify acute oral LD50 values for DCAN of 270 to 339 mg/kg. Clinical signs observed include
ataxia, depressed respiration, depressed activity, and coma. However, LD50 studies are not
suitable for the development of one-day health advisories and no other adequate acute studies
were located for DCAN. In the absence of such data, the Ten-day HA value is recommended as a
conservative estimate of an appropriate One-day HA value.

B.3.2 Ten-Day Health Advisory for DCAN

One general toxicity study of suitable duration for the Ten-day HA was located. Hayes et
al. (1986) conducted a 14-day study of DCAN toxicity in rats. Body weight decreases were

EPA/OW/OST/HECD

VTII-24

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

observed in both males and females. Males were more sensitive to this effect than females. In
males, a decrease in body weight of greater than 10% was observed at 45 and 90 mg/kg/day,
although these results were not statistically significant. Several serum markers for organ toxicity
were increased in treated animals. Significantly increased SGPT levels in females at 90
mg/kg/day, and ALP levels at 90 mg/kg/day in males and at 45 and 90 mg/kg/day in females were
reported, possibly indicative of hepatotoxicity. Although the authors did not consider these
changes to be compound-related adverse effects (no reason provided), these changes were
considered adverse for this assessment, based on the magnitude of the changes, and the
supporting data for DCAN in females in the subchronic study. No remarkable findings were
observed at necropsy; however, relative liver weight was significantly increased (p < 0.05) in male
and female rats. The observed increase in serum levels of hepatic enzyme activity at higher doses
than those associated with liver weight gives greater weight to the potential toxicological
significance of the liver weight changes, even though the absence of histopathology data makes it
difficult to determine conclusively if the effects were adverse at low doses. Based on this
uncertainty, both decreased body weight and increased relative liver weight are considered
toxicologically-relevant responses. The more sensitive of these endpoints was selected as the
critical effect for this study. Therefore, the lowest dose tested of 12 mg/kg/day is the study
LOAEL for increased relative liver weight in males, and no NOAEL is determined. BMD
modeling was conducted for decreased body weight and increased relative liver weight in both
sexes to identify alternative critical effect levels for this study. A BMDL of 5 mg/kg/day for
increased relative liver weight in males was selected as the most appropriate modeling result to
serve as the basis for the quantitative dose-response assessment (see Appendix A).

EPA/OW/OST/HECD

VTII-25

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

As a follow-up to their earlier single-dose study, Smith et al. (1989) reported that doses of
25 to 45 mg/kg/day administered for 12 days during gestation resulted in fetotoxicity and
teratogenicity in rats. However, use of this study for dose-response assessment is not
appropriate, because it may not accurately reflect the toxicity of DC AN in drinking water. This
conclusion is based on the observation of embryotoxicity of the tricaprylin vehicle in this study
and later work by this laboratory which suggests that tricaprylin may act synergistically with
TCAN to enhance developmental toxicity (Christ el al., 1996).

Based on increased relative liver weight in male rats (Hayes et al., 1986) the Ten-day HA
was calculated as shown below and summarized in Table VIII-5. An uncertainty factor of 10 is
used to account for extrapolation from an animal study and an uncertainty factor of 10 is used to
account for inter-individual variability in human sensitivity. An additional factor of 3 was used to
account for extrapolation from a minimal LOAEL. A factor of 10 was not used since the adverse
effect (increased relative liver weight) was of marginal severity (i.e. no clinical chemistry findings
were observed at this dose). This additional factor was not used for derivation of the Ten-day
HA when the BMDL was used as the point of departure, since the BMDL often approximates a
NOAEL as indicated by the lower value of the BMDL for the same effect the critical study. The
composite uncertainty factor used is 300 when the LOAEL was used as the point of departure,
and 100 when the BMDL was used as the point of departure.

Derivation of the Ten-day HA based on the study LOAEL.

T J TTA (12 mg/kg/day) (10 kg)

Ten-day HA = (300)(1 L/day) = 0.4 mg/L

EPA/OW/OST/HECD VIII-26 Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

where:

12 mg/kg/day = LOAEL, based on increased relative liver weight in males supported by
clinical chemistry findings at higher doses in rats exposed to DCAN by gavage for 14 days
(Hayeses a/., 1986).

10 kg = assumed body weight of a child.

300 = composite uncertainty factor, chosen to account for extrapolation from a minimal
LOAEL in animals, and inter-individual variability in humans.

1 L/day = assumed daily water consumption by a 10-kg child.

Derivation of the Ten-day HA based on the study BMDL.

^ , T T a (5 mg/kg/day) (10 kg)

Ten-day HA = (100)(1 L/day) = 0.5 mg/L

where:

5 mg/kg/day = BMDL, based on increased relative liver weight in males supported by
clinical chemistry findings at higher doses in rats exposed to DCAN by gavage for 14 days
(Hayeses a/., 1986).

10 kg = assumed body weight of a child.

100 = composite uncertainty factor, chosen to account for extrapolation from a BMDL in
animals, and inter-individual variability in humans.

1 L/day = assumed daily water consumption by a 10-kg child.

B.3.3 Longer-Term Health Advisory for DCAN

Only one study of suitable duration for the derivation of a longer-term HA was located.
Hayes et al. (1986) conducted a 90-day subchronic toxicity study in rats. Body weight was
significantly decreased in high-dose male and female rats and in male rats at 33 mg/kg/day.
Relative liver weights were statistically significantly elevated in males beginning at 33 mg/kg/day
and in females beginning at 8 mg/kg/day. However, relative liver weight increases were greater

EPA/OW/OST/HECD	VIII-27	Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
than 10% at 8 mg/kg/day in both males and females, and therefore the increase in relative liver

weight at this dose was considered toxicologically relevant for both sexes. The observed increase

in serum levels of ALP activity in the subchronic study, and the increase in both ALP and SGPT

observed in the 14-day study support the toxicological relevance of the liver weight findings. The

absence of histopathology data makes it difficult to determine conclusively if the effects were

adverse at low doses. Based on this uncertainty, both decreased body weight and increased

relative liver weight are considered toxicologically-relevant responses. The more sensitive of

these endpoints was selected as the critical effect. Therefore, the lowest dose tested of 8

mg/kg/day is the study LOAEL for increased relative liver weight in males and females, and no

NOAEL is determined. BMD modeling was conducted for decreased body weight and increased

relative liver weight in both sexes to identify alternative critical effect levels for this study. A

BMDL of 4 mg/kg/day for increased relative liver weight in males was selected as the most

appropriate modeling result to serve as the basis for the quantitative dose-response assessment,

using a benchmark response (BMR) of a one standard deviation decrease in relative liver weight

(see Appendix A).

Based on the increased relative liver weight in rats (Hayes et al., 1986), the Longer-term
HA value for a 10-kg child may be calculated as shown below and summarized in Table VIII-5.
Derivation of the health advisories are shown when either the study LOAEL (8 mg/kg/day) or
BMDL (4 mg/kg/day) is selected as the point of departure. An uncertainty factor of 10 is used to
account for extrapolation from an animal study and an uncertainty factor of 10 is used to account
for human variability in sensitivity, in the absence of sufficient data to depart from these defaults.
An uncertainty factor of 10 is used to account for insufficiencies in the database. This factor was

EPA/OW/OST/HECD

VTII-28

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

chosen because only one subchronic toxicity study in a single species was identified for derivation
of the Longer-term HA (Table VIII-4). The absence of a systemic toxicity study of suitable
duration in a second species, the lack of histopathology data in the existing 90-day study, and
failure to investigate effects associated with thiocyanate (an identified metabolite) or cyanide (the
likely precursor of thiocyanate), such as thyroid or central nervous system effects, further
weakens the database. In addition, no adequate studies on reproductive or developmental toxicity
were reported. The only available developmental toxicity study testing multiple dose levels
(Smith et al., 1989) was compromised by the use of tricaprylin as the solvent vehicle and was
judged as inadequate for use in the quantitative dose-response assessment. If the LOAEL is
selected as the point of departure, an additional factor of 3 is used to account from extrapolation
from a LOAEL for minimally adverse liver effects, since no clinical chemistry changes were
observed at this dose to accompany the observed increases in liver weight. The composite
uncertainty factor used is 3000 with the LOAEL as the point of departure and 1000 with the
BMDL as the point of departure, based on full factors of 10 each for interspecies extrapolation
and inter-individual human variability, a factor of 10 for database insufficiencies, and, for the
LOAEL only, a partial factor of 3 for a minimal LOAEL.

Derivation of the Longer-term HA based on the study LOAEL.

Longer-Term HA = (8 ^^Q^L/day^ = ° °27 mg/L (rounded t0 0 03 mS/L)
where:

8 mg/kg/day = LOAEL, based on increased relative liver weight in male and female rats

exposed to DCAN by gavage for 90 days (Hayes et al., 1986).

EPA/OW/OST/HECD

VTII-29

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

10 kg = assumed body weight of a child.

3000 = composite uncertainty factor, chosen to account for extrapolation from a LOAEL
in animals, inter-individual variability in humans, and insufficiencies in the database.

1	L/day = assumed daily water consumption by a 10-kg child.

The Longer-term HA for a 70-kg adult consuming 2 L/day of water is calculated as
follows:

Longer-Term HA = (8	= 0 093 m8/L (rounded to 0.09 mg/L)

(3000)(2 L/day)

where:

8 mg/kg/day = LOAEL, based on increased relative liver weight in male and female rats
exposed to DCAN by gavage for 90 days (Hayes et al., 1986).

70 kg = assumed body weight of an adult

3000 = composite uncertainty factor, chosen to account for extrapolation from a LOAEL
in animals, inter-individual variability in humans, and insufficiencies in the database.

2	L/day = assumed daily water consumption by a 70-kg adult

Derivation of the Longer-term HA based on the study BMDL.

t	t ua (4 mg/kg/day) (10 kg)

Longer-Term HA = (1000)(1 L/day) = 0 04 mg/L

where:

4 mg/kg/day = BMDL, based on increased relative liver weight in male rats exposed to
DCAN by gavage for 90 days (Hayes et al., 1986).

10 kg = assumed body weight of a child.

1000 = composite uncertainty factor, chosen to account for extrapolation from a BMDL
in animals, inter-individual variability in humans, and insufficiencies in the database.

1 L/day = assumed daily water consumption by a 10-kg child.

EPA/OW/OST/HECD

VTII-30

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

The Longer-term HA for a 70-kg adult consuming 2 L/day of water is calculated as
follows:

Longer-Term HA = (4 ^= °-14 m8/L (rounded to 0.1 mg/L)

(1000)(2 L/day)

where:

4 mg/kg/day = BMDL, based on increased relative liver weight in male rats exposed to
DCAN by gavage for 90 days (Hayes et al., 1986).

70 kg = assumed body weight of an adult

1000 = composite uncertainty factor, chosen to account for extrapolation from a BMDL
in animals, inter-individual variability in humans, and insufficiencies in the database.

2 L/day = assumed daily water consumption by a 70-kg adult

B.3.4 Lifetime Health Advisory for DCAN

No chronic studies of DCAN toxicity were located. In the absence of such data, the
subchronic (90-day) LOAEL of 8 mg/kg/day or the BMDL of 4 mg/kg/day for increased relative
liver weight in rats reported in the study by Hayes et al. (1986) may be employed to derive the
RfD as shown below and summarized in Table VIII-5. An uncertainty factor of 10 is used to
account for extrapolation from an animal study and an uncertainty factor of 10 is used to account
for inter-individual variability in human sensitivity, in the absence of sufficient data to depart from
these defaults.

An additional factor of 3 is used to account for less-than-lifetime exposure for DCAN.
This factor of 3 was used to replace the default factor of 10 for extrapolation from a subchronic
study. In selecting a factor of 3, comparison of the critical effect levels between the 14-day and
90-day studies (Hayes et al., 1986) is not helpful since the low dose was the LOAEL for

EPA/OW/OST/HECD

VTII-31

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
increased liver weight in both cases. An added complication is that the same dose levels were not

used in the 14-day and 90-day studies, which makes comparison of the severity of effects with

duration more difficult. However, examination of the liver weight changes across durations

suggests that the uncertainty factor for extrapolation from a subchronic-to-chronic study would

be adequately accounted by an uncertainty factor of 3. This conclusion is supported by the

observation that for males a similar degree of change in relative liver weight (increase of 13%)

was observed over a 1.5-fold change in dose (8 mg/kg/day for 90 days versus 12 mg/kg/day for

14 days). Furthermore, in females the increase in relative liver weight was roughly two-fold

greater at 8 mg/kg/day in the 90-day study than at 12 mg/kg-day in the 14-day study, suggesting a

differences in responsiveness of 3-fold on an equal dose basis (1.5-fold decrease in dose x 2-fold

increase in effect). A similar type of evaluation for higher doses suggests that the difference in the

magnitude of liver weight increases is not likely to increase by ten-fold, with increasing exposure

duration, although some increase in magnitude of the effect is observed with longer-term

exposure. Finally, the BMDL calculated for the 14-day study (5 mg/kg/day) is nearly identical to

the BMDL calculated for the 90-day study (4 mg/kg/day) for the same endpoint. Since the

BMDL is based on a defined change in mean and not dependent on the dose levels, the similarity

of the BMDLs across study duration suggest minimal progression.

An uncertainty factor of 10 is used to account for insufficiencies in the database. This
factor was chosen because only one subchronic toxicity study in a single species was identified for
derivation of the Lifetime HA (Table VIII-4). The absence of a systemic toxicity study of suitable
duration in a second species, the lack of histopathology data in the existing 90-day study, and
failure to investigate effects associated with thiocyanate (an identified metabolite) or cyanide (the

EPA/OW/OST/HECD

VTII-32

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
likely precursor of thiocyanate), such as thyroid or central nervous system effects, further

weakens the database. In addition, no sufficient studies on reproductive or developmental toxicity

were reported. The only available multiple-dose developmental toxicity study (Smith et. al.,

1989) was compromised by the use of tricaprylin as the solvent vehicle and was judged as

inadequate for use in the quantitative dose-response assessment.

If the LOAEL is selected as the point of departure an additional factor of 3 is used to
account from extrapolation from a LOAEL for minimally adverse liver effects, since no clinical
chemistry changes were observed at this dose to accompany the observed increases in relative
liver weight. When the LOAEL is used as the point of departure, the composite uncertainty
factor is based on three full factors of 10 and two partial factors of 3. This is equivalent to four
full areas of uncertainty. Based on EPA policy (Dourson, 1994) the composite uncertainty factor
for four individual factors of 10 is 3000, reflecting overlap in the individual factors. Therefore,
the composite uncertainty factor is 3000 when the LOAEL is used as the point of departure.

When the BMDL is used as the point of departure, the composite uncertainty factor is based on
three full factors of 10 and one partial factor of 3. Therefore, the composite uncertainty factor is
also 3000 when the BMDL is used as the point of departure.

Derivation of the Lifetime HA based on the study LOAEL.

Step 1: Determination of RfD for DC AN

RfD = ^ (3000)^aV' = ^-0027 mg/kg/day (rounded to 0.003 mg/kg/day)
where:

EPA/OW/OST/HECD

VTII-33

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

8 mg/kg/day = LOAEL, based on increased relative liver weight in male and female rats
exposed to DCAN by gavage for 90 days (Hayes et al., 1986).

3000 = composite uncertainty factor chosen to account for extrapolation from a LOAEL
in animals, inter-individual variability in humans, less-than-lifetime exposure, and
insufficiencies in the database.

Step 2: Determination of the Drinking Water Equivalent Level (DWEL) for DCAN

(0.0027 mg/kg/day) (70 kg)	„ . . .	„.

DWEL = 	(2 L/day) 	 = 0 094 mg/L (rounded to 0.09 mg/L)

where:

0.0027 mg/kg/day = RfD.

70 kg = assumed body weight of an adult.

2 L/day = assumed water consumption of a 70-kg adult.

Step 3: Determination of Lifetime HA for DCAN

Lifetime HA = (0.094 mg/L) (20%) = 0.019 mg/L (rounded to 0.02 mg/L)

where:

0.094 mg/L = DWEL

20% = assumed relative source contribution from water

Derivation of the Lifetime HA based on the study BMDL.

Step 1: Determination of RfD for DCAN

RfD = ^ (3000)^aV' = 0.0013 mg/kg/day (rounded to 0.001 mg/kg/day)

where:

4 mg/kg/day = BMDL, based on increased relative liver weight in male rats exposed to
DCAN by gavage for 90 days (Hayes et al., 1986).

EPA/OW/OST/HECD

VTII-34

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

3000 = composite uncertainty factor chosen to account for extrapolation from a BMDL
in animals, inter-individual variability in humans, less-than-lifetime exposure, and
insufficiencies in the database.

Step 2: Determination of the Drinking Water Equivalent Level (DWEL) for DCAN

where:

0.0013 mg/kg/day = RfD (before rounding)

70 kg = assumed body weight of an adult.

2 L/day = assumed water consumption of a 70-kg adult.

Step 3: Determination of Lifetime HA for DCAN

Lifetime HA = (0.046 mg/L) (20%) = 0.0092 mg/L (rounded to 0.009 mg/L)

DWEL

(0.0013 mg/kg/dav) (70 kg)
(2 L/day)

0.046 mg/L (rounded to 0.05 mg/L)

where:

0.046 mg/L = DWEL

20% = assumed relative source contribution from water

EPA/OW/OST/HECD

VTII-35

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Table VIII-4 Summary of Oral Studies of DC AN Toxicity

Reference

Species/
Strain

Route/
Dose

Exposure
Duration

Endpoints Evaluated

NOAEL
(mg/kg/day)

LOAEL (mg/kg/day)

Hayes et
al. (1986)

Mouse-
B6C3F1

Gavage in
corn oil

25 - 3,200
mg/kg/day

Acute

Lethality

NDa

LD50 = 270 (M)
279 (F)



Rat-
CD

Gavage in
corn oil

25 - 1,600
mg/kg/day

Acute

Lethality

ND

LD50 =339 (M)
330 (F)



Rat-
CD

Gavage in
corn oil

0, 12, 23,
45, 90
mg/kg/day

14 Days

Body weight, organ
weight, serum
chemistry,
hematology,
urinalysis, gross
necropsy

ND

12 (Increased liver
weight)



Rat-
CD

Gavage in
corn oil

0, 8,33,65
mg/kg/day

90 Days

Body weight, organ
weight, serum
chemistry,
hematology,
urinalysis, gross
necropsy

ND

8 (Increased liver
weight)

Meier et
al. (1985)

Mouse-
B6C3F1

Gavage in
water

0, 12.5,25,
or 50

mg/kg/day

5 Days

Sperm head
abnormalities

50 mg/kg

(Free-standing
NOAEL)

ND

Smith et
al. (1987)

Rat-

Long-
Evans
Hooded

Gavage in
tricapyrlinb

55

mg/kg/day

Gestation
days 7 to
21

Maternal weight,
reproductive success,
pup viability and
growth

ND

Maternal: 55
(Decreased maternal
weight)

Development: 55
(Decreased pregnancy
rate; decreased viable
litters; increased
litters resorbed;
decreased fetal
weight)

EPA/OW/OST/HECD

VTII-36

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Reference

Species/

Route/

Exposure

Endpoints Evaluated

NOAEL

LOAEL (mg/kg/day)



Strain

Dose

Duration



(mg/kg/day)



Smith et

Rat-

Gavage in

Gestation

Maternal weight,

Maternal: 15

Maternal: 25

al. (1989)



tricapyrlinb

days 6 to

reproductive success,



(increased liver



Long-



18

pup viability and



weight)



Evans

0, 5, 15,



growth







Hooded

25,45
mg/kg/day





Developmental:
15

Development: 25
(Increased post-
implantation loss,
increased soft-tissue
malformations)

a.	ND = not determined.

b.	Due to the demonstrated ability of tricaprylin to enhance developmental toxicity of TCAN, this study was considered in
derivation of the Health Advisories.

Table VIII-5 Summary of Development of the Health Advisories for DC AN

Study

Critical Effect

Critical
Effect
Level

Uncertainty
Factors3

RfD

(mg/kg/day)

Health
Advisory
(mg/L)

Ten-day

Hayes et al.
(1986)

Increased relative
liver weight

12

mg/kg/day
(LOAEL)

300 (10H, 10A,
3l)



0.4

Hayes et al.
(1986)

Increased relative
liver weight

5

mg/kg/day
(BMDL)

100 (10H, 10A)



0.5

Longer-term

Hayes et al.
(1986)

Increased relative
liver weight

8

mg/kg/day
(LOAEL)

3000 (10H, 10A,

3l, 10d)



Child 0.03
Adult 0.09

Hayes et al.
(1986)

Increased relative
liver weight

4

mg/kg/day
(BMDL)

1000 (10H, ioA,

10D)



Child 0.04
Adult 0.1

Lifetime

Hayes et al.
(1986)

Increased relative
liver weight

8

mg/kg/day
(LOAEL)

3000 (10H, 10A,
3S, 3l, 10d)

0.003

0.02

Hayes et al.
(1986)

Increased relative
liver weight

4

mg/kg/day
(BMDL)

3000 (10H, 10A,

3S, 10D)

0.001

0.009

a. Areas of uncertainty addressed by uncertainty factors are: animal to human extrapolation (A); intrahuman variability and
protection of sensitive subpopulations (H); extrapolation from a LOAEL to a NOAEL(L); extrapolation from a subchronic to
chronic exposure (S); and lack of a complete database (D)

EPA/OW/OST/HECD

VTII-37

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

B.4 TCAN

B.4.1 One-Day Health Advisory for TCAN

The oral toxicity data for TCAN are summarized in Table VIII-6. An oral LD50 of 360
mg/kg has been reported (Smyth et al., 1962). However, LD50 studies are not suitable for the
development of one-day health advisories and no other adequate acute studies were located for
TCAN. In the absence of suitable acute data, the Ten-day HA value is recommended as a
conservative estimate of the One-day HA value.

B.4.2 Ten-Day Health Advisory for TCAN

Two developmental studies have been conducted which evaluate the short-term effects of
TCAN. In Smith et al. (1988), oral exposure of pregnant rats to TCAN in tricaprylin on gestation
days 6 through 18 resulted in significant fetotoxic and teratogenic effects at doses of 7.5
mg/kg/day or higher. Although the increased incidence of teratogenic effects was not statistically
significant at a dose of 1 mg/kg/day, the authors expressed concern that this might be of biological
significance. However, further evaluation of these data based on litter incidences did not reveal a
treatment-related effect at 1 mg/kg/day. In a later study by Christ et al. (1996), oral exposure of
pregnant rats to TCAN in corn oil on gestation days 6 to 18 resulted in significantly reduced
maternal weight gain at doses of 35 mg/kg/day and higher. Fetotoxic and teratogenic effects were
not observed until doses of 55 mg/kg/day and higher. Because of the confounding effect of
tricaprylin toxicity, as identified by Christ et al. (1996), exposure to TCAN in corn oil is more
appropriate basis for a health advisory. Therefore, the Christ et al. (1996) study is selected as the
critical study for derivation of the Ten-day HA as shown below and summarized in Table VIII-7.
In general, developmental toxicity endpoints arising from in utero exposure have limited

EPA/OW/OST/HECD

VTII-38

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
application as the basis for the derivation of the Ten-day HA, since the Ten-day HA is based on

water consumption by children directly, and not from maternal exposure. However, when the

critical effect is a general systemic endpoint (e.g., decreased maternal weight gain) that is likely to

be relevant to exposed children, it is appropriate to use these data to derive the Ten-day HA.

Based on this consideration, the study by Christ et al. (1996) is judged as an adequate basis for

the Ten-day HA. Based on the critical effect of decreased maternal body weight gain, the

NOAEL is 15 mg/kg/day and the LOAEL is 35 mg/kg/day. BMD modeling was conducted to

identify alternative critical effect levels for this study. A BMDL of 17 mg/kg/day for decreased

adjusted maternal weight gain was selected as the most appropriate modeling result to serve as

the basis for the quantitative dose-response assessment (see Appendix A).

For derivation of the Ten-day HA, an uncertainty factor of 10 is used to account for
extrapolation from an animal study and an uncertainty factor of 10 is used to account for inter-
individual variability in human sensitivity, in the absence of sufficient data to depart from these
defaults. The composite uncertainty factor used is 100.

Derivation of the Ten-day HA based on the study NOAEL.

„ , (15 mg/kg/day) (10 kg) „ . , ,

Ten-day HA = q00) (1 L/day) m§ (rounded t0 2 mS/L)

where:

15 mg/kg/day = NOAEL, based on the absence of decreased adjusted maternal weight
gain in pregnant rats exposed by gavage on days 6 to 18 of gestation, with a
corresponding LOAEL of 35 mg/kg/day (Christ et al., 1996).

10 kg = assumed body weight of a child.

EPA/OW/OST/HECD

VTII-39

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

100 = composite uncertainty factor, chosen to account for extrapolation from a NOAEL
from a study in animals, and inter-individual variability in humans.

1 L/day = assumed water intake by a 10-kg child.

Derivation of the Ten-day HA based on the study BMDL.

„ ,	(17 mg/kg/dav) (10 kg)	„ . , ,

Ten-day HA = (100)(1 L/day)	= 17 m§/L (rounded to 2 mg/L)

where:

17 mg/kg/day = BMDL, based on the absence of decreased adjusted maternal weight gain
in pregnant rats exposed by gavage on days 6 to 18 of gestation (Christ et al., 1996).

10 kg = assumed body weight of a child.

100 = composite uncertainty factor, chosen to account for extrapolation from a NOAEL
from a study in animals, and inter-individual variability in humans.

1 L/day = assumed water intake by a 10-kg child.

B.4.3 Longer-Term and Lifetime Health Advisories for TCAN

No data on the effects of longer-term or chronic exposure to TCAN were located. In the
absence of suitable data, no values can be derived for the Longer-term or Lifetime HAs for
TCAN.

EPA/OW/OST/HECD

VTII-40

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Table VIII-6 Summary of Oral Studies of TCAN Toxicity

Reference

Species/
Strain

Route

Exposure
Duration

Endpoints
Evaluated

NOAEL
(mg/kg/day)

LOAEL (mg/kg/day)

Smyth et
al. (1962)

Rat-
Wistar

Gavage

0.19-0.32
mg/kg/day

Acute

Lethality

NDa

LD50 = 360

Meier et
al. (1985)

Mouse-
B6C3F1

Gavage in
water

0, 12.5,25,
or 50

mg/kg/day

5 Days

Sperm head
abnormalities

50 mg/kg (Free-
standing
NOAEL)

ND

Smith et
al. (1987)

Rat
Long-
Evans
Hooded

Gavage in
tricaprylinb

55

mg/kg/day

Days 7 to
21 of
gestation

Maternal weight,
reproductive
success, pup
viability and
growth

Maternal: ND

Developmental:
ND

Maternal: 55 (FEL for
maternal death; decrease
maternal weight gain)

Development: 55
(Decreased pregnancy
rate; decreased viable
litters; increased litters
resorbed; decreased fetal
weight)

Smith et
al. (1988)

Rat
Long-
Evans
Hooded

Gavage in
tricaprylinb

0, 1,7.5,
15, 35, 55
mg/kg/day

Days 6 to
18 of
gestation

Maternal weight,
reproductive
success, pup
viability and
growth,
malformations

Maternal: 35
Developmental: 1

Maternal: 55 (FEL for
maternal death; decrease
maternal weight gain)

Developmental: 7.5
(Increased full-liter
resorptions; increased
cardiovascular
malformations)

Christ et
al. (1996)

Rat

Long-

Evans

Gavage in
corn oilc

0, 15, 35,

55,75

mg/kg/day

Days 6 to
18 of
gestation

Maternal body and
organ weight,
reproductive
success, pup
viability and
growth,
malformations

Maternal: 15

Developmental:
35

Maternal: 35 (Decreased
maternal weight gain;
organ weight changes)

Development: 55
(increased post-
implantation loss,
cardiovascular and
cranio-facial
malformations;
decreased live fetuses
per litter, fetal body
weight, crown-rump
length.

a.	ND = not determined

b.	Due to the demonstrated ability of tricaprylin to enhance developmental toxicity of TCAN, this study was not considered
in derivation of the Health Advisories.

c.	Only data relating to the corn oil control are reported in the table, since the developmental toxicity reported in the groups
administered tricaprylin were not considered in derivation of the Health Advisories.

EPA/OW/OST/HECD

VTII-41

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Table VIII-7. Summary of Development of the Health Advisories for TCAN.

Study

Critical Effect

Critical

Uncertainty

RfD

Health





Effect

Factors3

(mg/kg/day)

Advisory





Level





(mg/L)

Ten-day

Christ et al.

Decreased adjusted

15

100 (10H, 10A)

.

2

(1996)

maternal body weight

mg/kg/day









gain

(NOAEL)







Christ et al.

Decreased adjusted

17

100 (10H, 10A)

.

2

(1996)

maternal body weight

mg/kg/day









gain

(BMDL)







Longer-term

Inadequate data to derive Health Advisory

Lifetime

Inadequate data to derive Health Advisory

a. Areas of uncertainty addressed by uncertainty factors are: animal to human extrapolation (A); intrahuman variability and
protection of sensitive subpopulations (H).

C. Carcinogenic Effects

No epidemiological studies have evaluated directly the carcinogenic potential of HANs in
humans. Rather, studies have evaluated the carcinogenic potential of chlorinated versus
unchlorinated drinking water or the presence of trihalomethanes as a marker of chlorination by-
products (IARC, 1999; Mills el al., 1998). No standard cancer bioassays of HANs have been
done in animals, although DBAN is on test for a full bioassay as part of the National Toxicology
Program (NTP, 2002). Limited short-term exposure data from the mouse skin assay (Bull et al.,
1985) and the mouse lung assay (Bull and Robinson, 1985) indicate that all four compounds
(BCAN, DBAN, and TCAN) may be tumorigenic, although DCAN, DBAN, and TCAN were
reported to be negative in the rat liver GGT-foci assay (Herren-Freund and Pereira, 1986). QSTR
predictions of the carcinogenicity of the HANs have produced mixed results (Moudgal et al.,

EPA/OW/OST/HECD

VTII-42

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
2000). For example, BCAN was predicted as positive in male and female mice, but negative in

both sexes of rats. DBAN was negative in both sexes of mice, and female rats, but indeterminate

in male rats. No carcinogenicity predictions for TCAN or DCAN were included in the study.

Taken together, the screening bioassays and QSTR predictions provide at least limited indications

of potential carcinogenicity of HANs, although the existing data are not adequate to demonstrate

positive carcinogenicity in animals.

The HANs or their metabolites are reactive compounds that can bind macromolecules
including DNA and proteins (Daniel et al., 1986; Lin et al., 1992). Glutathione conjugation may
be an important cellular protection against these reactive compounds (Ahmed et al., 1991),
although no glutathione conjugates or their metabolites have been identified. The genotoxicity of
each of the HAN compounds BCAN, DBAN, DCAN, and TCAN have been evaluated in at least
one, and in most cases a variety of different assays. Although some of the data have provided
contradictory results, all of the tested compounds appear to have some capacity to induce
genotoxic effects. For example, each compound has been found to generate a positive result in at
least one reported Salmonella!microsome assay, except DBAN. In addition, while not uniformly
consistent, a variety of other assays for DNA damage (i.e. DNA strand breaks) or responses to
DNA damage (sister chromatid assays, tests for gene recombination in yeast, and SOS
chromotest) have yielded positive results for some of the HANs. Overall, these data suggest that
HANs induce genotoxicity through direct interactions with DNA. Evidence for direct
chromosome effects are weaker, with inconsistent results reported in the limited number of
studies that evaluated formation of micronuclei and aneuploidy.

EPA/OW/OST/HECD

VTII-43

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

The International Agency for Research on Cancer (IARC, 1999) reviewed the data for

HANs and found that there is inadequate evidence for carcinogenicity in experimental animals for
all compounds. As a results, IARC classified these compounds as "not classifiable as to its
carcinogenicity to humans" (Group 3).

Following EPA's 1986 Guidelines for Carcinogen Risk Assessment (U.S. EPA, 1986),
BCAN, DBAN, DCAN, and TCAN are appropriately classified as Group D - Not Classifiable as
to Human Carcinogenicity. This classification is appropriate when there is inadequate evidence
of carcinogenicity in humans or animals. Following EPA's Draft 1999 Guidelines for Carcinogen
Risk Assessment (U.S. EPA, 1999), the data for the HANs can best be described as Data Are
Inadequate for an Assessment of Human Carcinogenic Potential.

D. Characterization of Uncertainties and Data Gaps

The available data are very limited for the HANs included in this Criteria Document.
Adequate human data are not available to evaluate noncancer or cancer effects. Full chronic
toxicity and carcinogenicity studies in animals were not available for any of the HANs. The
absence of adequate carcinogenicity testing is a significant data gap, particularly in light of the
mixed results in screening bioassays, genotoxicity testing, and QSTR modeling, which indicate at
least some potential for HANs to induce tumorigenic responses. Available subchronic studies for
DCAN did not include a full histopathology evaluation of relevant tissues. The absence of
histopathology evaluations following longer-term exposures resulted in significant uncertainties in
the assessment of noncancer toxicity, since the adversity of the increases in liver weight induced

EPA/OW/OST/HECD

VTII-44

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
by DCAN (Hayes el al., 1986) could not be substantiated, and detailed evaluations of known

targets for the oxidative pathway metabolites cyanide and thiocyanate was not possible.

Although the developmental toxicity of HANs has been evaluated in numerous animal
studies, the confounding effects of the tricaprylin solvent vehicle that was used in the earlier
studies greatly limits the usefulness of most of the data. The mechanism by which tricaprylin
impacted the developmental toxicity of the HANs in these studies remains unclear. Due to
confounding by tricaprylin, adequate developmental toxicity data are available only for TCAN
(Christ et al., 1996). No multi-generation reproductive studies were available for any of the
HANs. Only one study was available that evaluated the developmental effects of HANs in
humans (Klotz and Pyrch, 1999), and no association was observed between HANs exposure and
developmental effects. The potential reproductive and developmental toxicity of the HANs
remains a significant area of uncertainty in the current assessment, and represents a major data
gap in light of the reproductive and developmental effects attributed to other disinfectant
byproducts in humans (Neiuwenhuijsen et al., 2000).

The available data on HAN toxicokinetics and toxicodymanics was not sufficient to move
away from default uncertainty factor values for extrapolation from animal studies or for inter-
individual variability in human sensitivity. Basic research on toxicokinetics is needed for most of
the HANs; only DBAN (NTP, 2002) and DCAN have been studied in detail (Roby et al., 1986),
and only the study for DCAN was available for review at the time this document was prepared.
In particular, further understanding of HAN metabolism is needed. Research in this area could
clarify the relative contribution of glutathione conjugation and oxidative metabolism pathways to

EPA/OW/OST/HECD

VTII-45

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
the observed spectrum of toxicity for HANs. Identification of the toxic moiety and the GST or

CYP isoforms important in catalyzing HAN metabolism could contribute to characterization of

potential human susceptibility based on age, gender, or genetic predisposition.

Based on the clear limitations in the database and gaps in understanding of the mechanisms
of toxicity for HANs, the derived RfD and HA values are best characterized as low in confidence.

EPA/OW/OST/HECD

VTII-46

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Chapter IX. References

Abdel-Aziz, A.A., S.Z. Abdel-Rahman, A.M. Nouraldeen, S.A. Shouman, J.P. Loh and A.E.
Ahmed. 1993. Effect of glutathione modulation on molecular interaction of
[14C]-chloroacetonitrile with maternal and fetal DNA in mice. Reprod Toxicol 7(3): 263-272.

Ahmed, A.E. and M.Y. Farooqui. 1982. Comparative toxicities of aliphatic nitriles. Toxicol Lett
12(2-3): 157-163.

Ahmed, A.E., S.A. Soliman, J.P. Loh and G.I. Hussein. 1989. Studies on the mechanism of
haloacetonitriles toxicity: inhibition of rat hepatic glutathione S-transferases in vitro. Toxicol
Appl Pharmacol 100(2): 271-279.

Ahmed, A.E., G.I. Hussein, J.P. Loh and S.Z. Abdel-Rahman. 1991. Studies on the mechanism
of haloacetonitrile-induced gastrointestinal toxicity: interaction of dibromoacetonitrile with
glutathione and glutathione-S-transferase in rats. J Biochem Toxicol 6(2): 115-121.

Ahmed, A.E., J. Aronson and S. Jacob. 2000. Induction of oxidative stress and TNF-alpha
secretion by dichloroacetonitrile, a water disinfectant by-product, as possible mediators of
apoptosis or necrosis in a murine macrophage cell line (RAW). Toxicol In Vitro 14(3): 199-210.

Arora, H., M.W. LeChavallier, and K.L. Dixon. 1997. DBP Occurrence Survey. J. AWWA
89(6): 60-68.

AWWARF. Disinfection by-products database and model project. American Water Works
Association Research Foundation. Denver, Colorado, 1991. (Cited in WHO, 2000)

Bieber TI, Trehy ML. Dihaloacetonitriles in chlorinated natural waters. In: Jolley RJ, Brungs
WA, Cotruvo JA, Cumming RB, Mattice JS, Jacobs VA, eds. Water Chlorination: Environmental
Impact and Health Effects, Vol. 4, Book 1: Chemistry and Water Treatment. Ann Arbor, MI:
Ann Arbor Science Publisher, Inc., 1983, pp. 85-96. (Cited in WHO, 2000)

Boorman, G.A., V. Dellarco, J.K. Dunnick, R.E. Chapin, S. Hunter, F. Hauchman, H. Gardner,
M. Cox and R.C. Sills. 1999. Drinking water disinfection byproducts: Review and approach to
toxicity evaluation. Environ. Health Perspect. 107(Suppl. 1): 207-217.

Budavari S, O'Neill M, Smith A, eds. The Merck index. An encyclopedia of chemicals, drugs,
and biologicals, 1989, 11th ed. Rahway, NJ, Merck. (Cited in WHO, 2000)

Bull, R.J. and M. Robinson. 1985. Carcinogenic Activity of Haloacetonitrile and Haloacetone
Derivatives in the Mouse Skin and Lung. In: Jolley, R.L., R.J. Bull, W.P. Davis, S. Katz, M.H.
Roberts and V. A. Jacobs, eds. Water Chlorination: Chemistry, Environmental Impact and Health
Effects, Vol 5. Chelsea, MI: Lewis Publishers, Inc., pp. 221-227.

EPA/OW/OST/HECD

IX-1

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Bull, R.J., J.R. Meier, M. Robinson, H.P. Ringhand, R.D. Laurie and J.A. Stober. 1985.
Evaluation of mutagenic and carcinogenic properties of brominated and chlorinated acetonitriles:
by-products of chlorination. Fundam Appl Toxicol 5: 1065-1074.

Bull, R.J., J.M. Brown, E.A. Meierhenry, T.A. Jorgenson, M. Robinson and J.A. Stober. 1986.
Enhancement of the hepatotoxicity of chloroform in B6C3F1 mice by corn oil: Implications for
chloroform carcinogenesis. Environ Health Perspect 69:49-58.

Caravati, E.M. and T.L. Litovitz. 1988. Pediatric cyanide intoxication and death from an
acetonitrile-containing cosmetic. JAMA 260(23): 3470-3473.

Christ, S.A., E.J. Read, J.A. Stober and M.K. Smith. 1995. The developmental toxicity of
bromochloroacetonitrile in pregnant Long-Evans rats. Int J Environ Health Res 5(2): 175-88.

Christ, S.A., E.J. Read, J.A. Stober and M.K. Smith. 1996. Developmental effects of
trichloroacetonitrile administered in corn oil to pregnant Long-Evans rats. J Toxicol Environ
Health 47(3): 233-47.

Daniel, F.B., K.M. Schenck, J.K. Mattox, E.L.C. Lin, D.L. Haas and M.A. Pereira. 1986.
Genotoxic properties of haloacetonitriles: drinking water by-products of chlorine disinfection.
Fund Appl Toxicol 6: 447-453.

Dix, K.J., G.L. Kedderis and S.J. Borghoff. 1997. Vehicle-dependent oral absorption and target
tissue dosimetry of chloroform in male rats and female mice. Toxicol Lett 91(3): 197-209.

Dourson, M.L. 1994. Methods for establishing oral reference doses (RfDs). In: Merts, W., C.O.
Abernathy and S.S. Olin, eds. Risk assessment of essential elements. Washington, DC: ILSI
Press, pp. 51-61.

DOW Chemical Company. 1992a. Initial submission: acute toxicity evaluation of
dibromoacetonitrile in rats (with cover letter dated 050792). Washington, DC: EPA-OPPTS
Document # 88-920002501.

DOW Chemical Company. 1992b. Initial submission: toxicity report: dibromoacetonitrile (with
cover letter dated 082692). Washington, DC: EPA-OPPTS Document # 88-920009101.

Eastman Kodak Company. 1992. Initial submission: toxicity report: dibromoacetonitrile with
cover letters dated 082692. Submitted to: Office of Pollution Prevention and Toxics, U.S.
Environmental Protection Agency. Document No. OTS0546384.

Farooqui, M.Y., B. Ybarra, J. Piper and A. Tamez. 1995. Effect of dosing vehicle on the toxicity
and metabolism of unsaturated aliphatic nitriles. J Appl Toxicol 15(5): 411-420.

Gee, P., C.H. Sommers, A.S. Melick, X.M. Gidrol, M.D. Todd, R.B. Burris, M.E. Nelson, R.C.
Klemm and E. Zeiger. 1998. Comparison of responses of base-specific salmonella tester strains

EPA/OW/OST/HECD

IX-2

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

with the traditional strains for identifying mutagens: The results of a validation study. Mutat Res
412(2): 115-30.

Gordon, D.A., T.K. Wessendarp, W. Crocker, M.K. Smith and A.C. Roth. 1991. Comparative
absorption and distribution of radiolabeled trichloroacetonitrile (TCAN) in pregnant rats from
corn oil (CO) and tricaprylin (TCAP) vehicles. Teratology 43(5): 427.

Geller, R.J., B.R. Ekins and R.C. Iknoian. 1991. Cyanide toxicity from acetonitrile-containing
false nail remover. AmJEmergMed 9(3): 268-270.

Hayes, J.R., L.W. Condie and J.F. Borzelleca. 1986. Toxicology of haloacetonitriles. Environ
Health Perspect 69: 183-202.

Hechenbleikner, I. 1946. American Cyanamid. Substitution on halogenated acetonitriles. Chem
Abstr 40: 2927.

Herren-Freund, S.L. and M.A. Pereira. 1986. Carcinogenicity of by-products of disinfection in
mouse and rat liver. Environ Health Perspect 69: 59-65.

IARC. 1999. International Agency for Research of Cancer. IARC monographs on the
evaluation of the carcinogenic risk of chemicals to humans: Re-evaluation of some organic
chemicals, hydrazine and hydrogen peroxide (part three). Lyon, France: World Health
Organization.

Jacangelo, J.G., N.L. Patania, K.M. Reagan, E.M. Aieta, S.W. Krasner and M.J. McGuire.
1989. Ozonation: assessing its role in the formation and control of disinfection by-products. J
Am Water Works Assn. 81: 74-84.

Kier, L.E., D.J. Brusick, A.E. Auletta, E.S. Von Halle, M.M. Brown, V.F. Simmon, V. Dunkel, J.
McCann, K. Mortelmans, L.E. Kier LE, et al. 1986. The Salmonella tryhimurium/mammalian
microsomal assay: A report of the U.S. Environmental Protection Agency Gene-Tox Program.
Mutat Res 168: 69-240.

Klotz, J.B. and L.A. Pyrch. 1999. Neural tube defects and drinking water disinfection
by-products. Epidemiology 10(4): 383-390.

Krasner, S.W., M.J. McGuire, J.G. Jacangelo, N.L. Patania, K.M. Reagan and E.M. Aieta. 1989.
The occurrence of disinfection by-products in U.S. drinking water. J. Am. Water Works Assn.
81: 41-53.

Kurt, T.L., L.C. Day, W.G. Reed and W. Gandy. 1991. Cyanide poisoning from glue-on nail
remover. AmJEmergMed 9(3): 271-272.

EPA/OW/OST/HECD

IX-3

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Larson, J.L., D.C. Wolf and B.E. Butterworth. 1994. Induced cytotoxicity and cell proliferation
in the hepatocarcinogenicity of chloroform in female B6C3F1 mice: Comparison of administration
by gavage in corn oil vs ad libitum in drinking water. Fundam Appl Toxicol 22(1): 90-102.

LeCurieux, F., S. Giller, L. Gauthier, F. Erb and D. Marzin. 1995. Study of the genotoxic
activity of six halogenated acetonitriles, using the Sos chromotest, the Ames-fluctuation test and
the Newt micronucleus test. MutatRes 341(4): 289-302.

Lide, D.R., ed. 1992. CRC Handbook of Chemistry and Physics. 73rd Edition. Boca Raton:
CRC Press, Inc.

Lilly, P.D., J.E. Simmons and R.A. Pegram. 1994. Dose-dependent vehicle differences in the
acute toxicity of bromodichloromethane. Fundam Appl Toxicol 23(1): 132-140.

Lilly, P.D., M.E. Andersen, T.M. Ross and R.A. Pegram. 1998. A physiologically based
pharmacokinetic description of the oral uptake, tissue dosimetry, and rates of metabolism of
bromodichloromethane in the male rat. Toxicol Appl Pharmacol 150(2): 205-217.

Lin, E.L.C., F.B. Daniel, S.L. Herren-Freund and M.A. Pereira. 1986. Haloacetonitriles:
metabolism, genotoxicity, and tumor-initiating activity. Environ Health Perspect 69: 67-71.

Lin, E.L. and C.W. Guion. 1989. Interaction of haloacetonitriles with glutathione and
glutathione-S-transferase. Biochem Pharmacol 38(4): 685-688.

Lin, E.L.C., T.V. Reddy and F.B. Daniel. 1992. Macromolecular adduction by
trichloroacetonitrile in the Fischer 344 rat following oral gavage. Cancer Lett 62(1): 1-9.

Lykins, Jr., B.W., W.E. Koffskey, and K.S. Patterson. 1994. Alternative disinfectants for
drinking water treatment. J. Environ. Eng. 120(4):745-758.

Meier, J.R., R.J. Bull, J.A. Stober and M.C. Cimino. 1985. Evaluation of chemicals used for
drinking water disinfection for production of chromosomal damage and sperm-head abnormalities
in mice. Env Mutag 7: 201-211.

Mills, C.J., R.J. Bull, K.P. Cantor, J. Reif, S.E. Hrudey and P. Huston. 1998. Workshop report.
Health risks of drinking water chlorination by-products: Report of an expert working group.
Chronic Dis Can 19(3): 91-102.

Miltner, R.J., E.W. Rice and A. A. Stevens. 1990. Pilot-scale investigation of the formation and
control of disinfection byproducts. In: 1990 Annual Conference Proceedings. AWWA Annual
Conference, Cincinnati, OH. 2:1787-1802.

Mink, F.L., W.E. Colman, J.W. Munch, W.H. Kaylor and H.P. Ringhand. 1983. In vivo
formation of halogenated reaction products following peroral sodium hypochlorite. Bull Environ
Contam Toxicol 30: 394-399.

EPA/OW/OST/HECD

IX-4

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Mohamadin, A.M. and A.B. Abdel-Naim. 1999. Chloroacetonitrile-induced toxicity and
oxidative stress in rat gastric epithelial cells. Pharmacol Res 40(4): 377-383.

Moudgal, C.J., J.C. Lipscomb and R.M. Bruce. 2000. Potential health effects of drinking water
disinfection by-products using quantitative structure toxicity relationship. Toxicology 147(2):
109-131.

Narotsky, M.G., R.A. Pegram and R.J. Kavlock. 1997. Effect of dosing vehicle on the
developmental toxicity of bromodichloromethane and carbon tetrachloride in rats. Fundam Appl
Toxicol 40(1): 30-36.

NAS. 1977. National Academy of Sciences. Drinking water and health. Washington, DC.

NAS. 1980. National Academy of Sciences. Drinking water and health, Vol. 3. Washington, DC.

Nieuwenhuijsen, M.J., M.B. Toledano, N.E. Eaton, J. Fawell and P. Elliott. 2000. Chlorination
disinfection byproducts in water and their association with adverse reproductive outcomes: A
review. Occup Environ Med 57(2): 73-85.

Nouraldeen, A.M. and A.E. Ahmed. 1996. Studies on the mechanisms of
halocentronitrile-induced genotoxicity IV: In vitro interaction of haloacetonitriles with DNA.
Toxicol in Vitro 10:17-26.

NTP. 2000. National Toxicology Program. Safe Drinking Water Program Factsheet. National
Institute of Environmental Health Sciences, National Institutes of Health.

NTP. 2002. National Toxicology Program. Abstracts from subacute and subchronic studies of
Dibromoacetonitrile (DBAN) on Fischer-344 rats and B6C3F1 mice, and update on test status.
Research Triangle Park, NC. Available at http://ntp-server.niehs.nih.gov/

Oliver, B.G. 1983. Dihaloacetonitriles in drinking water: algae and fulvic acid as precursors.
Environ Sci Technol 17: 80-83.

Oliver, B.G. Dihaloacetonitriles in drinking water: algae and fulvic acids as precursors.
Environmental science technology, 1983, 15: 1075-1080.

O'Neil, M.J., A. Smith, P.E. Heckelman (eds.). 2001. The Merck Index: An Encyclopedia of
Chemicals, Drugs, and Biologicals, 13th Edition. Merck & Co., Inc. Whitehouse Station, NJ.

Osgood, C. and D. Sterling. 1991. Dichloroacetonitrile, a by-product of water chlorination,
induces aneuploidy in Drosophila. Mutat Res 261(2): 85-91.

Pereira, M.A., C.L. Luan-Ho and J.K. Mattox. 1984. Haloacetonitrile excretion as thiocyanate
and inhibition of dimethylnitrosamine demethylase: A proposed metabolic scheme. J Tox Environ
Health 13: 633-641.

EPA/OW/OST/HECD

IX-5

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

R.O.W. Sciences, Inc. 1997. Final report on the reproductive toxicity of dibromoacetonitrite
(CAS No. 3252-43-5) administered in diet to SD rats. NTIS/PB97-143127

Raymond, P. and G.L. Plaa. 1997. Effect of dosing vehicle on the hepatotoxicity of CC14 and
nephrotoxicity of CHC13 in rats. J Toxicol Environ Health 51(5): 463-476.

Reckhow, D.A. and P.C. Singer. 1990. Chlorination by-products in drinking water: from
formation potentials to finished water concentrations. J. AWWA 82(4): 173-180.

Reckhow, D.A., P.C. Singer and R.L. Malcolm. 1990. Chlorination of humic materials: byproduct
formation and chemical interpretations. Environ. Science and Technology 24(11): 1655-1664.

Richardson, S.D. 1998. Identification of drinking water disinfection byproducts. In: John
Wiley's Encyclopedia of Environmental Analysis & Remediation, R.A. Meyers Ed. 3:1398-1421.

Roby MR, Carle S, Pereira MA, Carter DE. Excretion and tissue disposition of
dichloroacetonitrile in rats and mice. Environmental health perspectives, 1986, 69:215-220.

Roth, A.C., D.A. Gordon and M.K. Smith. 1990. Absorption and distribution of radiolabeled
trichloroacetonitrile (TCAN) from two vehicles. Teratology. 41(5): 588.

Saillenfait, A.M. and J.P. Sabate. 2000. Comparative developmental toxicities of aliphatic
nitriles: in vivo and in vitro observations. Toxicol Appl Pharmacol 163(2): 149-163.

Siddiqui MS, Amy GL. Factors affecting DBP formation during ozone-bromide reactions.

Journal of the American water works association, 1993, 85(1): 63-72. (Cited in WHO, 2000)

Silver, E.H., S.H. Kuttab, T. Hasan and M. Hassan. 1982. Structural considerations in the
metabolism of nitriles to cyanide in vivo. Drug Metab Dispos 10(5): 495-498.

Simmon, V.F., K. Kauhanen and R.G. Tardiff. 1977. Mutagenic activity of chemical identified in
drinking water. Dev Toxicol Environ Sci 2: 249-258.

Simmon, V.F. and K. Kauhanen. 1978. In vitro microbiological mutagenicity assays of
dichloroacetonitrile. Report LSU-5612. 16 pp. (As cited in Kier et al., 1986).

Smith, M.K., H. Zenick and E.L. George. 1986. Reproductive toxicology of disinfection
by-products. Environ Health Perspect 69: 177-182.

Smith, M.K., E.L. George, H. Zenick, J.M. Manson and J.A. Stober. 1987. Developmental
toxicity of halogenated acetonitriles: Drinking water by-products of chlorine disinfection.
Toxicology 46: 83-93.

Smith, M.K., J.L. Randall, L.D. Ford, D.R. Tocco and R.G. York. 1988. Teratogenic effects of
trichloroacetonitrile in the Long-Evans rat. Teratology. 38: 113-120.

EPA/OW/OST/HECD

IX-6

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Smith, M.K., J.L. Randall, J.A. Stober and E.J. Read. 1989. Developmental toxicity of
dichloroacetonitrile: A by-product of drinking water disinfection. Fundam Appl Toxicol 12(4):
765-772.

Smyth, H.F., C.P. Carpenter, C.S. Weil, U.C. Pozzani and J.A. Striegel. 1962. Range-finding
toxicity data: List VI. AIHAJ 23:95-107.

U.S. EPA. 1986. U.S. Environmental Protection Agency. Guidelines for carcinogen risk
assessment. Fed Reg 51(185): 33992-34003.

U.S. EPA. 1987. Rough Final Draft for the Drinking Water Criteria Document on
Haloacetonitriles, Chloropicrin and Cyanogen Chloride. Prepared by ICAIR, Life Systems, Inc.,
for Criteria and Standards Division, Office of Drinking Water, Washington, DC, under Contract
68-03-3279.

U.S. EPA. 1991. U.S. Environmental Protection Agency. Guidelines for developmental toxicity
risk assessment. Fed Reg 56(234): 63798-63826.

U.S. EPA. 1993. U.S. Environmental Protection Agency. Reference dose (RfD): Description
and use in health risk assessments. Integrated Risk Information System (IRIS). Online.
Intra-Agency Reference Dose (RfD. Work Group, Office of Health and Environmental
Assessment, Environmental Criteria and Assessment Office. Cincinnati, OH.

U.S. EPA. 1994. Final Draft for the Drinking Water Criteria Document on Chlorinated
Acids/Aldehydes/Ketones/Alcohols. Prepared for Health and Ecological Criteria Division Office
of Science and Technology, Office of Water, U.S. Environmental Protection Agency,

Washington, DE 20460. EPA 68-C2-0139.

U.S. EPA. 1999. U.S. Environmental Protection Agency. Guidelines for Carcinogen Risk
Assessment. NCEA-F-0644. Washington, DC.

U.S. EPA. 2000a. U.S. Environmental Protection Agency. Stage 2 Occurrence and Exposure
Assessment for Disinfectants and Disinfection Byproducts (D/DBPs) in Public Drinking Water
Systems. Washington, DC: Office of Ground Water and Drinking Water.

U.S. EPA. 2000b. U.S. Environmental Protection Agency. ICR Data Analysis Plan.

Washington, DC: Office of Water.

U.S. EPA. 2000c. U.S. Environmental Protection Agency. Benchmark dose technical guidance
document. External review draft. Washington, DC: Risk Assessment Forum.

U.S. EPA. 2000d. U.S. Environmental Protection Agency. Methodology for Deriving Ambient
Water Quality Criteria for the Protection of Human Health. Washington, DC: Office of Water,
Office of Science and Technology.

EPA/OW/OST/HECD

IX-7

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

U.S. EPA. 2001. U.S. Environmental Protection Agency. Help manual for Benchmark dose
software version 1.3. Washington, DC: Office of Research and Development. EPA
600/R-00/014F.

U.S. EPA. 2002a. U.S. Environmental Protection Agency. Information collection rule (ICR) -
technical working group data analysis website. Available online at
http://www.cadmusonline.net/twg/mainmenu.htm

U.S. EPA. 2002b. Information Collection Rule (ICR) database. Available online at
http://www.cadmusonline.net/twg/ Downloaded June, 2002.

U.S. EPA. 2002c. Draft Drinking Water Criteria Document for Cyanogen Chloride and Its
Potential Metabolites. Washington, DC: Office of Water, Office of Science and Technology.

WHO. 2000. World Health Organization. Environmental Health Criteria: 216 Disinfectants and
Disinfectant By-products. World Health Organization, Geneva. International Programme on
Chemical Safety (IPCS).

Withey, J.R., B.T. Collins and P.G. Collins. 1983. Effect of vehicle on the pharmacokinetics and
uptake of four halogenated hydrocarbons from the gastrointestinal tract of the rat. J Appl Toxicol
3(5): 249-253.

Zimmermann, F.K., R.C. von Borstel, E.S. von Halle, J.M. Parry, D. Siebert, G. Zetterberg, R.
Barale and N. Loprieno. 1984. Testing of chemicals for genetic activity with Saccharomyces
cerevisiae: a report of the U.S. Environmental Protection Agency Gene-Tox Program. Mutat Res
133: 199-244.

Zimmermann, F.K. and A. Mohr. 1992. Formaldehyde, glyoxal, urethane, methyl carbamate,
2,3-butanedione, 2,3-hexanedione, ethyl acrylate, dibromoacetonitrile and 2-hydroxypropionitrile
induce chromosome loss in Saccharomyces cerevisiae. Mutat Res 270(2): 151-66.

EPA/OW/OST/HECD

IX-8

Final Draft

-------
Drinking Water Criteria Document for Cyanogen Chloride and Potential Metabolites

Chapter X. References

Abajo, P., C. Feal, T. Sanz-Sanchez, J. Sanchez-Perez and A. Garcia-Diez. 1999. Eczematous
erythroderma induced by cyanamide. Contact Dermatitis. 40(3): 160-1.

ACGIH. 1996. American Council of Governmental Industrial Hygienists. Documentation of
threshold limit values and biological exposure indices. Hydrogen cyanide and cyanide salts.

ACGIH. 1991. American Council of Governmental Industrial Hygienists. Documentation of
threshold limit values and biological exposure indices. Cyanogen chloride.

Aitken, D., D. West, F. Smith, W. Poznanski, J. Cowan, J. Hurtig, E. Peterson and B. Benoit.
1977. Cyanide toxicity following nitroprusside-induced hypotension. Can Anaesth Soc J 24:651-
660. (As cited in ATSDR, 1997.)

Ajima, M., K. Usuki, A. Igarashi, R. Okazaki, K. Hamano, A. Urabe and Y. Totsuka. 1997.
Cyanamide-induced granulocytopenia. Internal Med 36(9): 640-2.

Albert, R.E., A.R. Sellakumar, S. Laskin, M. Kuschner, N. Nelson and C.A. Snyder. 1982.
Gaseous formaldehyde and hydrogen chloride induction of nasal cancer in rats. J Natl Cancer Inst
68(4): 597-603.

Aldridge, W.N. and C.L. Evans. 1946. The physiological effects and fate of cyanogen chloride.
Q J Exp Physiol 33: 241-266.

Aldridge, W.N. 1951. The conversion of cyanogen chloride to cyanide in the presence of blood
proteins and sulphydryl compounds. Biochem J 48: 271-276.

Aletor, V.A. 1993. Cyanide in garri. 1. Distribution of total, bound and free hydrocyanic acid in
commercial garri, and the effect of fermentation time on residual cyanide content. Int J Food Sci
Nutr 44(4):281-287.

Ali, F., and T.N. Persaud. 1987. Mechanisms of alcohol embryopathy: Role of acetaldehyde.
Teratology 36(1): 20A. (abstract)

Anderson, R.C. and K.K. Chen. 1940. Absorption and toxicity of sodium and potassium
thiocyanates. J Am Pharm Assoc 29: 152-161.

Andrews, J.M., E.S. Sweeney, T.C. Grey and T. Wetzel. 1989. The biohazard potential of
cyanide poisoning during postmortem examination. J Foresic Sci 34:1280-1284. (As cited in
ATSDR, 1997.)

Anonymous. 1990. Health assessment for ALSCO Anaconda National priorities list. NTIS
BP90-100330. (As cited in ATSDR, 1997.)

EPA/OW/OST/HECD

X-l

Final draft

-------
Drinking Water Criteria Document for Cyanogen Chloride and Potential Metabolites

ATSDR. 1997. Toxicological Profile for Cyanide (Update). U.S. Department of Health and
Human Services, Public Health Service. September.

Ayman, D. 1931. Potassium thiocyanate in the treatment of essential hypertension. J Am Med
Assoc 96: 1852-1857.

Bala, T.S., M.K. Janardanasarma and M. Raghunath. 1996. Dietary goitrogen-induced changes
in the transport of 2-deoxy-D-glucose and amino acids across the rat blood-brain barrier. Int J
DevNeurosci 14(5): 575-83.

Ballantyne, B. 1988. Toxicology and hazard evaluation of cyanide fumigation powders. Clin
Toxicol 26:325-335 (As cited in ATSDR, 1997)

Banerjee, K.K., P. Marimuthu, P. Bhattacharyya and M. Chatterjee. 1997. Effect of thiocyanate
ingestion through milk on thyroid hormone homeostasis in women. BrJNutr 78(5): 679-81.

Barker, M.H. 1936. The blood cyanates in the treatment of hypertension. J Am Med Assoc
106(10): 762-767.

Barker, M.H., H. A. Lindberg and M.H. Wald. 1941. Further experiences with thiocyanates. J
AmMedAssoc 117(19): 1591-1594.

Barrow, C.S., Y. Alarie, J.C. Warrick and M.F. Stock. 1977. Comparison of the sensory
irritation response in mice to chlorine and hydrogen chloride. Arch Environ Health 32(2): 68-76.

Basu, T.K. 1983. High-dose ascorbic acid decreases detoxification of cyanide derived from
amygdalin (laetrile): Studies in guinea pigs. Can J Physiol Pharmacol 61:1426-1430.

Beamish, R.E., W.F. Perry and V.M. Storrie. 1954. Observations on thyroid function in
hypertensive patients treated with potassium thiocyanate. Am Heart J 48: 433-438.

Beaumont, J.J., J. Leveton, K. Knox, T. Bloom, T. McQuiston, M. Young, R. Goldsmith, N.K.
Steenland, D.P. Brown and W.E. Halperin. 1987. Lung cancer mortality in workers exposed to
sulfuric acid mist and other acid mists. J Nat Cancer Inst 79(5): 911-921.

Bhattacharya, R., P. Kumar and A.S. Sachan. 1994. Cyanide induced changes in dynamic
pulmonary mechanics in rats. Indian J Physiol Pharmacol 38(4): 281-4.

Blackburn, C.M., F.R. Keating and S.F. Haines. 1951. Radioiodine tracer studies in thiocyanate
myxedema. J Clin Endocrinol 11: 1503-1511.

Blanc, P., M. Hogan, K. Malin, D. Hryhorczuk, S. Hessl and B. Bernard. 1985. Cyanide
Intoxication Among Silver-Reclaiming Workers. JAMA 253(3): 367-371.

EPA/OW/OST/HECD

X-2

Final draft

-------
Drinking Water Criteria Document for Cyanogen Chloride and Potential Metabolites

Bond, G.G., G.H. Flores, B.A. Stafford and G.W. Olsen. 1991. Lung cancer and hydrogen
chloride exposure: results from a nested case-control study of chemical workers. J Occup Med
33(9): 958-961.

Bookallil, M.J. 2001. PH of the Blood: Acid-Base Balance. The University of Sydney.
http://www.usyd.edu.au/su/anaes/lectures/acidbase_mjb/frameversion.html

Boorman, G.A. 1998. Memorandum to Dr. Stokes, Chairperson, Protocol Review Committee
regarding cyanogen chloride studies. April 24, 1998.

Bottoms, S.F., B.R. Kuhnert, P.M. Kuhnert and A.L. Reese. 1982. Maternal passive smoking
and fetal serum thiocyanate levels. Am J Ob stet Gynecol 144(7):787-791. (As cited in AT SDR,
1997.)

Boulos, B.M., F. Hanna, L.E. Davis and A.H. Almond. 1973. Placental transfer of antipyrine and
thiocyanate and their use in determining maternal and fetal body fluids in a maintained pregnancy.
Arch Int Pharm Acodyn Ther 201: 42-51.

Buckley, L.A., X.Z. Jiang, R.A. James, K.T. Morgan and C.S. Barrow. 1984. Respiratory tract
lesions induced by sensory irritants at the RD50 concentration. Toxicol Appl Pharmacol 74:
417-429.

Bujan, J.J.G., J.Y. Bayona, and R.S. Arechavala. 1994. Allergic contact dermatitis from
cyanamide: Report of 3 cases. Contact Dermatitis 31: 331-332.

Burleigh-Flayer, H., K.L. Wong and Y. Alarie. 1985. Evaluation of the pulmonary effects of
hydrochloric-acid using carbon dioxide challenges in guinea-pigs. Fund Appl Toxicol 5(5): 978-
985.

Callahan, M.A., M.W. Slimak, N.W. Gabel, et al. 1979. Water-related environmental fate of 129
priority pollutants. Vol. 1. U.S. Environmental Protection Agency, Office of Water Planning and
Standards, Office of Water and Waste Management, Washington, DC. (As cited in ATSDR,
1997.)

Carella, F., M.P. Grassi, M. Savoiardo, P. Contri, B. Rapuzzi and A. Mangoni. 1988. Dystonic-
Parkinsonian Syndrome After Cyanide Poisoning: Clinical and MRI Findings. J Neurol
Neurosurg Psychiatry 51:1345-1348.

Chan, J.C. 1983. Acid-base disorders and the kidney. AdvPediatr 30:401-71.

Chan, S. and M.D. Kilby. 2000. Thyroid hormone and central nervous system development. J
Endocrin 165:1-8

Chandra, H., B.N. Gupta and N. Mathur. 1988. Threshold limit value of cyanide: A reappraisal
in Indian context. Indian J Environ Protection 8:170-174. (As cited in ATSDR, 1997.)

EPA/OW/OST/HECD

X-3

Final draft

-------
Drinking Water Criteria Document for Cyanogen Chloride and Potential Metabolites

Chandra, H., B.N. Gupta, S.K. Bhargava, S.H. Clerk, and P.N. Mahendra. 1980. Chronic
cyanide exposure - A biochemical and industrial hygiene study. J Analyt Toxicol 4:161-165.

Charache, S., T.P. Duffy, N. Jander, J.C. Scott, M. Bedine and R. Morrell. 1975. Toxic-
therapeutic ratio of sodium cyanate. Arch Internal Med 135(8): 1043-7.

Chemfinder. 2001. Onlineatwww.chemfinder.com.

Chen, Y., L.L. Pederson and N.M. Lefcoe. 1990. Exposure to environmental tobacco smoke
(ETS) and serum thiocyanate level in infants. Arch Environ Health 45(3): 163-167. (As cited in
AT SDR, 1997.)

Cicerone, R.J. and Zellner R. 1983. The atmospheric chemistry of hydrogen cyanide (HCN). J
GeophysRes 88:10689-10696. (As cited in AT SDR, 1997.)

Cipollaro, M., G. Corsale, A. Esposito, E. Ragucci, N. Staiano, G.G. Giordano and G. Pagano.
1986. Sublethal pH decrease may cause genetic damage to eukaryotic cell: a study on sea urchins
and Salmonella typhimurium. Teratogen Carcinogen Mutagen 6(4): 275-87.

Clausing, P. and M. Gottschalk. 1989. Effects of drinking water acidification, restriction of
water supply and individual caging on parameters of toxicological studies in rats. Zeitschrift Fur
Versuchstierkunde 32(3): 129-34.

Cliff, J., D. Nicala, F. Saute, R. Givragy, G. Azambuja, A. Taela, L. Chavane, and J. Howarth.
1997. Konzo associated with war in Mozambique. Trop Med Int Health 2(11): 1068-1074.

Cole, R.H., R.E. Frederick, R.P. Healy et al. 1984. Preliminary findings of the priority pollutant
monitoring project of the nationwide urban runoff program. J Water Pollut Control Fed 56:898-
908. (As cited in ATSDR, 1997.)

Conde-Salazar, L., D. Guimaraens, L. Romero and A. Harto. 1981. Allergic contact dermatitis
to cyanamide (carbodiimide). Contact Dermatitis 7(6): 329-330.

Crutzen, P.J. and G.R. Carmichael. 1993. Modeling the influence of fires on atmospheric
chemistry. In: Crutzen, P.J. and J.G. Goldammer, eds. Fire in the Environment: The Ecological,
Atmospheric and Climatic Importance of Vegetation Fires. John Wiley and Sons, LTD. pp. 89-
105. (As cited in ATSDR, 1997.)

Dahlberg, P.A., A. Bergmark, L. Bjorck, A. Bruce, L. Hambraeus and O. Claesson. 1984. Intake
of thiocyanate by way of milk and its possible effect on thyroid function. Am J Clin Nutr 39(3):
416-20.

Darmer, K.I., E.R. Kinkead and L.C. Dipasquale. 1974. Acute toxicity in rats and mice exposed
to hydrogen chloride gas and aerosols. Am Ind Hyg Assoc J 35(10): 623-631.

EPA/OW/OST/HECD

X-4

Final draft

-------
Drinking Water Criteria Document for Cyanogen Chloride and Potential Metabolites

De Flora, S. 1981. Study of 106 organic and inorganic compounds in the Salmonella/microsome
test. Carcinogenesis 2:283-298. (As cited in ATSDR, 1997)

De Flora, S., A. Camoirano, P. Zanacchi and C. Bennicelli. 1984. Mutagenicity testing with
TA97 and TA102 of 30 DNA-damaging compounds, negative with other Salmonella strains.
Mutat Res. 134:159-165. (As cited in ATSDR, 1997).

De Groot, A.P., M.I. Willems and R.H. de Vos. 1991. Effects of high levels of brussels sprouts
in the diet of rats. Food Chem Toxicol 29(12): 829-37.

Deitrich, RA., P.A. Troxell, W.S. Worth, and V.G. Erwin. 1976. Inhibition of aldehyde
dehydrogenase in brain and liver by cyanamide. Biochem Pharmacol 25:2733-2737. (As cited in
Mertschenk et al., 1991).

Deschamps, D., P. Soler, N. Rosenberg, F. Baud and P. Gervais. 1994. Persistent asthma after
inhalation of a mixture of sodium hypochlorite and hydrochloric acid. Chest 105(6): 1895-6.

Devlin, D. J., J.W. Mills and R.P. Smith. 1989a. Histochemical Localization of Rhodanese
Activity in Rat Liver and Skeletal Muscle. Toxicol Appl Pharmacol 97:247-255.

Devlin, D.J., R.P. Smith and C.D. Thron. 1989b. Cyanide metabolism in the isolated, perfused,
bloodless hindlimbs or liver of the rat. Toxicol Appl Pharmacol 98(2): 338-49.

Domzalski, C.A., L.C. Kolb and E.A. Hines. 1953. Delerious reactions secondary to thiocyanate
therapy of hypertension. Staff Meetings of the Mayo Clinic 28(9): 272-280.

Dourson, M.L. 1994. Methods for establishing oral reference doses (RfDs). In: Merts, W., C.O.
Abernathy and S.S. Olin, eds. Risk assessment of essential elements. Washington, DC: ILSI Press,
pp. 51-61.

Drawbaugh, R. B. and T. C. Marrs. 1987. Interspecies differences activity in liver, kidney and
plasma. Comp Biochem Physiol 85B(2): 307-310.

Drinker. P. 1932. Hydrocyanic acid gas poisoning by absorption through the skin. J Ind Hyg
14: 1-2.

Edwards, J.O., T.E. Erstfeld, K.M. Ibne-Rasa, G. Levey, and M. Moyer. 1986. Reaction rates
for nucleophiles with cyanogen chloride; Comparison with two other digonal carbon compounds.
Int J Chem Kinetics 18:165-180.

Einhorn, I.N. and G. Moore. 1990. Utilization of neurophysiological protocols to characterize
soldier response to irritant gases. Phase 1. Northeast Research Institute, Inc. p.91
NTIS/AD-B160773.

EPA/OW/OST/HECD

X-5

Final draft

-------
Drinking Water Criteria Document for Cyanogen Chloride and Potential Metabolites

El-Ghawabi, S.H., M.A. Gaafar, A.A. El-Saharti, S.H. Ahmed, K.K. Malash and R. Fares. 1975.
Chronic Cyanide Exposure: A Clinical, Radioisotope and Laboratory Study. Br J Ind Med
32:215-219.

Ermans, A.M., F. Delange, M. Vander Velden and J. Kinthaert. 1980. Possible Role of Cyanide
and Thiocyanate in The Etiology of Endemic Cretinism. Adv Exp Med Biol 30: 455-486.

Ferguson, H.C. 1962. Dilution of dose and acute oral toxicity. Toxicol Appl Pharmacol.
4:759-762. (as cited in AT SDR, 1997)

Fiksel, J., C. Cooper, A. Eschenroeder, et al. 1981. Exposure and risk assessment for cyanide.
EPA/440/4-85/008. NTIS PB85-220572. (As cited in AT SDR, 1997.)

Flury, F. and F. Zernik. 1931. Noxious gases: Vapors, mist, smoke and dust particles. With
authorized use of the work Noxious gases by Henderson and Haggard. Berlin: Publishing house
of Julius Springer, pp. 340-351.

GEOMET Technologies, Inc. 1981. Hydrogen chloride: Report 4, occupational hazard
assessment. U.S. Department of Health and Human Services, NIOSH, Cincinnati, OH. NTIS
PB83-105296.

Goday-Bujan, J.J., I. Yanguas-Bayona and R. Soloeta-Arechavala. 1994. Allergic contact
dermatitis from cyanamide: Report of 3 cases. Contact Dermatitis 31(5): 331-2.

Goldring, W. and H. Chasis. 1931. The use of sulphocyanate in the treatment of hypertension.
NY State J Med 31: 1322-1324.

Gorman, W.F, E. Messinger and M. Herman. 1949. Toxicity of thiocyanates used in treatment
of hypertension. Ann Intern Med 30: 1054-1059.

Grandas, F., J. Artieda and J. A. Obeso. 1989. Clinical and CT Scan Findings in a Case of
Cyanide Intoxication. Movement Disorders 4(2): 188-193.

Graziano, J.H., Y.S.Thornton, J.K. Leong and A. Cerami. 1973. Pharmacology of cyanate. 2.
Effects on the endocrine system. J Pharmacol Exp Ther 185: 667-675.

Great Lakes Water Quality Board. 1983. An inventory of chemical substances identified in the
Great Lakes ecosystem. Vol. 1. Summary Report to the Great Lakes Water Quality Board.
Windsor Ontario, Canada 195. (As cited in ATSDR, 1997.)

Guillen, F.J. and J.J. Vazquez. 1984. Cyanamide-induced liver cell injury. Experimental study in
the rat. Lab Invest 50(4): 385-93.

EPA/OW/OST/HECD

X-6

Final draft

-------
Drinking Water Criteria Document for Cyanogen Chloride and Potential Metabolites

Haddow, J. E., G. E. Palomake, W. C. Allan, J. R. Williams, G. J. Knight, J. Gagnon, C. E.
O'Heir, M. L. Mitchell, R. J. Hermos, S. E. Waisbren, J. D. Faix and R. Z. Klein. 1999. Maternal
thyroid deficiency during pregnancy and subsequent neuropsychological development of the child.
NEJM 341(8): 549-555.

Haut, M.J., P.P. Toskes, P.K. Hildenbrandt, B.E. Glader and M.E. Conrad. 1975. In vivo
hepatic and intestinal toxicity of sodium cyanate in rats: cyanate-induced alterations in hepatic
glycogen metabolism. J Lab Clin Med 85(1): 140-54.

Hauth, J.C., J. Hauth, R,B, Drawbaugh, L.C. Gilstrap III and W.P. Pierson. 1984. Passive
smoking and thiocyanate concentrations in pregnant women and newborns. Obstet Gynecol
63(4):519-522. (As cited in AT SDR, 1997.)

Haymaker, W., A.M. Ginzler and R.L. Ferguson. 1952. Residual Neuropathologial Effects of
Cyanide Poisoning: A Study of the Central Nervous System of 23 Dogs Exposed to Cyanide
Compounds. The Military Surgeon 111(4): 231-246.

Health Canada. 2001. Summary of guidelines for Canadian drinking water quality. Available at
http://www.hc-sc.gc.ca/ehp/ehd/bch/water_quality.htm Supporting documentation dated 7/91.

Heidelberger, C., A.E. Freeman, R.J. Pienta, A. Sivak, J.S. Bertram, B.C. Casto, V.C. Dunkel,
M.W. Francis, T. Kakunaga, J.B. Little, and L.M. Schechtman. 1983. Cell transformation by
chemical agents - a review and analysis of the literature. A report of the U.S. Environmental
Protection Agency Gene-Tox Program. Mutat Res 114:283-385.

Heydens, W.F. 1985. Effects of thiocyanate on fetal and postnatal development and thyroid
function in rats. Diss Abstr Int B 45:2124.

Hill, R.N., L.S. Erdreich, O.E. Paynter, P.A. Roberts, S.L. Rosenthal and C.F. Wilkinson. 1989.
Review: Thyroid follicular cell carcinogenesis. Fund Appl Toxicol 12: 629-697.

Himwich, W. A. and J. P. Saunders. 1948. Enzymatic Conversion of Cyanide to Thiocyanate.
Am J Physiol 153:348-354.

Hollowell, J. G., N. W. Staehling, W. H. Hannon, D. W. Flanders, E. W. Gunter, G. F. Maberly,
L. E. Braverman, S. Pino, D. T. Miller, P. L. Garbe, D. M. DeLozier and R. J. Jackson. 1998.
Iodine nutrition in the United States. Trends and public health implications: iodine excretion data
from National Health and Nutrition Examination Surveys I and III (1971-1974 and 1988-1994). J
Clin Endocrinol Metab 83(10): 3401-3408.

Holthman, S.B. and F. Schutz. 1948. Cyanase, a new enzyme catalyzing the hydrolysis of
cyanate. Experientia 4:398-399 (As cited in Johnson et al., 1985).

Honig, D.H., M.E. Hockridge, R.M. Gould and J.J. Rackis. 1983. Determination of cyanide in
soybeans and soybean products. J Agric Food Chem 31:272-275. (As cited in ATSDR, 1997.)

EPA/OW/OST/HECD

X-7

Final draft

-------
Drinking Water Criteria Document for Cyanogen Chloride and Potential Metabolites

Howard, J.W. and R.F. Hanzal. 1955. Chronic toxicity for rats of food treated with hydrogen
cyanide. Agric Food Chem 3(4): 325-329.

Huang, J., H. Niknahad, S. Khan and P.J. O'Brien. 1998. Hepatocyte-catalysed detoxification of
cyanide by L-and D-cysteine. Biochem Pharmacol 55(12): 1983-90.

IARC. 1992. International Agency for Research on Cancer. Occupational exposures to mists and
vapours from sulfuric acid and other strong inorganic acids. IARC Monographs on the evaluation
of the carcinogenic risk of chemicals to humans. 54: 41-130.

Jackson, L.C. 1988. Behavioral effects of chronic sublethal dietary cyanide in an animal model:
Implications for humans consuming cassava {Manihot esculentd). Hum Biol 60: 597-614.

Johnson, J.D., P.R. Mayer and G.E. Isom. 1985. Pharmacokinetic analysis of potassium cyanate
in mice using whole blood and expired air. Drug Metabol Dispos 13 (2): 260-2.

Kamalu, B.P. 1993. Pathological changes in growing dogs fed on a balanced cassava (Manihot
esculenta Crantz) diet. BrJNutr 69(3): 921-934.

Kanno, J., C. Matsuoka, A.K. Furuta, H. Onodera, H. Miyajima, A. Maekawa and Y. Hayashi.
1990. Tumor promoting effect of goitrogens on the rat thyroid. Toxicol Pathol 18(2): 239-246.

Kaplan, H.L. 1987. Effects of irritant gases on avoidance/escape performance and respiratory
response of the baboon. Toxicology 47(1-2): 165-79.

Kaplan, H.L., A. Anzueto, W.G. Switzer and R.K. Hinderer. 1988. Effects of hydrogen chloride
on respiratory response and pulmonary function of the baboon. J Toxicol Environ Health 23(4):
473-93.

Kato, T., M. Kameyama, S. Nakamura, M. Inada and H. Sugiyama H. 1985. Cyanide
metabolism in motor neuron disease. Acta Neurol Scand 72: 151-156.

Kawana, S. 1997. Drug eruption induced by cyanamide (carbimide): A clinical and
histopathologic study of 7 patients. Dermatology 195(1): 30-4.

Kern, H.L., R.W. Bellhorn and C.M. Peterson. 1977. Sodium cyanate-induced ocular lesions in
the beagle. J Pharmacol Exp Therap 200(1): 10-6.

Khandekar, J.D. and H. Edelman. 1979. Studies of amygdalin (laetrile) toxicity in rodents. J Am
Med Assoc 242:169-171. (As cited in AT SDR, 1997.)

Kier, L.D. 1988. Comments and perspective on the EPA workshop on "The relationship
between short-term information and carcinogenicity." Environ Mol Mutagen 11: 147-157. (As
cited in Rosenkranz and Klopman, 1990.)

EPA/OW/OST/HECD

X-8

Final draft

-------
Drinking Water Criteria Document for Cyanogen Chloride and Potential Metabolites

Kihlman, B.A. 1957. Experimentally induced chromosome aberrations in plants. 1. The
production of chromosome aberrations by cyanide and other heavy metal complexing agents. J
Biophys Biochem Cytol 3:363-380.

Kilburn, K.H. 1996. Effects of a hydrochloric acid spill on neurobehavioral and pulmonary
function. J Occup Environ Med 38(10): 1018-25.

Knowles, C.J. 1988. Cyanide utilisation and degradation by microorganisms. In: Cyanide
compounds in biology. CIBA foundation symposium 140., Wiley, Chichester, pp. 3-15. (As cited
in AT SDR, 1997.)

Koibuchi, N. and W.W. Chin. 2000. Thyroid Hormone Action and Brain Development. Trends
Endocrin Metab 11:123-128.

Krasner, S.W., M.J. McGuire, J.G. Jacangelo, et al. 1989. The occurrence of disinfection by-
products in U.S. drinking water. J Am Water Works Assoc 81:41-53.

Kreutler, P.A., V. Varbanov, W. Goodman, G. Olaya and J.B. Stanbury. 1978. Interactions of
Protein Deficiency, Cyanide, and Thiocyanate on Thyroid Function in Neonatal and Adult Rats.
Am J clin Nutr 31:282-289.

Kushi, A., T. Matsumoto, and D. Yoshida. 1983. Mutagen from the gaseous phase of protein
pyrolyzate. Agric Biol Chem 47:1979-1982. (As cited in ATSDR, 1997)

Laesch, E.E. and R. El Shawa. 1981. Multiple cases of cyanide poisoning by apricot kernels in
children from Gaza. Pediatrics 68(1): 5-7.

Les, E.P. 1968. Effect of acidified-chlorinated water on reproduction in C3H/HEJ and
C57BL/6J mice. Lab Anim Care 18:210-213.

Leuschner, J., A. Winkler and F. Leuschner. 1991. Toxicokinetic aspects of chronic cyanide
exposure in the rat. Toxicol Lett 57(2): 195-201.

Levin, B.C., M. Paabo, J.L. Gurman and S.E. Harris. 1987. Effect of exposure to single or
multiple combinations of the predominant toxic gases and low oxygen atmospheres produced in
fires. Fundam Appl Toxicol 9:236-250.

Levis, A.G. and F. Majone. 1981. Cytotoxic and clastogenic effects of soluble and insoluble
compounds containing hexavalent and trivalent chromium. Br J Cancer 44: 219-235.

Lewis, J.L., C.E. Rhoades, P.G. Gervasi, W.C. Griffith and A.R. Dahl. 1991. The cyanide-
metabolizing enzyme rhodanese in human nasal respiratory mucosa. Toxicol Appl Pharmacol
108(1): 114-20.

Lide, D.R., ed. 1992. CRC Handbook of Chemistry and Physics, 73rd Edition. CRC Press, Inc.

EPA/OW/OST/HECD

X-9

Final draft

-------
Drinking Water Criteria Document for Cyanogen Chloride and Potential Metabolites

Liebowitz, D. and H. Schwartz. 1948. Cyanide poisoning: Report of a case with recovery. Am J
Clin Pathol 18:965-970.

Lijinsky W. and M.D. Reuber. 1982. Transnitrosation by nitrosamines in vivo. N-nitroso
compounds: occurence and biological effects. IARC Scientific Publications 41: 625-631.

Lijinsky, W. and R.M. Kovatch. 1989. Chronic toxicity tests of sodium thiocyanate with sodium
nitrite in F344 rats. Toxicol Ind Health 5(1): 25-29.

Lindberg, H.A., M.H. Wald and M.H. Barker. 1941. Observations on the pathologic effects of
thiocyanate: an experimental study. Am Heart J 21: 605-616.

Lobert, J.M. and J. Warnatz. 1993. Emission from the combustion process in vegetation. In:
P.J. Crutzen and J.G. Goldammer, eds.. Fire in the environment: The ecological, atmospheric and
climatic importance of vegetation fires. John Wiley and Sons, pp. 15-37. (As cited in ATSDR,
1997.)

Loomis, C.W. and J.F. Brien. 1983. Inhibition of hepatic aldehyde dehydrogenase in the rat by
calcium carbimide (calcium cyanamide). Can J Physiol Pharmacol 61: 1025-1034 (As cited in
Obach et al., 1989).

Loveless, L.E., E. Spoerl and T.H. Weisman. 1954. Survey of effects of chemicals on division
and growth of yeast and escherichia coli. J Bacteriol 68: 637-644.

Lucas, A.D. 1992. Health hazards associated with the cyanotype printing process. J Environ
Pathol Toxicol Oncol 11(1): 18-20. (As cited in ATSDR, 1997.)

McCarroll, N.E., C.E. Piper and B.H. Keech. 1981a. An /•]. coli microsuspension assay for the
detection of DNA damage induced by direct-acting agents and promutagens. Environ Mutagen 3:
429-444.

McCarroll, N.E., B.H. Keech and C.E. Piper. 1981b. Microsuspension adaptation of the Bacillus
subtilis rec assay. Environ Mutagen 3: 607-616.

McDougall, P.T., N.S. Wolf, W.A. Stenback, and J.J. Trentin. 1967. Control of Pseudomonas
aeruginosa in an experimental mouse colony. Lab Anim Care 17: 204-214.

McMahon, T. and L. Birnbaum. 1990. Age-related changes in biotransformation and toxicity of
postassium cyanide (KCN) in male C57BL/6N mice [Abstract], In: Proceedings of the 29th
Annual Meeting of the Society of Toxicology, Miami Beach, Fl.

MelzerM.S., R.T. Christian, J.F. Dooley, B. Schumann, H.L. Su, and S. Samuels. 1983.
Mutagenicity of cyanate, a decomposition product of mnu. Mutat Res 116: 281-287.

EPA/OW/OST/HECD

X-10

Final draft

-------
Drinking Water Criteria Document for Cyanogen Chloride and Potential Metabolites

Menargues, A., R. Obach and J.M. Valles. 1984. An evaluation of the mutagenic potential of
cyanamide using the micronucleus test. Mutat Res 136: 127-129.

Mertschenk, B., W. Bornemann, J.G. Filser, L. von Meyer, U. Rust, J.C. Schneider and C.
Gloxhuber. 1991. Urinary excretion of acetylcyanamide in rat and human after oral and dermal
application of hydrogen cyanamide (H2NCN). Arch Toxicol 65(4): 268-72.

Michigan Department of Public Health. 1977. Communication to the TLV committee.
November. (As cited in ACGIH, 1991).

Midwest Research Institute. 1997. In vitro study of cyanogen chloride in blood - amended
report. NIEHS Contract No. N01-ES-55385. April 30, 1997.

Migridichian, V. 1946. The chemistry of organic cyanogen compounds. New York: Reinhold
Publishing Corporation, pp. 102-116. As cited in U.S. EPA, 1990.

Mlingi, N., V.D. Vassey, A.B.M. Swai, et al. 1993. Determinants of cyanide exposure from
cassava in konzo-affected population in northern Tanzania. Int J Food Sci Nutr 44(2): 137-144.
(As cited in AT SDR, 1997.)

Mlingi, N., N.H. Poulter and H. Rosling. 1992. An outbreak of acute intoxications from
consumption of insufficiently processed cassava in Tanzania. Nutr Res 12(6):677-687. (As cited
in AT SDR, 1997.)

Morita, T., Y. Watanabe, K. Takeda and K. Okumura. 1989. Effects of pH in the in vitro
chromosomal aberration test. Mutat Res 225:55-60.

Morreale de Escobar, G., M. J. Obregon and F. Escobar del Ray. 2000. Is neuropsychological
development related to maternal hypothyroidism or to maternal hypothyroxinemia? J Clin
Endocrinol Metab 85:3975-3987.

Moudgal, C.J., J.C. Lipscomb, and R.M. Bruce. 2000. Potential health effects of drinking water
disinfection by-products using quantitative structure toxicity relationship. Toxicology 147: 109-
131.

Miiller, A.J. 1965. Arabidopsis Information Service. 6: 22-24. Unpublished article and no
further information available. (As cited in Rede, 1982).

Myers, P.R., D.E. Rawlings, D.R. Woods and G.G. Lindsey. 1993. Isolation and
characterization of a cyanide dihydratase from Bacillus pumilus CI. J Bacteriol 175(19):6105-
6112. (As cited in AT SDR, 1997.)

Nagasawa, H., R. Yanai, Y. Nakajima, H. Namiki, S. Kikuyama and K. Shiota. 1980. Inhibitory
effects of potassium thiocyanate on normal and neoplastic mammary development in female mice.
Eur J Cancer 16(4): 473-80.

EPA/OW/OST/HECD

X-ll

Final draft

-------
Drinking Water Criteria Document for Cyanogen Chloride and Potential Metabolites

NAS. 1980. National Academy of Sciences. Drinking water and health, Vol. 3. Washington,
DC: National Academy Press.

NAS. 1977. National Academy of Sciences. Drinking water and health. Washington, DC:
National Academy Press.

NCI. 1979. National Cancer Institute. Bioassay of calcium cyanamide for possible carcinogenicity.
Technical report series No. 163. National Toxicology Program.

Nicholson, D.H., D.R. Harkness, W.E. Benson and C.M. Peterson. 1976. Cyanate-induced
cataracts in patients with sickle-cell hemoglobinopathies. Arch Ophthalmol 94(6): 927-30.

NIOSH. 1990. National Istitute for Occupational Safety and Health. Unpublished provisional
data as of 7/1/90, National Occupational Exposure Survey (1981-83). Cincinnati, OH. (As cited
in AT SDR, 1997.)

NIOSH. 1982. National Institute for Occupational Safety and Health. In depth survey report of
American Airlines plating facility, NIOSH, Cincinnati, OH. PB83-187799, Springfield, VA. (As
cited in ATSDR, 1997.)

NIOSH. 1976. National Institute for Occupational Safety and Health. Health Hazard Evaluation
Report No. 74-129-268, U.S. Dept. of Health, Education, and Welfare, Center for Disease
Control, NIOSH, Cincinnati, OH. (As cited in ATSDR, 1997.)

NRC. National Research Council. 1989. Recommended Dietary Allowances 10th edition,
National Academy Press.

NTP. 1993. Technical Report on toxicity studies of sodium cyanide (CAS No. 143-33-9)
administered in drinking water to F344/N rats and B6C3FI mice. N.I.H. Publication 94-3386.
Public Health Service, National Institutes of Health, U.S. Department of Health and Human
Services, National Toxicology Program.

Nylander, O., L. Holm, E. Wilander and A. Hallgren. 1994. Exposure of the duodenum to high
concentrations of hydrochloric acid. Effects on mucosal permeability, alkaline secretion, and
blood flow. Scand J Gastroentero 29(5): 437-44.

Obach, R., A. Rives, J.M. Valles, A. Menargues, J. Prunonosa and J. Valles. 1985. Lack of
hepatotoxicity after long-term administration of cyanamide in rats: a histological and biochemical
study. Acta Pharmacol Toxicol 57(4): 279-84.

Obach, R., C. Valenti, J. Valles, J.M. Valles and J. Domenech. 1986a. Bioavailability of
cyanamide in fasted and unfasted rats. Biopharm Drug Dispos 7(3): 273-80.

EPA/OW/OST/HECD

X-12

Final draft

-------
Drinking Water Criteria Document for Cyanogen Chloride and Potential Metabolites

Obach, R., A. Menargues, J. Valles, J.M. Valles and J.A. Garcia-Sevilla. 1986b. Effects of
cyanamide on body weight and brain monoamines and metabolites in rats. Eur J Pharmacol
127(3): 225-31.

Obach, R., H. Colom, J. Arso, C. Peraire, and J. Prunonosa. 1989. Pharmacokinetics of
cyanamide in dog and rat. J Pharm Pharmacol 41:624-627

O'Brien, G.M., L. Mbome, A.J. Taylor, et al. 1992. Variations in cyanogen content of cassava
during village processing in Cameroon. Food Chem 44(2): 131-136. (As cited in AT SDR, 1997.)

Oh, M.S. 1994. Water, electrolyte, and acid-base balance. In: Shiles, M.E., J.A. Olson, and M.
Shike., eds. Modern nutrition in health and disease, 8th edition. Philadelphia: Lea & Febiger, pp.
134-135.

Ohio River Valley Water Sanitation Commission. 1982. Assessment of the water quality
conditions: Ohio river mainstream 1980-81. Ohio River Valley Water Sanitation Comm.,
Cincinnati, OH. (As cited in AT SDR, 1997.)

Ohnishi, A., C.M.Peterson and P.J. Dyck. 1975. Axonal degeneration in sodium cyanate-induced
neuropathy. Arch Neurol 32(8): 530-4.

Ohya, T. and S. Kanno. 1987. Formation of cyanogen chloride during the chlorination of water
containing aromatic compounds and ammonium ion. J Pharm Sci 76(11).

Okoh, P.N. 1983. Excretion of 14-C-labeled cyanide in rats exposed to chronic intake of
potassium cyanide. Toxicol Appl Pharmacol 70:335-339. (As cited in ATSDR, 1997).

O'Neil, M.J., A. Smith, and P.E. Heckelman, eds. 2001. The Merck index-an encyclopedia of
chemicals, drugs, and biologicals. 13th ed. Whitehouse Station, NJ: Merck&Co. P. 467.

Painter R.B. and R. Howard. 1982. The HeLa DNA-synthesis inhibition test as a rapid screen for
mutagenic carcinogens. Mutat Res 92:427-437. (As cited in ATSDR, 1997).

Palmer, I.S. and O.E. Olson. 1979. Partial prevention by cyanide of selenium poisoning in rats.
Biochm Biophys Res Commun 90(4): 1379-1386.

Palmer, R.S. and H.B. Sprague. 1929. Four cases illustrating the untoward symptoms which may
be produced by the use of potassium sulfocyanate in the treatment of hypertension. M Clin North
America 13: 215-220.

Palmer, R.S., L.S. Silver and P.D. White. 1929. Clinical use of potassium sulfocyanate in
hypertension: a preliminary report of 59 cases. New Engl J Med 201: 709-714.

Pavlova, T.E. 1976. Disturbance of development of the progeny of rats exposed to hydrogen
chloride. Bull Exp Biol Med (USSR) 82: 1078-1081.

EPA/OW/OST/HECD

X-13

Final draft

-------
Drinking Water Criteria Document for Cyanogen Chloride and Potential Metabolites

Peterson, C.M., P. Tsairis, A. Onishi, Y.S. Lu and R. Grady. 1974. Sodium cyanate induced
polyneuropathy in patients with sickle-cell disease. Ann Intern Med 81(2): 152-8.

Philbrick, D.J., J.B. Hopkins, D.C. Hill, J.C. Alexander and R.G. Thomson. 1979. Effects of
prolonged cyanide and thiocyanate feeding in rats. J Toxicol Environ Health 5(4): 579-92.

Porterfield, S. P. 1994. Vulnerability of the developing brain to thyroid abnormalities:
Environmental insults to the thyroid system. Environ Health Perspect 102(Suppl. 2): 125-130.

Porterfield, S. P. 2000. Thyroidal dysfunction and environmental chemicals-Potential impact on
brain development. Environ Health Perspect 108(Suppl. 3): 433-438.

Potter, A.L. 1950. The successful treatment of two recent cases of cyanide poisoning. BrJInd
Med 7:125-130.

Prentiss, A.M. 1937. Chemicals in war: A treatise on chemical warfare. Chapter VIII. Systemic
toxic agents. New York: McGraw-Hill Book Company, pp. 170-176.

Price, C.C., T.E. Larson, K.M. Beck, F.C. Harrington, L.C. Smith, and I. Stephanoff. 1947.
Hydrolysis and chlorinolysis of cyanogen chloride. J Am Chem Soc 69: 1640-1644.

Purser, D.A., P. Grimshaw, and K.R. Berrill. 1984. Intoxication by cyanide in fires: a study in
monkeys using polyacrylonitrile. Arch Environ Health 39:394-400.

Pyska, H. 1977. Effect of thiocyanate on mammary gland growth in rats. J Dairy Res 44(3): 427-
31.

Raghunath, M. and T.S. Bala. 1998. Diverse effects of mild and potent goitrogens on blood-
brain barrier nutrient transport. Neurochem Int 33(2): 173-7.

Rawson, R.W., J.F. Tannheimer and W. Peacock. 1944. The uptake of radioactive iodine by the
thyroids of rats made goiterous by potassium thiocyanate and by thiouracil. Endocrinology 34:
245-253.

Redei, G.P. 1982. Mutagen assay with Arabidopsis. A report of the U.S.

Environmental Protection Agency Gene-Tox Program. Mutat Res 99:243-255.

Reed, C.I. 1920. Chronic poisoning from cyanogen chloride. J Ind Hyg 2(4): 140-143.

Research Triangle Institute. 1999. Perchlorate peer review workshop report. Reseach Triangle
Park, NC: U.S. Environmental Protection Agency, Office of Solid Waste and Emergency
Response.

EPA/OW/OST/HECD

X-14

Final draft

-------
Drinking Water Criteria Document for Cyanogen Chloride and Potential Metabolites

Richardson, S.D. 1998. Identification of drinking water disinfection byproducts. In: R.A.
Meyers, ed. John Wiley's Encyclopedia of Environmental Analysis & Remediation. 3:1398-1421.
(As cited in AT SDR, 1997.)

Robertson, S. A. 1989. Simple acid-base disorders. Vet Clin North Am Small Anim Pract 19(2):
289-306.

Rosenberg, N.L., J.A. Myers, and W.R.W. Martin. 1989. Cyanide induced parkinsonism clinical
MRI and 6 fluorodopa Fd positron emission tomography pet studies. Neurology 39: 142-144.

Rosenkranz, H.S. and G. Klopman. 1990. Structural alerts to genotoxicity: The interaction of
human and artificial intelligence. Mutagenesis 5:333-361.

Rosling, H. 1987. Cassava toxicity and food security. A report for UNICEF African household
food security programme. 2nd ed. Uppsala, Sweden: Tryck kontakt. pp. 1-40. (As cited in
AT SDR, 1997).

Russell, W.O. and W.C. Stahl. 1942. Fatal poisoning from potassium thiocyanate treatment of
hypertension. JAmMedAssoc 119(15): 1177-1181.

Saincher, A., N. Swirsky and M. Tenenbein. 1994. Cyanide overdose: Survival with fatal blood
concentration without antidotal therapy. J Emerg Med 12(4): 555-557.

Seigler, D.S. 1991. Cyanide and cyanogenic glycosides. In: Rosenthal, G.A., and M.R.
Berenbaum, eds. Herbivores: Their interaction with secondary plant metabolites. New York,
NY: Academic Press, pp. 35-77. (As cited in AT SDR, 1997.)

Sellakumar, A.R., C.A. Snyder, J.J. Solomon and R.E. Albert. 1985. Carcinogenicity of
formaldehyde and hydrogen chloride in rats. Toxicol Appl Pharmacol 81: 401-406.

Shirota, F.N., E.G. DeMaster and H.T. Nagasawa. 1987. Cyanide is a product of the catalase-
mediated oxidation of the alcohol deterrent agent, cyanamide. Toxicol Lett 37(1): 7-12.

Shirota, F.N., H.T. Nagasawa, C.H. Kwon, and E.G. DeMaster. 1984. N-acetylcyanamide, the
major urinary metabolite of cyanamide in rat, rabbit, dog, and man. Drug Metabol Dispos
12:337-344. (As cited in Obach et al., 1989).

Siemiatycki, J., ed. 1991. Risk factors for cancer in the workplace. Boca Raton, FL: CRC Press.
(As cited by IARC, 1992)

Sina, J.F., C.L. Bean, G.R. Dysart, V.I. Taylor, and M.O. Bradley. 1983. Evaluation of the
alkaline elution/rat hepatocyte asssay as a predictor of carcinogenic/mutagenic potential. Mutat.
Res 113:357-391.

EPA/OW/OST/HECD

X-15

Final draft

-------
Drinking Water Criteria Document for Cyanogen Chloride and Potential Metabolites

Singh, B.M., N. Coles, P. Lewis, R.A. Braithwaite, M. Nattrass and M.G. FitzGerald. 1989. The
metabolic effects of fatal cyanide poisoning. Postgraduate Med J 65(770): 923-5.

Smith, A.G. and R.D. Rudolf. 1928. The use of sulphocyanate of soda in high blood pressure.
Canadian Med Assoc J 19: 288-292.

Smyth, H.F., C.P. Carpenter, C.S. Weil, U.C. Pozzani, J.A. Striegel and J.S. Nycum. 1969.
Range-finding toxicity data: List VII. Am Ind Hyg Assoc J 30:470-476. (As cited in ATSDR,
1997.)

Stadelman, W. 1976. Content of hydrocyanic acid in stone fruit juices. Fluess Obst 43:45-47.
(As cited in ATSDR, 1997/)

Stavert, D.M., D.C. Archuleta, M.J. Behr and B.E. Lehnert. 1991. Relative acute toxicities of
hydrogen fluoride, hydrogen chloride, and hydrogen bromide in nose- and pseudo-mouth-
breathing rats. Fund Appl Toxicol 16(4): 636-55.

Stevens, B., J.Q. Koenig, V. Rebolledo, Q.S. Hanley and D.S. Covert. 1992. Respiratory effects
from the inhalation of hydrogen chloride in young adult asthmatics. J Occup Med 34: 923-929.

Swain, E., C.P. LI and J.E. Poulton. 1992. Development of the potential for cyanogenesis in
maturing black cherry (Prunus serotina Ehrh.) fruits. Plant Physiol (BETHESDA) 98(4): 1423-
1428. (As cited in ATSDR, 1997)

Sylvester, M. and C. Sander. 1990. Immunohistochemical localization of rhodanese. Histochem J
22(4): 197-200.

Taubman, G. and R. Heilborn. 1930. Untersuchungen zur Toxicologie des Natriumrhodanids.
Arch Exper Path Pharmakol 152: 250. (As cited in Lindberg, et al., 1941; study not available for
review)

Teisseire, B.P., C.C. Vieilledent, L.J. Teisseire, M.O. Vallez, R.A. Herigault and D.N. Laurent.
1986. Chronic sodium cyanate treatment induces "hypoxia-like" effects in rats. J Appl Physiol
60(4): 1145-9.

Tellez, I., D. Johnson, R.L. Nagel and A. Cerami. 1979. Neurotoxicity of sodium cyanate. New
pathological and ultrastructural observations in maccaca nemestrina. Acta Neuropathologica.
47(1): 75-9.

Tellez-Nagel, I., J.K. Korthals, H.V. Vlassara and A. Cerami. 1977. An ultrastructural study of
chronic sodium cyanate-indiuced neuropathy. JNeuropathol Exp Neurol 36(2): 352-63.

Tewe, O.O. and J.H. Maner. 1981. Long-term and carry-over effect of dietary inorganic cyanide
(KCN) in the life cycle performance and metabolism of rats. Toxicol Appl Pharmacol 58:1-7.

EPA/OW/OST/HECD

X-16

Final draft

-------
Drinking Water Criteria Document for Cyanogen Chloride and Potential Metabolites

Throssell, D., J. Brown, K.P. Harris and J. Walls. 1995. Metabolic acidosis does not contribute
to chronic renal injury in the rat. Clin Sci 89(6): 643-50.

Throssell, D., K.P. Harris, A. Bevington, P.N. Furness, A.J. Howie and J. Walls. 1996. Renal
effects of metabolic acidosis in the normal rat. Nephron 73(3): 450-5.

Tober-Meyer, B.K., H.J. Bieniek and I.R. Kupke. 1981. Studies on the hygiene of drinking
water for laboratory animals. 2. Clinical and biochemical studies in rats and rabbits during
long-term provision of acidified drinking water. Lab Anim 15(2): 111-7.

Towill, L.E., J.S. Drury, B.L. Whitfield, et al. 1978. Reviews of the environmental effects of
pollutants. V. Cyanide. Cincinnati, OH: EPA Health Effects Research Laboratory, Office of
Research and Development. NTIS PB28-9920. (As cited in AT SDR, 1997)

ToxiGenics, Inc. 1984. 90-Day inhalation study of hydrogen chloride gas in B6C3F1 mice,
Sprague-Dawley rats, and Fischer-344 rats. Study conducted for CIIT, Research Triangle Park,
NC. CIIT Docket No. 20915.

Tylleskar, T., M. Banea, N. Bikangi, R.D. Cooke, N.H. Poulter and H. Rosling. 1992. Cassava
cyanogens and konzo, an upper motoneuron disease found in Africa [published erratum appears
Lancet 1992 Feb 15:339(8790):440], Lancet 339(8787):208-211. (As cited in AT SDR, 1997)

Uitti, R.J., A.H. Rajput, E.M. Ashenhurst and B. Rozdilsky. 1985. Cyanide-induced
parkinsonism: A clinicopathologic report. Neurology 35(6): 921-925.

U.S. EPA. 2001a. U.S. Environmental Protection Agency. Integrated Risk Information System
(IRIS). Online. Cyanide reference dose (RfD). http://www.epa.gov/iris/ Verified 1985.

U.S. EPA. 2001b. U.S. Environmental Protection Agency. Help manual for Benchmark dose
software version 1.3. Washington, DC: Office of Research and Development. EPA
600/R-00/014F. Available at http://cfpub.epa.gov/ncea/cfm/bmds.cfm?ActType=default

U.S. EPA. 2001c. U.S. Environmental Protection Agency. National Drinking Water Occurrence
Database, http://www.epa.gov/ncod/

U.S. EPA. 2000a. U.S. Environmental Protection Agency. ICR data analysis plan. Office of
Water.

U.S. EPA. 2000b. U.S. Environmental Protection Agency. Information collection rule (ICR)-
technical working group data analysis website. U.S. EPA. Available online at
http://www.cadmusonline.net/twg/mainmenu.htm

U.S. EPA. 2000c. U.S. Environmental Protection Agency. Stage 2 Occurrence and exposure
assessment for disinfectants and disinfection byproducts (D/DBPs) in public drinking water
systems. Office of Ground Water and Drinking Water, Washington, D.C.

EPA/OW/OST/HECD

X-17

Final draft

-------
Drinking Water Criteria Document for Cyanogen Chloride and Potential Metabolites

U.S. EPA. 2000d. U.S. Environmental Protection Agency. Methodology for deriving ambient
water quality criteria for the protection of human health. Office of Water, Office of Science and
Technology, Washington, D.C. EPA-822-B-00-004.

U.S. EPA. 2000e. U.S. Environmental Protection Agency. Benchmark dose technical guidance
document. External review draft. Washington, DC: Risk Assessment Forum.

U.S. EPA. 1999. U.S. Environmental Protection Agency. Guidelines for carcinogen risk
assessment. Draft. NCEA-F-0644.

U.S. EPA. 1997a U.S. Environmental Protection Agency. Risk assessment issue paper for:
Derivation of a provisional RfD for thiocyanate. Superfund Technical Support Center, National
Center for Environmental Assessment, Cincinnati, Ohio.

U.S. EPA. 1997b. U.S. Environmental Protection Agency. Volume I - General Factors,
Exposure Factors Handbook, Update to Exposure Factors Handbook. EPA/600/P-95/002Fa.
Volume III - Activity Factors, Exposure Factors Handbook, Update to Exposure Factors
Handbook - EPA/600/P-95/002Fc. EPA/600/8-89/043 - May 1989, Office of Research and
Development, National Center for Environmental Assessment, U.S. Environmental Protection
Agency, Washington, DC.

U.S. EPA. 1996. U.S. Environmental Protection Agency. Proposed Guidelines for Carcinogen
Risk Assessment. Office of Research and Development, Washington, DC. EPA/600/P-92/003C.
April.

U.S. EPA. 1994a. U.S. Environmental Protection Agency. Final Draft for the Drinking Water
Criteria Document on Chlorinated Acids/Aldehydes/Ketones/Alcohols. Prepared for Health and
Ecological Criteria Division Office of Science and Technology, Office of Water. Washington,
DC. EPA 68-C2-0139.

U.S. EPA. 1994b. Environmental Protection Agency. Methods for derivation of inhalation
reference concentrations and application of inhalation dosimetry. EPA/600/8-90/066F.

U.S. EPA. 1993. U.S. Environmental Protection Agency. Reference dose (RfD): Description
and use in health risk assessments. Integrated Risk Information System (IRIS). Online.
Intra-Agency Reference Dose Work Group, Office of Health and Environmental Assessment,
Environmental Criteria and Assessment Office. Cincinnati, OH.

U.S. EPA. 1992. U.S. Environmental Protection Agency. Drinking water criteria document for
cyanide. Prepared by the Office of Health and Environmental Assessment, Environmental Criteria
and Assessment Office, Cincinnati, OH for the Office of Water. ECAO-CIN-442.

U.S. EPA. 1990. U.S. Environmental Protection Agency. Reactions of cyanogen chloride.

Office of Drinking Water, Criteria and Standards Division. TR-1242-5 8.

EPA/OW/OST/HECD

X-18

Final draft

-------
Drinking Water Criteria Document for Cyanogen Chloride and Potential Metabolites

U.S. EPA. 1988. U.S. Environmental Protection Agency. Recommendations for and
documentation of biological values for use in risk assessment. Environmental Criteria and
Assessment Office, Office of Research and Development. EPA/600/6-87/008, PB88-179874

U.S. EPA. 1987. U.S. Environmental Protection Agency. Rough Final Draft for the Drinking
Water Criteria Document on Haloacetonitriles, Chloropicrin and Cyanogen Chloride. Prepared by
ICAIR, Life Systems, Inc., for Criteria and Standards Division, Office of Drinking Water,
Washington, DC, under Contract 68-03-3279.

U.S. EPA. 1986. U.S. Environmental Protection Agency. Guidelines for carcinogen risk
assessment. Fed Reg 51(185): 33992-34003.

U.S. EPA. 1980. U.S. Environmental Protection Agency. Water quality criteria documents:
Availability. Fed Reg 45:79318-79379.

U.S. FDA. 2002. U.S. Food and Drug Administration. History of FDA's Total Diet Study.
http://vm.cfsan.fda.gov/~comm/tds-hist.html. Downloaded June 2002.

U.S. FDA. 1999. U.S. Food and Drug Administration. Economic characterization of the dietary
supplement industry final report. Center for Food Safety and Nutrition. Available online:
http://www.csfan.fda.gov/~comm/ds-econ4. html.

Valerdiz, S. and J.J. Vazquez. 1989. Cyanamide and its calcium form: do they differ with
respect to their action on the liver cell? Experimental study in the rat. Appl Pathol 7(6): 344-9.

Valles, J. R. Obach, A. Menargues, J.M. Valles and A. Rives. 1987. A two-generation
reproduction-fertility study of cyanamide in the rat. Pharmacol Toxicol 61(1): 20-5.

Vazquez, J.J., F.J. Guillen, J. Zozaya and M. Lahoz. 1983. Cyanamide-induced liver injury. A
predictable lesion. Liver 3(4): 225-30.

Venier, P., A. Montaldi, L. Busi, C. Gava, L. Zentilin, G. Tecchio, V. Bianchi and A.G. Levis.
1985. Genetic Effects of Chromium Tannins. Carcinogenesis (London) 6: 1327-1335.

Vesey, C.J., P.V. Cole and P.J. Simpson. 1976. Cyanide and thiocyanate concentrations
following sodium nitroprusside infusion in man. Br J Anaesth 48(7):651-660. (As cited in
AT SDR, 1997)

Voldrich, M. and V. Kyzlink. 1992. Cyanogenesis in canned stone fruits. J Food Sci 57(1): 161-
162,189. (As cited in ATSDR, 1997)

WHO. 2000. World Health Organization. Environmental Health Criteria: 216 Disinfectant
By-products. Geneva, Switzerland.

EPA/OW/OST/HECD

X-19

Final draft

-------
Drinking Water Criteria Document for Cyanogen Chloride and Potential Metabolites

WHO. 1998. World Health Organization. Guidelines for drinking-water quality, 2nd ed. Vol. 2.
Health criteria and other supporting information, 1996 (pp. 940-949) and Addendum to Vol. 2 .
1998 (pp. 281-283). Geneva, Switzerland.

WHO. 1992. World Health Organization. Technical Report Series 828. Evaluation of certain
food additives and naturally occurring toxicants. Thirty-ninth report of the Joint FAO/WHO
Expert Committee on food additives. Geneva, Switzerland. (As cited in ATSDR, 1997)

Willhite, C.C., and R.P. Smith. 1981. The role of cyanide liberation in the acute toxicity of
aliphatic nitriles. Toxicol Appl Pharmacol 59:589-602. (As cited in ATSDR, 1997)

Wilson, J. 1983. Cyanide in human disease: a review of clinical and laboratory evidence.

Fundam Appl Toxicol 3:397-399.

Wing, D.A. and S.I. Baskin. 1992. Modifiers of mercaptopyruvate sulfurtransferase catalyzed
conversion of cyanide to thiocyanate in vitro. J Biochem Toxicol 7(2): 65-72.

Wolff, J. 1998. Perchlorate and the thyroid gland. Pharmacol Rev 50:89-105.

Wolff, J. and J.R. Maurey. 1963. Thyroidal iodide transport. IV. The role of ion size. Biochim
Biophys Acta 69: 48-58.

Wolff, J., I.L. Chaikoff, A. Taurog and L. Rubin. 1946. The disturbance in iodine metabolism
produced by thiocyanate: The mechanism of its goitrogenic action with radioactive iodine as
indicator. Endocrinology 39: 140-148.

Wolfsie, J.H. and C.B. Shaffer. 1958. Hydrogen cyanide. Hazards, toxicology, prevention and
management of poisoning. J Occup Med 2: 289 (as cited by ACGIH, 1996).

Wood, J.L. 1975. Biochemistry. In: Newman, A.A., ed. Chemistry and biochemistry of
thiocyanic acid and its derivatives. London: Academic Press, pp. 156-221.

Wu, W.W., P.A. Chadik, W.M. Davis, D.H. Powell and J.J. Delfino. 1998. Disinfection
byproduct formation from the preparation of instant tea. J Agric and Food Chem 46(8):3272-
3279.

Xie, Y., and D.A. Reckhow. 1993. A rapid and simple analytical method for cyanogen chloride
and cyanogen bromide in drinking water. Water Res 27(3):507-511. (As cited in ATSDR, 1997)

Yamanaka, S., S. Takaku, Y. Takaesu and M. Nishimura. 1991. Validity of salivary thiocyanate
as an indicator of cyanide exposure from smoking. Bull Tokyo Dental Coll 32(4): 157-163. (As
cited in ATSDR, 1997.)

EPA/OW/OST/HECD

X-20

Final draft

-------
Drinking Water Criteria Document for Cyanogen Chloride and Potential Metabolites

Yoon, J.S., J.M. Mason, R. Valencia, R.C. Woodruff and S. Zimmering. 1985. Chemical
mutagenesis testing in drosophila. 4. Results of 45 coded compounds tested for the national
toxicology program. Environ Mutagen 7:349-367.

Young, M.S. and P.C. Uden. 1994. By-products of the aqueous chlorination of purines and
pyrimidines. Env Sci Tech 28(9): 1755-1758.

Zeiger, E. 1987. Carcinogenicity of mutagens: Predictive capability of the salmonella mutagenesis
assay for rodent carcinogenicity. Cancer Res 47:1287-1296.

Zeiger, E., B. Anderson, S. Haworth, T. Lawlor and K. Mortelmans. 1988. Salmonella
mutagenicity tests: IV. Results from the testing of 300 chemicals. Environ Mol Mutagen 11(12):
1-158.

EPA/OW/OST/HECD

X-21

Final draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Appendix A. Benchmark Dose Modeling Results
Introduction

Benchmark dose (BMD) modeling was performed to identify potential critical effect levels
as alternatives to the study NOAEL/LOAELs for derivation of the HAs for DBAN, DCAN, and
TCAN, No adequate studies were available to support a quantitative dose-response assessment
for BCAN, Individual modeling output for each endpoint is provided in Appendix B.

Methods

Benchmark Dose

The haloacetonitrile data sets considered for dose-response modeling were all continuous
endpoints. The modeling was conducted according to draft EPA guidelines (U.S. EPA, 2000c)
using Benchmark Dose Software (BMDS version 1.3.1), available from the U.S. EPA (U.S. EPA,
2001). The methods and models applied to the continuous endpoints are presented here.

The continuous endpoints of interest with respect to haloacetonitrile toxicity were
quantitatively summarized by group means and measures of variability (standard errors or
standard deviations). Since all of the endpoints that were modeled were continuous rather than
quantal (e.g., incidence data) in nature, the Hill, power, and polynomial models were used for
each data set. Linear fits to the data were incorporated into the analysis by allowing the power
and polynomial models to simplify to linear equations as dictated by the data (for short-term
studies). The linear model option in BMDS was also run separately for the longer-term studies,

EPA/OW/OST/HECD

A-l

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

since this provided the advantage of obtaining goodness-of-fit p-values (since the number of
parameters is smaller for the linear model) for the longer-term data sets in which the high dose
was removed; insufficient degrees of freedom were available for calculation of p-values for the
power or polynomial models for these data sets. An attempt to fit the data using a hybrid
modeling approach for the longer-term studies failed to compute a BMDL estimate. The hybrid
approach defines the benchmark response (BMR) directly in terms of risk, as opposed to the
standard approach, which defines the BMR in terms of a change in the mean. Furthermore, the
hybrid model software in BMDS is still undergoing Beta-testing, and was not considered
sufficiently validated to have used a BMDL from this model as the basis for the quantitative dose-
response assessment.

These mathematical models fit to the data are defined here. In all cases, |i(d) indicates the
mean of the response variable following exposure to dose d.

The polynomial model is defined as:

n(d) = p0 + p1d + ... + Mn

where the degree of the polynomial, n, was set less than or equal to the number of dose groups in
the experiment being analyzed. Note that U.S. EPA (2000c) recommends the use of the most
parsimonious model that provides an adequate fit to the data. It may appear that the use of a
polynomial model with degree possibly as great as the number of dose groups would not yield the

EPA/OW/OST/HECD

A-2

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
most parsimonious model. However, allowing the model to have that degree is not the same as
forcing the model to have that degree; in the model fitting, if fewer parameters (e.g., a lower
degree polynomial) is adequate and consistent with the data, then the fitting will reflect that fact
and a more parsimonious model will be the result. For these analyses, the values of the (3
parameters allowed to be estimated were constrained to be either all nonnegative or all
nonpositive (as dictated by the data set being modeled, i.e., nonnegative if the mean response
increased with increasing dose or nonpositive if the mean response decreased with increasing
dose).

The power model is represented by the equation:

|i(d) = y + (3da

where the parameter a is restricted to be nonnegative. [The linear model is obtained when a is
fixed at a value of 1. The linear model was not separately fit to the data; if the result of fitting the
power model does not result in the linear form, a = 1, then the linear model does not fit as well
as the more general power model, by definition.]

The Hill model is given by the following equation:

|i(d) = y+(vdn)/(dn + kn))
where the parameters n and k are restricted to be positive (in fact, n > 1).

EPA/OW/OST/HECD

A-3

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
In the case of continuous endpoints, one must assume something about the distribution of
individual observations around the dose-specific mean values defined by the above models. The
assumptions imposed by BMDS were used in this analysis: individual observations were assumed
to vary normally around the means with variances given by the following equation:

o;2 = a2 x[(i(di)]p

where both a2 and p were parameters estimated by the model.

Given those assumptions about variation around the means, maximum likelihood methods
were applied to estimate all of the parameters, where the log-likelihood to be maximized is
(except for an additive constant) given by

L = X [(Ni/2)xln(oi2) + (N, - l)S|72o,2 + N,{m, - M(d,)J2/2o,2]

where N; is the number of individuals in group i exposed to dose dp and m, and s; are the observed
mean and standard deviation for that group. The summation runs over i from 1 to k (the number
of dose groups).

Goodness of Fit Analyses

For these continuous models, goodness of fit was determined based on a likelihood ratio
statistic. In particular, the maximized log-likelihood associated with the fitted model was

EPA/OW/OST/HECD

A-4

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

compared to the log-likelihood maximized with each dose group considered to have a mean and
variance completely independent of the means and variances of the other dose groups.1 It is
always the case that the latter log-likelihood will be at least as great as the model-associated log-
likelihood, but if the model does a reasonable job of fitting the data, the difference between the
two log-likelihoods will not be too great. A formal statistical test reflecting this idea uses the fact
that twice the difference in the log-likelihoods is distributed as a chi-square random variable. The
degrees of freedom associated with that chi-squared test statistic are equal to the difference
between the number of parameters fit by the model (including the parameters a2 and p defining
how variances change as a function of mean response level) and twice the number of dose groups
(which is equal to the number of parameters estimated by the model assuming independence of
dose group means and variances). Parameters hitting boundary values were not included for
determining degrees of freedom.

Acceptable fit was defined as a goodness-of-fit p-value greater than or equal to 0.1, or a
perfect fit when there were no degrees of freedom for a statistical test of fit. Choice of 0.1 is
consistent with current U.S. EPA guidance for BMD modeling (U.S. EPA, 2000c). If a model
was judged to provide a reasonable BMDL estimate, but the p-value criterion of 0.1 was not met,
the rationale for waiving the p-value criterion is provided in the discussion of the results.

1 If and when BMDS suggested that a homogeneous-variance model was appropriate, the log-likelihood
of the fitted model was compared to the likelihood maximized assuming independent means but a single, constant
variance for all dose groups (the fitted model also assumed that to be the case in such cases').	

EPA/OW/OST/HECD	A-5	Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Goodness-of-fit statistics are not designed to compare different models, particularly if the
different models have different numbers of parameters. Within a family of models, adding
parameters generally improves the fit. BMDS reports the Akaike Information Criterion (AIC) to
aid in comparing the fit of different models. The AIC is defined as -2L+2p, where L is the log-
likelihood at the maximum likelihood estimates for the parameters, and p is the number of model
parameters estimated (ignoring parameters assuming values at the boundaries of their allowable
ranges). When comparing the fit of two or more models to a single data set, the model with the
lesser AIC was considered to provide a superior fit.

Definition of the BMR and Corresponding BMD and BMDL

For the continuous models, BMDs were implicitly defined as follows:

I |i(BMD) - |i(0) | = 8a,

where o, is the model-estimated standard deviation in the control group. In other words, the
BMR was defined as a change in mean corresponding to some multiplicative factor of the control
group standard deviation.

The value of 8 used in this analysis was 1.0. This value was chosen based on EPA draft
guidelines for BMD analyses (U.S. EPA, 2000c), in the absence of a clear biological rationale for
selecting an alternative response level. It is roughly consistent with (though slightly more
conservative than) a choice of 1.1, which according to Crump (1995) corresponds to an additional

EPA/OW/OST/HECD

A-6

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

risk of 10% when the background response rate was assumed to be 1%, with normal variation
around the mean (and constant standard deviation).

Choice of BMDL

The following guidance was followed with regard to the choice of the BMDL to use as a
point of departure for calculation of a health advisory. This guidance is consistent with
recommendations in U.S. EPA (2000c). For each endpoint, the following procedure is
recommended:

1.	Models with an unacceptable fit are excluded.

2.	If the BMDL values for the remaining models for a given endpoint are within a factor of 3,
no model dependence is assumed, and the models are considered indistinguishable in the
context of the precision of the methods. The models are then ranked according to the
AIC, and the model with the lowest AIC is chosen as the basis for the BMDL.

3.	If the BMDL values are not within a factor of 3, some model dependence is assumed, and
the lowest BMDL is selected as a reasonable conservative estimate, unless it is an outlier
compared to the results from all of the other models. Note that when outliers are
removed, the remaining BMDLs may then be within a factor of 3, and so the criteria given
in item 2. would be applied.

4.	The BMDL values from all modeled endpoints are compared, along with any NOAELs or
LOAELs from data sets that were not amenable to modeling, and the lowest NOAEL or
BMDL is chosen.

EPA/OW/OST/HECD

A-7

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
Modeling Results for Short-term Studies

This modeling was done to support the derivation of the Ten-Day HAs. BMD modeling
was conducted only for toxicologically-relevant endpoints that could be used to derive the HAs.
Adequate short-term studies for modeling were available only for DBAN, DC AN, and TCAN,
No suitable studies were available for derivation of the Ten-day HA for BCAN. As a result,
BMD modeling was not performed for BCAN. The BMD modeling results for the short-term
studies are presented in Table A-l and described below.

DBAN

The endpoints modeled for DBAN in the Hayes et al. (1986) 14-day study were body
weight in males and relative liver weight in females. Modeling was done for relative liver weight
for completeness, although as discussed in Chapter V, the relative liver weight changes for DBAN
were not considered to be sufficiently adverse to serve as the basis for the HA.

The body weight response to DBAN in males did not appear to have constant variance;
the variability in the highest dose group was much greater than that observed in the other groups.
Even when a dose-dependent variance was included, however, the polynomial model did not
predict standard deviations that matched well with the observed values, contributing to the
significant lack of fit of that model (p = 0.004). The power model did a much better job of fitting
the observed standard deviations, as well as the observed means, yielding a p-value for goodness-
of-fit of 0.07. The Hill model did not provide an adequate fit. The BMD and BMDL (26 and 16
mg/kg/day, respectively) from the power model are preferred for this data set. The goodness-of-

EPA/OW/OST/HECD

A-8

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
fit statistic for the power model was below the current p-value criterion of 0.1 recommended in
current EPA guidance (U.S. EPA, 2000c). However, the results of the power model were
considered adequate for use in the quantitative dose-response assessment. This consideration was
based on the observation that the visual fit to the data was reasonably good, and that the model-
predicted means and standard deviations were similar to the observed values (i.e. as indicated by
low values for the chi-square residuals).

For the relative liver weight in female rats, all the models fit the data extremely well (p-
values all greater than 0.45). Because of the extra parameters in the Hill model (which visually
looked essentially the same as the polynomial or power models), the AIC for the Hill model was
substantially higher than the AICs for the polynomial or power models. In addition, the BMDL
calculation failed for the Hill model. The polynomial mode gave a slightly better fit than the
power model, and consequently the BMD and BMDL from the polynomial model (31 and 17
mg/kg/day, respectively) are the estimates of choice for this data set.

In summary, for the DBAN Ten-day HA, only the modeling results for decreased body
weight reported in Hayes et al. (1986) were considered for use in the quantitative dose-response
assessment. The estimate of choice for this data set was the BMD of 26 mg/kg/day with the
corresponding BMDL of 16 mg/kg/day obtained from the power model.

EPA/OW/OST/HECD

A-9

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

DCAN

The endpoints modeled for DCAN in the Hayes et al. (1986) 14-day study were body
weight in males and relative liver weight in males and females. Modeling was done for each of
these effects since they were considered to be toxicologically relevant.

A constant variance model was used for the analysis of the DCAN body weight endpoint
in male rats (Hayes et al., 1986). The mean weights showed a non-monotonic dose response,
with the lowest positive dose group having a mean weight greater than that in controls. This
caused some difficulty in fitting the models (p-values all less than or equal to 0.02). However, in
this case, the fits might be judged to be adequate for BMD estimation for several reasons. First,
the poor statistical fit was driven largely by one point, while the visual fits were reasonable.
Second, the BMD estimates (ranging from 32-36 mg/kg/day) were very consistent among the
models, suggesting that the fits were not model-dependent. Third, the BMDL estimates were
associated with a response that corresponded almost exactly to one standard deviation below the
control group mean, using the observed values of the control group mean and standard deviation.
The AIC for the power model was slightly better than those for the other models, and the BMD
of 36 mg/kg/day and BMDL of 25 mg/kg/day from the power model are recommended as the best
estimates for this data set.

For the female rats exposed to DCAN, a non-monotonic dose-response was observed for
increased relative liver weight (with, essentially, a plateau of effect for the highest three dose
groups after little change in the low-dose group compared to controls). Moreover, the observed

EPA/OW/OST/HECD

A-10

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

standard deviation in the high-dose group was three to four times greater than the standard
deviation in the other groups. As a result, the variance modeling approaches available in BMDS
were not capable of fitting the observed dose-variance pattern.2 The linear models to which the
polynomial and power models defaulted could not fit the dose-response pattern. However, the
Hill model did reflect the dose-response pattern well enough and the predicted standard deviation
was reasonably close to the observed standard deviation for the control group, which would make
that model acceptable for BMD estimation. The BMD estimate was 14 mg/kg/day, but the Hill
model failed to derive a BMDL estimate. As an alternative, the high dose group was removed
and the dose-response models were refit to the reduced data set. Without the highest group,
BMDS suggests that a constant variance model is appropriate. The polynomial and power models
still defaulted to linearity, and still did not fit the data (p = 0.002). However, the Hill model
passes exactly through all of the observed means and yields a BMD and BMDL of 13 and 11
mg/kg/day, respectively. These estimates are reasonable values to use for this data set, given the
similarity of the BMD estimate for the truncated data set as compared to that of the Hill model fit
to all the doses.

For the male rats exposed to DCAN, a monotonic dose-response pattern was observed for
the relative liver weight, but again the responses tended to plateau at the top two dose levels. The
best estimates that the polynomial or power models can produce in such cases are based on a

2 A bug in BMDS resulted in three different values for the log-likelihood of model A3
(independent means but modeled variances) across the three model runs. The fit of model A3
should be the same regardless of the model being fit, so it is not possible to know with certainty
what the correct likelihood for A3 is. In all cases, however, it was significantly worse than that
for model A2 (independent means and independent variances).

EPA/OW/OST/HECD A-11 Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

linear fit which, in this case, severely over-estimated the response observed at the high dose and
underestimated the response at the penultimate dose. The Hill model provides an excellent fit to
the data, but failed to estimate a BMDL estimate to go with the BMD estimate of 8.4 mg/kg/day.
For consistency, the male data were also considered with the high dose group ignored (as was
done for the females exposed to DCAN), The results of analyzing the reduced male data set were
as follows. The linear model to which the polynomial and power models defaulted provided an
adequate fit to the data (p = 0.14). The Hill model also fit the data pretty well (p = 0.06), but
because it required additional parameters, its AIC value was slightly greater than that for the
linear model. The elimination of the highest dose resulted in an acceptable linear model fit but
reduced the BMD to 7.8 mg/kg/day, which was similar to the BMD from the Hill model that fit
the complete data set very well. Given that similarity, that BMD and the associated BMDL of 5.0
mg/kg/day from the linear model fit to the reduced data set are considered to be adequate values
to characterize this data set.

In summary, for the Ten-day HA for DCAN, the modeling results for the endpoints of
decreased body weight and increased relative liver weight (Hayes et al., 1986) were considered
for use in the quantitative dose-response assessment. Changes in relative liver weight in males
was the most sensitive endpoint. The estimate of choice for the short-term data sets for DCAN
was the BMD of 7.8 mg/kg/day with the corresponding BMDL of 5.0 mg/kg/day obtained from
the power and polynomial models for increased relative liver weight in males.

EPA/OW/OST/HECD

A-12

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

TCAN

For TCAN, the developmental study by Christ et al. (1996) was selected as the most
appropriate basis for deriving the Ten-day HA. A number of maternal and developmental
parameters were affected by TCAN. Comparison of NOAELs across these endpoints suggested
that adjusted maternal body weight gain was the most sensitive effect. Therefore, BMD modeling
results are described in detail for this endpoint. Results of modeling with BMDS suggested a
constant-variance approach. The polynomial model defaulted to a linear fit, which was very good
(p = 0.70). The power model estimated a power just greater than 1 (1.04), and so was not
exactly linear. The power model fit was not substantially better than the polynomial model, so the
AIC for the power model was greater than that for the linear model (because of the extra
parameter). Similarly, the Hill model used extra parameters to achieve a marginally better fit to
the data; it too yielded an AIC greater than that for the linear (polynomial) model. Since the extra
parameters resulted in only marginal improvements in fit, the linear model estimates derived from
the polynomial model run are the preferred ones, with a BMD of 21 mg/kg/day and a BMDL of
17 mg/kg/day.

Visual inspection of the data for other maternal or developmental endpoints affected by
TCAN in this study suggested that the critical effect levels for these additional endpoints would
likely be greater than for adjusted maternal body weight gain. To verify this, BMD modeling was
performed for the following endpoints: maternal relative liver weight, post-implantation loss, live
fetuses/litter, male and female fetal body weight, male and female crown-rump length, and
incidence of external malformations. Best-fit BMDL estimates for all of these endpoints were

EPA/OW/OST/HECD

A-13

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

greater than the BMDL of 17 mg/kg/day for maternal adjusted weight gain, and thus these other
endpoints would not be appropriate as the basis for deriving the Ten-day HA. For this reason, we
do not describe these additional modeling results in detail here, but the output from BMDS for
these endpoints are presented in Appendix B.

In summary, for the TCAN Ten-day HA modeling results for numerous maternal and
developmental endpoints reported in Christ et al. (1996) were considered for use in the
quantitative dose-response assessment. The most appropriate endpoint to serve as the basis for
the Ten-day HA was adjusted maternal body weight gain. The BMD was 21 mg/kg/day with a
corresponding BMDL of 17 mg/kg/day, derived using the linear model estimates from the
polynomial model.

EPA/OW/OST/HECD

A-14

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Table A-l Benchmark Dose Modeling Results for DBAN, DCANa, and TCANb

Endpoint and Model

AIC

P-value

BMDC

BMDL

DBAN - Body Weight Males

Hill

323.095

<0.001d

Failed6

Failed6

Power

279.347

0.07

26

16

Polynomial

284.303

0.004

30

15

DBAN - Relative Liver Weight Females

Hill

-11.853

0.52d

32

Failed

Power

-13.885

0.45

31

17

Polynomial

-14.037

0.53

31

17

DCAN - Body Weight Males

Hill

343.446

0.01

32

22

Power

343.290

0.02

36

25

Polynomial

343.538e

0.02

33

25

DCAN - Relative Liver Weight Females

Hill

61.811

0.06

14

Failed

Power

65.850

0.04e

13

8.4

Polynomial

65.582

0.01e

13

8.4

DCAN - Relative Liver Weight Females (without highest dose)

Hill

15.440

1.0f

13

11

Power

24.058e

0.002e

15

11

Polynomial

24.058e

0.002e

15

11

DCAN - Relative Liver Weights Males

Hill

10.291

0.93

8.4

Failed

Power

26.962

0.0001

16

8.4

Polynomial

26.962

0.0001

16

8.4

DCAN - Relative Liver Weight Males (without highest dose)

Hill

4.948

0.06d

8.4

Failed

EPA/OW/OST/HECD

A-15

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Endpoint and Model

AIC

P-value

BMDC

BMDL

Power

4.859e

0.14e

7.8

5.0

Polynomial

4.859

0.14e

7.8

5.0

TCAN - Adjusted % Maternal Weight Gain

Hill

320.867

0.89d

Power

319.562e

0.40

Polynomial

317.5826

0.70e

11 Modeling for BCAN and DCAN based on data from Hayes et al. (1986).

b Adjusted % Weight Gain for TCAN based on data from Christ et al. (1996).

c BMD and BMDL are based on benchmark response of 1SD. Results are presented in units of mg/kg/day.
BMD and BMDL estimates in bold type are the estimates judged to be the best estimates to use for the
quantitative dose-response assessment for each chemical. Failed indicates that BMDS was unable to produce
the estimate or the information required to be able to present a value.

d Based on a comparison of the fitted model to the model maximizing the likelihood (i.e. model with
independent means and variances for each dose group, model A2 from BMDS).

e Corrected from erroneous BMDS output. Errors were identified in the degrees of freedom (DF) provided in
the output for the fitted model in several cases. For these cases, the AIC was calculated independently using the
log likelihoods provided in the output and the correct number of DF. Similarly, the goodness-of-fit p-values
were corrected by calculating manually the chi square p-value using the appropriate number of DF.

f A fit that maximizes the likelihood is assigned a p-value of 1.0, even if there were no degrees of freedom for a
formal statistical test. The maximized likelihood is given by model Al for constant variance models and model
A2 for non-constant variance models. Models Al and A2 are independent of the model chosen to fit the data
(e.g., power, polynomial, Hill model) and provide the best match possible to the mean and standard deviation
for each dose level.

Modeling Results for Longer-term Studies

BMD modeling was done to support the derivation of the Longer-term and Lifetime HAs.
Adequate longer-term studies that would support BMD modeling were available only for DBAN
and DCAN. No studies of suitable duration were available for derivation of Longer-term or Life-
time HAs for BCAN or TCAN. As a result, BMD modeling was not performed for these two

EPA/OW/OST/HECD

A-16

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
HANs. The BMD modeling results for the longer-term studies are presented in Table A-2 and
described below.

I) BAN

As noted in Chapter V, the only endpoint that was considered to be toxicologically
relevant for DBAN in the single available subchronic study (Hayes el al., 1986) was the observed
decrease in body weight in male rats. A constant variance model was appropriate for modeling
the data set. The Hill, polynomial, and power models all gave similar BMD and BMDL estimates.
Visual inspection in the region of the BMDL indicated that the fit was adequate for all three of
these models, and was better for these models than for the linear model. The goodness-of-fit p-
values for the polynomial and power models were very good, and model fit for the Hill model was
not very good. The polynomial model provided a slightly better fit than the polynomial model,
with fewer parameters as indicated by the lower AIC and higher p-value, and the BMD of 29
mg/kg/day and the BMDL of 20 mg/kg/day for the polynomial model was selected as the best
estimate for this data set.

In summary, decreased body weight in males was the only endpoint judged to be of
sufficient toxicological significance to be modeled for DBAN. Therefore, the BMDL of 20
mg/kg/day for this endpoint is the most appropriate basis for the Longer-term and Lifetime HAs.

EPA/OW/OST/HECD

A-17

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

DCAN

Decreased body weight reported in Hayes et al. (1986) was considered to be a
toxicologically-relevant response to DCAN. A constant-variance model was appropriate for
modeling the data set for body weight in males. However, none of the continuous models
provided an adequate visual fit in the dose range of interest. Only the Hill models had an
adequate goodness-of-fit statistic, but this model was not selected due to the model dependence
of the BMDL estimate, which relied on a sigmoid curve that could not be supported based on the
underlying biology. The poor fits were due to the non-monotonic nature of the data; the mean
body weight was higher in the low dose group than in the controls. Therefore, no adequate
BMDL estimate for decreased body weight in males was obtained. Further optimization of the
models for decreased body weight that provided poor fits was not done since it became apparent
in the course of the preliminary modeling that increased relative liver weight would yield
significantly lower BMDLs than decreased body weight, and thus the modeling for body weight
would not drive the assessment.

Initial modeling for body weight in females was conducted separately assuming either
constant or non-constant variance. For all four mathematical models fit to the data, modeling
results using the constant variance model suggested that a non-homogenous variance model
should be used. However, when this was done, the test statistic for the variance model (test 3)
was inadequate. This result is consistent with the absence of a clear dose-dependent (i.e. mean-
dependent) trend in the group standard deviations. The outcome of these modeling efforts
indicates that neither the constant variance nor modeled variance options in BMDS provided an

EPA/OW/OST/HECD

A-18

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles
adequate estimate of the variance around the best fit curve. An adequate estimate of the variances
is critical for relying on the BMDL, since the computation of the BMDL is dependent on the
estimated variance for the fitted model. (As noted above, the BMR was defined as a change in
the mean of one standard deviation, a measure related to the variance.) As a result, even though
the linear, power, and polynomial models yielded adequate goodness-of-fit p-values ranging from
0.11 to 0.22, none of these models was viewed as providing a reliable estimate of the BMDL.

The NOAEL/LOAEL analysis indicated that increased relative liver weight was the most
sensitive indicator of toxicity for DC AN (see discussion in Chapter V). Since increases in liver
weight were observed in both males and females and this effect was judged to be toxicologically
relevant, modeling was performed for the relative liver weight data for both sexes.

The relative liver weight data for males was modeled using a non-constant variance model,
based on visual inspection of the group standard deviations and BMD modeling results in initial
runs. The overall fit to the data achieved using the full data set was inadequate for all of the
continuous models that were run, except the Hill model, which did not calculate a BMDL
estimate. The poor model fits appeared to arise from the inability of these models to
accommodate the plateau in the dose-response curve at high doses. Therefore, consistent with
the approach used for the short-term studies, modeling was done using a truncated data set (i.e.
without the high dose). The Hill model was not run using the truncated data set, since this model
requires at least four data points to calculate a BMDL estimate and the truncated data set
contained only three doses. The linear, power, and polynomial models performed very well when

EPA/OW/OST/HECD

A-19

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

the high dose data were removed, based on visual inspection and review of the chi-square values
for residuals at the individual data points. Although goodness-of-fit statistics were not calculated
by BMDS for the polynomial and power models, independent calculation of the p- values (as
presented in Table A-2) confirms the good statistical fit. Since the BMDL estimates were the
same for all three models, the model with the lower AIC was selected as providing the best
estimate. Therefore, the BMD of 6 mg/kg/day and the BMDL of 4 mg/kg/day for the linear
model were selected as the best estimates for the dose-response assessment.

For modeling of relative liver weight in females, the initial BMDS results indicated that a
non-constant variance model would be most appropriate. However, for all four of the
mathematical models fit to the data, the test statistic for the variance model was inadequate with
this option selected, reflecting the absence of a clear dose-dependent (i.e. mean-dependent) trend.
As described above for modeling of female body weight for DCAN, the inability to get an
appropriate model of the variance precludes identifying the BMDL estimate with confidence.
Regardless of this consideration, none of the modeling results using the full data set provided an
adequate fit to the data. Although the Hill model yielded a curve that went through the points,
the curve had a sigmoid shape that was highly dependent on the model and that could not be
supported based on the underlying biology. Since there was an apparent plateau in the dose-
response data, a truncated data set was modeled with the high dose group removed. This
approach provided an adequate visual fit. However, similar to the modeling with the full data set,
the variance model could not adequately describe the data. Therefore, none of these models was
viewed as providing a reliable estimate of the BMDL.

EPA/OW/OST/HECD

A-20

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

In summary, based on the evaluation of these BMD modeling results for DCAN, increased
relative liver weight in males is the most sensitive endpoint. The BMD of 6 mg/kg/day and the
corresponding BMDL of 4 mg/kg/day for the linear model was selected as the most appropriate
basis for the Longer-term and Life-time HAs.

EPA/OW/OST/HECD

A-21

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Table A-2 Benchmark Dose Modeling Results for DBAN and DCANa

Endpoint and Model

AIC

P-value

BMDb

BMDL

DBAN - Body Weight in Males

Linear

716.297e

0.15

20

16

Hill

716.713

0.06c

30

21

Power

714.705

0.61

30

20

Polynomial

714.68 le

0.63

29

20

DCAN - Body Weight in Males

Linear

777.801e

0.022

17

15

Hill

775.374

0.12c

32

21

Power

776.405

0.028

25

18

Polynomial

777.462e

0.015

24

16

DCAN - Body Weight in Femalesd

Linear

717.62 le

0.22

46

34

Hill

721.120

<0.001c

55

31

Power

719.090

0.11

55

35

Polynomial

718.907e

0.13

52

33

DCAN - Liver Weight in Males

Linear

117.429

<0.0001

7

5

Hill

118.869

0.59c

8

Failed

Power

117.429e

0.02e

7

5

Polynomial

117.974

<0.0001

6

4

DCAN - Liver Weight in Males (without highest dose)

Linear

37.993

0.44

6

4

Power

39.388

1.0f

7

4

Polynomial

39.388

1.0f

7

4

DCAN - Liver Weight in Femalesd

Linear

86.720e

0.00039

32

25

EPA/OW/OST/HECD

A-22

Final Draft

-------
Drinking Water Criteria Document for Haloacetonitriles

Endpoint and Model

AIC

P-value

BMDb

BMDL

Hill

75.031

1.0f

Power

86.720e

0.0004e

Polynomial

86.720e

0.0004e

DCAN - Liver Weight in Females (without highest dose)d

Linear

64.218e

0.50

Power

64.218e

0.50

Polynomial

65.15T

1.0f

" Modeling was performed based on body and organ weight at terminal sacrifice in the subchronic study by
Hayes et al. (1986).

b BMD and BMDL are based on benchmark response of 1SD. Results are presented in units of mg/kg/day.
BMD and BMDL estimates in bold type are the estimates judged to be the best estimates to use for the
quantitative dose-response assessment for each chemical. Failed indicates that BMDS was unable to produce
the estimate or the information required to be able to present a value.

c Based on a comparison of the fitted model to the model maximizing the likelihood (i.e. model with
independent means and variances for each dose group, model A2 from BMDS).

dThe modeling results for DCAN shown here are for the constant variance model. Neither the constant nor
non-constant variance models yielded a reliable BMDL estimate.

EPA/OW/OST/HECD

A-23

Final Draft

-------