United States
Environmental Protection
1=1 m m Agency
EPA/690/R-09/046F
Final
9-08-2009
Provisional Peer-Reviewed Toxicity Values for
Phenyl isothiocyanate
(CASRN 103-72-0)
Superfund Health Risk Technical Support Center
National Center for Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
-------
COMMONLY USED ABBREVIATIONS
BMD
Benchmark Dose
IRIS
Integrated Risk Information System
IUR
inhalation unit risk
LOAEL
lowest-observed-adverse-effect level
LOAELadj
LOAEL adjusted to continuous exposure duration
LOAELhec
LOAEL adjusted for dosimetric differences across species to a human
NOAEL
no-ob served-adverse-effect level
NOAELadj
NOAEL adjusted to continuous exposure duration
NOAELhec
NOAEL adjusted for dosimetric differences across species to a human
NOEL
no-ob served-effect level
OSF
oral slope factor
p-IUR
provisional inhalation unit risk
p-OSF
provisional oral slope factor
p-RfC
provisional inhalation reference concentration
p-RfD
provisional oral reference dose
RfC
inhalation reference concentration
RfD
oral reference dose
UF
uncertainty factor
UFa
animal to human uncertainty factor
UFC
composite uncertainty factor
UFd
incomplete to complete database uncertainty factor
UFh
interhuman uncertainty factor
UFl
LOAEL to NOAEL uncertainty factor
UFS
subchronic to chronic uncertainty factor
1
-------
FINAL
9-8-2009
PROVISIONAL PEER-REVIEWED TOXICITY VALUES FOR
PHENYL ISOTHIOCYANATE (CASRN 103-72-0)
Background
On December 5, 2003, the U.S. Environmental Protection Agency's (U.S. EPA's) Office
of Superfund Remediation and Technology Innovation (OSRTI) revised its hierarchy of human
health toxicity values for Superfund risk assessments, establishing the following three tiers as the
new hierarchy:
1. U.S. EPA's Integrated Risk Information System (IRIS).
2. Provisional Peer-Reviewed Toxicity Values (PPRTVs) used in U.S. EPA's Superfund
Program.
3. Other (peer-reviewed) toxicity values, including:
~ Minimal Risk Levels produced by the Agency for Toxic Substances and Disease
Registry (ATSDR),
~ California Environmental Protection Agency (CalEPA) values, and
~ EPA Health Effects Assessment Summary Table (HEAST) values.
A PPRTV is defined as a toxicity value derived for use in the Superfund Program when
such a value is not available in U.S. EPA's IRIS. PPRTVs are developed according to a
Standard Operating Procedure (SOP) and are derived after a review of the relevant scientific
literature using the same methods, sources of data, and Agency guidance for value derivation
generally used by the U.S. EPA IRIS Program. All provisional toxicity values receive internal
review by two U.S. EPA scientists and external peer review by three independently selected
scientific experts. PPRTVs differ from IRIS values in that PPRTVs do not receive the
multiprogram consensus review provided for IRIS values. This is because IRIS values are
generally intended to be used in all U.S. EPA programs, while PPRTVs are developed
specifically for the Superfund Program.
Because new information becomes available and scientific methods improve over time,
PPRTVs are reviewed on a 5-year basis and updated into the active database. Once an IRIS
value for a specific chemical becomes available for Agency review, the analogous PPRTV for
that same chemical is retired. It should also be noted that some PPRTV documents conclude that
a PPRTV cannot be derived based on inadequate data.
Disclaimers
Users of this document should first check to see if any IRIS values exist for the chemical
of concern before proceeding to use a PPRTV. If no IRIS value is available, staff in the regional
Superfund and Resource Conservation and Recovery Act (RCRA) program offices are advised to
carefully review the information provided in this document to ensure that the PPRTVs used are
appropriate for the types of exposures and circumstances at the Superfund site or RCRA facility
in question. PPRTVs are periodically updated; therefore, users should ensure that the values
contained in the PPRTV are current at the time of use.
2
-------
FINAL
9-8-2009
It is important to remember that a provisional value alone tells very little about the
adverse effects of a chemical or the quality of evidence on which the value is based. Therefore,
users are strongly encouraged to read the entire PPRTV document and understand the strengths
and limitations of the derived provisional values. PPRTVs are developed by the U.S. EPA
Office of Research and Development's National Center for Environmental Assessment,
Superfund Health Risk Technical Support Center for OSRTI. Other U.S. EPA programs or
external parties who may choose of their own initiative to use these PPRTVs are advised that
Superfund resources will not generally be used to respond to challenges of PPRTVs used in a
context outside of the Superfund Program.
Questions Regarding PPRTVs
Questions regarding the contents of the PPRTVs and their appropriate use (e.g., on
chemicals not covered, or whether chemicals have pending IRIS toxicity values) may be directed
to the U.S. EPA Office of Research and Development's National Center for Environmental
Assessment, Superfund Health Risk Technical Support Center (513-569-7300), or OSRTI.
INTRODUCTION
Some isothiocyanates, such as sulforaphane, are natural products that have been shown to
inhibit carcinogenesis and tumorigenesis (for example, Talalay et al., 2007). However, other
isothiocyanates have less desirable properties. A chronic reference dose (RfD) for phenyl
isothiocyanate is not available in the U.S. Environmental Protection Agency's (EPA) Integrated
Risk Information System (IRIS; U.S. EPA, 2009), the Drinking Water Standards and Health
Advisories list (U.S. EPA, 2006), or the Health Effects Assessment Summary Tables (HEAST;
U.S. EPA, 1997). The Chemical Assessments and Related Activities (CARA) list (U.S. EPA,
1991, 1994) includes no documents for phenyl isothiocyanate. A toxicological review of phenyl
isothiocyanate is not available from the Agency for Toxic Substances and Disease Registry
(ATSDR, 2007) or the World Health Organization (WHO, 2007).
No chronic inhalation reference concentration (RfC) is available for phenyl
isothiocyanate on IRIS (U.S. EPA, 2009) or in the HEAST (U.S. EPA, 1997). The American
Conference of Governmental Industrial Hygienists (ACGIH, 2006), Occupational Safety and
Health Administration (OSHA, 2007), and the National Institute for Occupational Safety and
Health (NIOSH, 2007) have not established occupational health standards for phenyl
isothiocyanate.
A carcinogenicity assessment for phenyl isothiocyanate is not available on IRIS
(U.S. EPA, 2009) or in the HEAST (U.S. EPA, 1997). Phenyl isothiocyanate has not been
evaluated by the International Agency for Research on Cancer (IARC, 2007) nor is it included in
the National Toxicology Program's (NTP) 11th Report on Carcinogens (NTP, 2005).
Literature searches were conducted for studies relevant to the derivation of provisional
toxicity values for phenyl isothiocyanate in July 2007 in MEDLINE, TOXLINE special, and
DART/ETIC (1960s-July 2007); BIOSIS (2000-June 2007); TSCATS/TSCATS2, RTECS,
CCRIS, HSDB, and GENETOX (not date limited); and Current Contents
(September 2008-May 2009). The literature review was subsequently updated to May 2009.
3
-------
FINAL
9-8-2009
REVIEW OF PERTINENT DATA
Human Studies
No studies were located regarding the effects of subchronic or chronic exposure of
humans to phenyl isothiocyanate by oral or inhaled routes.
Animal Studies
Oral Exposure
No studies were located regarding the effects of long-term oral exposure to phenyl
isothiocyanate in animals.
In a 4-week study using SPF Wistar rats of the Riv:Tox [M] strain, Speijers et al. (1985)
administered phenyl isothiocyanate (99% pure) in sunflower seed oil by gavage to groups of
6 male rats on 5 days/week. Doses of 0, 2.5, 10, or 40 mg/kg-day were administered using a
constant intubation volume of 1 ml sunflower seed oil per 100 g body weight. The rats were
housed two to a cage and allowed free access to food (semi-purified diet with an iodine content
of 30 mg/kg diet) and water. Rat body weights were measured at the beginning of the
experiment and weekly thereafter. Initial body weights ranged from 30 to 50 g. Blood was
collected for hematology (erythrocyte count, leukocyte count [total and differential], packed cell
volume, mean corpuscular volume [MCV], mean corpuscular hemoglobin [MCH], mean
corpuscular hemoglobin concentration [MCHC]) and clinical chemistry (alanine
aminotransferase [ALT], aspartate aminotransferase [AST], and thyroxine [T4, total and free]).
The heart, liver, spleen, kidneys, thyroid, adrenals, and mesenteric lymph nodes were weighed
and examined histopathologically.
Terminal body weights of the rats in the highest-dose group (40 mg/kg-day) were
reduced approximately 11% compared to controls; however, these results are not statistically
significant. Food and water consumption data are not reported. Relative heart, liver, kidney, and
adrenal weights were significantly increased in the highest-dose group compared to control
values. The study authors suggested that the increases in these relative weights were probably
due to the lower body weights of the rats in this group. No changes in absolute weights of these
organs were noted, and no histopathological changes were observed. Hematological analysis
found that rats in the 10 mg/kg-day exposure group showed an increase in packed cell volume
(p < 0.05), while rats in the 40 mg/kg-day exposure group showed a decrease in packed cell
volume (p < 0.05). Erythrocyte and leukocyte counts and MCV among treated rats did not differ
significantly from controls, although there were significant increases in MCH (p < 0.05) and
MCHC (p < 0.01) in the highest-dose group. ALT and AST analyses demonstrated no
statistically significant changes in the treated groups. There were statistically significant
decreases in total serum T4 levels (at 40 mg/kg-day,/? < 0.01) and free serum T4 levels (at
10 mg/kg-day,/? < 0.05; at 40 mg/kg-day,/? < 0.001) following phenyl isothiocyanate exposure
(data presented in Table 1). The study authors suggested that the dose-related decreases in
T4 values indicate a potential hypothyroid effect. In general, because the free T4 assay is not
affected by serum protein levels, it is considered a more accurate reflection of thyroid hormone
function. While the decreased thyroid function was not found to be associated with an increase
in thyroid weight or with morphological changes in the thyroid, the study authors stated that the
change in the T4 concentration was the most sensitive indicator of an effect in this experiment.
4
-------
FINAL
9-8-2009
Table 1. Total and Free T4 Levels in Serum from Male Rats Given Phenyl Isothiocyanate
by Gavage 5 Days/Week for 4 Weeks3
Dose
Total T4 (nmol/liter)
Free T4 (pmol/liter)
(mg/kg-day)
No. of rats
Mean
SD
Mean
SD
0
6
124
21
27
4
2.5
6
123
33
25
5
10
6
113
14
22b
2
40
6
72c
25
16d
4
aSpeijers et al. (1985)
hp < 0.05
cp < 0.01
dp< 0.001
Other isothiocyanates have also been shown to affect thyroid function. In a short-term
gavage study in rats, allyl isothiocyanate affected thyroid function and produced antithyroid
effects in vitro (Langer and Greer, 1968, 1977). Methyl isothiocyanate has also been reported to
have antithyroid effects in vivo and in vitro (Langer and Greer, 1977).
Eastman Kodak Co. (1979) administered 0, 10, or 100 mg/kg-day phenyl isothiocyanate
(purity not reported) in corn oil by gavage to groups of five male rats (strain not reported) for up
to 15 days. As the rats in the high-dose group demonstrated weakness and poor general
condition, they either died or were terminated after 4 days. Blood was collected for hematology
(hemoglobin, hematocrit, white blood cell counts, and differential cell counts) and clinical
chemistry (glutamic oxaloacetic transaminase [SGOT = AST], glutamic pyruvic transaminase
[SGPT = ALT], lactic dehydrogenase [LDH], alkaline phosphatase, urea nitrogen, and glucose).
The liver and kidneys from rats in the low-dose group were weighed. A list of specific tissues
examined microscopically was not provided; however, it is apparent that the liver, kidneys, and
thymus were examined. No statistical analysis of the results was conducted.
Rats in the high-dose group demonstrated a severe depression in both feed intake and
body weight gain. Clinical signs included rough coats, pallor, ataxia, tremors, hypothermia, and
bloody urine. Differential blood counts showed relative neutrophilia and red blood cell
poikilocytosis. Serum LDH was "moderately" elevated and ALT was "greatly" elevated (no
quantitative data provided). Other hematological and clinical chemistry tests were not
performed, and relative organ weights were not determined for the 100 mg/kg-day group.
Histopathology revealed thymic necrosis in 3/5 rats in this group. No significant changes from
controls were reported in rats administered 10 mg/kg-day phenyl isothiocyanate.
In additional single-dose testing, groups of five male rats each received 100, 500, or
1000 mg/kg-day phenyl isothiocyanate by gavage for 1 day (Eastman Kodak Co., 1979).
Controls received 1000 mg/kg distilled water. Within 24 hours, animals in the high-dose group
developed rough hair coats, weakness, and tremors; all but one rat died. Animals in the low-dose
5
-------
FINAL
9-8-2009
group showed weakness and rough coats. The study authors reported the approximate oral LD50
(observation period not reported) for phenyl isothiocyanate as 141 mg/kg-day in male rats and
83 mg/kg-day in male mice; however, details of the mouse experiments were not provided.
Chung et al. (1984) fed groups of male F344 rats (3 to 4 per treatment group), initially
weighing 200 to 300 g each, NIH-07 diets for 2 weeks. Treated rats were maintained on the
NIH-07 diet containing 0.003 mmol phenyl isothiocyanate/g of diet. This concentration was
determined following initial range-finding experiments. Exposure to diets containing
0.03 mmol/g of diet demonstrated considerable weight loss (5 to 50 g). Control rats were fed
with the NIH-07 diet only. On average, the rats consumed about 15 g of diet per day. The study
authors reported that the total dose of phenyl isothiocyanate at which decreased body weight
(0.03 mmol/g-diet) was seen was about 21 mmol/kg of body weight. Based on the molecular
weight of phenyl isothiocyanate (135.19 g), this is equivalent to 2830 mg/kg body weight. The
animals were sacrificed on Day 15, and the esophagus and liver were removed and examined for
effects on nitrosamine metabolism. Pretreatment with phenyl isothiocyanate inhibited the
formation of alpha-hydroxylation products of N-nitrosopyrrolidine (NPYR) and
N'-nitrosonornicotine (NNN). In a follow-up study, Chung et al. (1985) fed groups of male
F344 rats (5 rats) NIH-07 diets for 2 weeks containing 0.003-mmol phenyl isothiocyanate/g of
diet. Following the 2-week feeding, the rats were administered N-nitrosodimethylamine
(NDMA) by i.p. injection (25 mg/kg in 0.9% NaCl w/v) or 4-(methylnitrosamino)-l-(3-pyridyl)-
1-butanone (NNK) (85 mg/kg in 0.9% NaCl w/v) by tail-vein injection. The rats were sacrificed
4 hours after injection, and their livers were removed. Pretreatment with phenyl isothiocyanate
caused a marked decrease in NDMA and NNK demethylation. The study authors suggested that
these results are indicative that phenyl isothiocyanate may help inhibit the carcinogenic activity
of NDMA and NNK, but further analysis is needed.
Becker and Plaa (1965) conducted an acute study of the lethality and hepatotoxicity of
single oral doses of phenyl isothiocyanate and other organic isothiocyanates in male Swiss
Webster mice. Only phenyl isothiocyanate was studied in any detail. Groups of 10 mice each
were treated with a single dose of phenyl isothiocyanate dissolved in corn oil and administered
(0.01 ml/g) by gavage. Most deaths from phenyl isothiocyanate occurred during the first
48 hours, although some mice died on the 3rd and 4th days. The total duration of observation was
not reported. The acute lethality study resulted in an LD50 value for 24 hours of 400 mg/kg
(95%) confidence limits: 282-570 mg/kg) and an LD50 value for 48 hours of 350 mg/kg
(95%o confidence limits: 246-497 mg/kg). These values were not statistically different. In
addition, phenyl isothiocyanate caused increases in plasma bilirubin and sulfobromophthalein
retention. Histopathology of livers from mice dosed with 300 mg/kg phenyl isothiocyanate
demonstrated irregularly distributed necrosis of the larger intrahepatic bile duct epithelia. The
study authors also noted some necrosis of the bile ductile epithelium and perivascular
inflammation.
Rothkopf-Ischebeck (1978) exposed male and female Wistar rats orally to phenyl
isothiocyanate dissolved in Neobee-0 oil (5 ml/kg volume) to investigate the effect of phenyl
isothiocyanate shown to induce profuse pleural cavity exudation (pleurisy) in a prior metabolism
study. A number of experiments were performed to determine an ED50, an LD50, a time course
of phenyl isothiocyanate, the age and weight dependency, and any antagonizing effects, each of
which was based on varying durations of exposure. The number of rats in each treatment group
was not reported. The ED50 values for pleurisy in male and female Wistar rats aged 88 and
6
-------
FINAL
9-8-2009
98 days, respectively, were 73.8 mg/kg (65.6-82.4) for males and 63.9 mg/kg (54.1-74.4) for
females. The LD50 values (14-day observation) for male and female rats aged 81 and 72 days,
respectively, were 110 mg/kg (99.9-121) for males and 86.9 mg/kg (78.2-96) for females. Both
male and female rats died between the 48th and 56th hours after provocation during the maximum
exudation period. The exudate volume produced by the same mg/kg dose of phenyl
isothiocyanate progressively increased with the age of the rats. Gross examination revealed no
changes in the kidneys, liver, heart, and spleen, but the examination did show tissue irritation of
the intestinal region at doses >100 mg/kg. Sporadic maculae were seen on the lungs of rats that
received lethal doses, and the frequency of thymus swelling increased with increased exudate
volumes. Histopathological examinations were not performed. Pleurisy was not induced in mice
or guinea pigs by oral phenyl isothiocyanate.
Inhalation Exposure
No data regarding the toxicity of phenyl isothiocyanate in animals following inhalation
exposure were located.
Other Studies
Dermal Studies
The results of the dermal studies indicate that phenyl isothiocyanate is readily absorbed,
highly irritating, and a sensitizer, although there is some inconsistency in results. Application of
phenyl isothiocyanate to the abdomen of guinea pigs under an impervious cuff for 24 hours
resulted in severe irritation. Death occurred within 24 hours in guinea pigs receiving doses of
5 ml/kg or higher on the skin. A single uncovered application of 0.5 ml to the clipped backs of
guinea pigs resulted in the death of five guinea pigs tested within 24 hours. Repeated application
of 0.1 ml/day in the same manner severely irritated the skin and resulted in either weight loss or
no weight gain. A skin sensitization test in guinea pigs gave a moderate response in four of the
five test animals and a weak response in one animal (Eastman Kodak Co., 1979).
In other studies, phenyl isothiocyanate was classified as a "moderate sensitizer" in the
guinea pig maximization test (Fregert et al., 1983), but, in another sensitization test, it did not
elicit a significant swelling response in mouse ears (Schmidt and Chung, 1993).
Parenteral Studies
In a screening developmental toxicity study of 48 compounds, phenyl isothiocyanate was
administered subcutaneously, through dissolution in dimethyl sulfoxide, on Gestation Days (GD)
6-14 at 48 mg/kg-day to C3H mice (3 litters) and at 25 mg/kg-day to BL6 mice (2 groups of
6 litters) and on GDs 6-15 at 25 mg/kg-day to AKR mice (12 litters) (Bionetics Research Labs,
1968). The numbers of litters in the corresponding control groups were 12 (C3H), 73 and
75 (BL6), and 33 (AKR). The number of dams was not specified. Maternal body weights were
unaffected by phenyl isothiocyanate. Fetal mortality was normal in all groups. In one group of
BL6 mice, the number of implantations per litter was increased, mean fetal weight was
decreased, and the incidence of abnormalities (as percent fetuses) was increased. In the other
group of BL6 mice, the number of implantations and live fetuses per litter were increased, but
fetal weight and the incidence of abnormalities were unaffected. In the AKR mice, the number
of live fetuses per litter and mean fetal weight were increased, but implantations per litter and
incidences of abnormalities were unaffected. The study authors stated that they had no
resolution for these inconsistencies.
7
-------
FINAL
9-8-2009
Tumor Inhibition
Many citations identified in the literature searches were studies investigating the potential
effects of phenyl isothiocyanate on tumor inhibition. The database demonstrates that phenyl
isothiocyanate may be protective in some systems but not in others. For example, phenyl
isothiocyanate inhibits melanoma formation (Manesh and Kuttan, 2003), mammary tumor
formation (Wattenberg, 1977), and tumor-specific angiogenesis (Thejass and Kuttan, 2007) in
mice. In the latter study, the study authors demonstrated that phenyl isothiocyanate
down-regulated serum nitrous oxide (NO) and tumor necrosis factor-alpha (TNF-a), both of
which are involved in angiogenesis (the process that is critical to the transformation of
premalignant lesions to the malignant phenotype). However, other studies with prostate cancer
cells (Xiao et al., 2004), HeLa cells (Yu et al., 1998), squamous cell carcinoma (Lui et al., 2003),
and lung cancer bioassays with mice (Morse et al., 1989, 1991; Smith et al., 1990;
Thomson et al., 2006), demonstrate that phenyl isothiocyanate does not induce apoptosis or
inhibit tumor formation processes.
Carcinogenicity
Musk and Johnson (1993) demonstrated the genotoxic potential of phenyl isothiocyanate
in comparison to three other isothiocyanates (i.e., allyl-, benzyl-, and phenethyl isothiocyanate).
Phenyl isothiocyanate induces chromosomal aberrations (chromatid gaps, breaks, and
rearrangements) in an SV/w-transformed Indian muntjac cell line (SVM) in the absence of an
exogenous metabolic activation system. Phenyl isothiocyanate was one order of magnitude less
cytotoxic than the other three isothiocyanates.
DERIVATION OF PROVISIONAL SUBCHRONIC AND CHRONIC ORAL RfD
VALUES FOR PHENYL ISOTHIOCYANATE
Subchronic p-RfD
The longest duration study based on oral exposure to phenyl isothiocyanate is the 4-week
study by Speijers et al. (1985). Results of this study indicate that interference with thyroid
hormone levels in rats treated orally with >10 mg/kg-day phenyl isothiocyanate production is a
sensitive effect of this chemical. Eastman Kodak Co. (1979) found no effects in rats treated
orally with 10 mg/kg-day phenyl isothiocyanate for 15 days, but this study did not specifically
evaluate for thyroid effects.
The Speijers et al. (1985) study was selected as the basis for the subchronic p-RfD. The
critical effect in this study is a decrease in thyroid function as measured by significantly
decreased serum T4 (total and free) levels. A LOAEL of 10 mg/kg-day was identified for the
most sensitive effect, decreased free T4 serum levels, with a corresponding NOAEL of
2.5 mg/kg-day. Models for continuous variables in U.S. EPA's Benchmark Dose Software
(BMDS) version 2.1 beta were fit to the total and free serum T4 data (see Table 1) in accordance
with U.S. EPA (2000) methodology. It was considered whether data are available to establish a
biologically based benchmark response (BMR) for changes in serum T4. The reference range for
T4 in adult male Wistar rats is approximately 2.5-7.0 (J,g/dL (32-90 nmol/L) (DePaolo and
Masoro, 1989). However, T4 levels in the control and low-dose rats in the Speijers et al. (1985)
study were considerably higher (124 ± 21 nmol/L), suggesting that the reference range presented
above is not applicable. In humans, there is considerable variation in serum T4 measured
8
-------
FINAL
9-8-2009
depending on methodology and population tested, and, as a result, it is recommended that
reference ranges be established by the analytic laboratory (Leise and Sibilia, 1993). Similarly, in
animals, serum T4 levels are known to be influenced by strain, sex, age, circadian rhythms, room
temperature, stress, and activity level (Christian and Trenton, 2003). Therefore, in accordance
with U.S. EPA (2000) guidelines and in the absence of any cogent basis for selecting a BMR for
the total and free T4 data, a BMR of 1 standard deviation (SD) from the control mean can be
used, as recommended as the standardized reporting level for comparisons for continuous data.
BMD modeling was performed using the doses administered in the study—not duration-adjusted
average daily doses. For total serum T4 data, the Linear model provides the best model fit, and
the estimated BMDisd and BMDLisd for total T4 are 16.66 and 11.68 mg/kg-day, respectively.
For free serum T4 data, the Hill model estimates a BMDisd and BMDLisd of 6.46 and
2.33 mg/kg-day, respectively. Details of BMD modeling and plots of the best-fitting models are
presented in Appendix B.
Potential points of departure (PODs) for derivation of the subchronic RfD for phenyl
isothiocyanate include the BMDLisd values of 11.68 and 2.33 mg/kg-day for reduced total and
free serum T4, respectively, and the NOAEL value of 2.5 mg/kg-day. Use of BMDL values is
preferred over use of the NOAEL because, in contrast to the NOAEL/LOAEL approach, the
BMD approach uses all of the data in the study, takes into account the shape of the dose-response
curve, and accounts for uncertainty in the observed dose-response due to the experimental
design. Thus, the BMDLisd for the most sensitive endpoint, 2.33 mg/kg-day for free T4, was
selected as the POD for derivation of the subchronic p-RfD.
The 4-week rat study by Speijers et al. (1985) involved phenyl isothiocyanate exposure
by oral gavage 5 days/week; therefore, the BMDLisd for free T4 was duration adjusted to
1.7 mg/kg-day. This duration-adjusted BMDLisd (BMDLisd [adj]) was divided by a composite
UF of 1000 to derive a provisional subchronic RfD for phenyl isothiocyanate, as follows:
Subchronic p-RfD = BMDLisd [adj] UF
= 1.7 mg/kg-day ^ 1000
= 0.002 mg/kg-day or 2 x 10 3 mg/kg-day
The UF of 1000 is composed of the following:
• A UFa of 10 is applied for interspecies extrapolation to account for potential
pharmacodynamic and pharmacokinetic differences between rodents and humans.
• A UFh of 10 is applied for extrapolation to potentially susceptible individuals in
the absence of quantitative information or information on the variability of
response in humans.
• A UFd of 10 is applied to account for database insufficiencies due to the lack of
multigenerational reproduction studies and developmental toxicity studies by oral
exposure.
Confidence in the principal study (i.e., Speijers et al., 1985) is low because, despite
adequate investigation of endpoints at multiple dose levels, only a small number of male rats
were tested and the experiment only lasted 4 weeks. Confidence in the database is low because
9
-------
FINAL
9-8-2009
the data set includes only short-term, subacute, and acute studies. No studies are available on the
potential of ingested phenyl isothiocyanate to induce developmental or reproductive effects.
Low confidence in the subchronic p-RfD follows.
Chronic p-RfD
There are no chronic oral studies available for use in developing a chronic p-RfD for
phenyl isothiocyanate. The BMDLisd value of 2.33 mg/kg-day (for free serum T4 levels)
identified from the Speijers et al. (1985) study and used to derive the subchronic p-RfD above
could also be used to derive a chronic p-RfD for phenyl isothiocyanate. However, in this case,
the composite UF would increase to 10,000. Based on current guidelines and standard operating
procedures, composite UFs >3000 cannot be considered for reference value derivation. As such,
while a chronic p-RfD cannot be derived here, Appendix A of this document contains an oral
"screening value" that may be useful in certain instances. Please refer to Appendix A for details.
FEASIBILITY OF DERIVING PROVISIONAL SUBCHRONIC AND CHRONIC
INHALATION RfC VALUES FOR PHENYL ISOTHIOCYANATE
There are no inhalation studies available for use in developing subchronic and/or chronic
provisional RfCs for phenyl isothiocyanate.
PROVISIONAL CARCINOGENICITY ASSESSMENT FOR
PHENYL ISOTHIOCYANATE
Weight-of-Evidence Descriptor
Studies evaluating the carcinogenic potential of oral or inhalation exposure to phenyl
isothiocyanate in humans are not available in the current literature. A single study on
genotoxicity is, however, available for phenyl isothiocyanate. This study (Musk and Johnson,
2003) demonstrates that phenyl isothiocyanate can induce chromosomal aberrations. Under the
2005 Guidelines for Carcinogen Risk Assessment (U.S. EPA, 2005), "Inadequate Information is
Available to Assess [the] Carcinogenic Potential' of phenyl isothiocyanate.
Quantitative Estimates of Carcinogenic Risk
The lack of suitable data precludes derivation of quantitative estimates of cancer risk for
phenyl isothiocyanate.
REFERENCES
ACGIH (American Conference of Governmental Industrial Hygienists). 2006. 2006 Threshold
Limit Values for Chemical Substances and Physical Agents and Biological Exposure Indices.
ACGIH, Cincinnati, OH.
ATSDR (Agency for Toxic Substances and Disease Registry). 2007. Internet
HazDat-Toxicological Profile Query. Online, http://www.emla.hu/korkep/chems/hazdat.html.
10
-------
FINAL
9-8-2009
Becker, B.A. and G.L. Plaa. 1965. Hepatotoxicity of alpha-naphthylisothiocyanate congeners
with particular emphasis on phenyl isothiocyanate. Toxicol. Appl. Pharmacol. 7(6):804-811.
Bionetics Research Labs. 1968. Evaluation of carcinogenic, teratogenic, and mutagenic
activities of selected pesticides and industrial chemicals. 2. Teratogenic study in mice and rats.
NTIS PB-223 160.
Christian, M.S. and N.A. Trenton. 2003. Evaluation of thyroid function in neonatal and adult
rats: the neglected endocrine mode of action. Pure Appl. Chem. 75(11-12):2055-2068.
Chung, F.L., A. Juchatz, J. Vitarius et al. 1984. Effects of dietary compounds on
alpha-hydroxylation of N-nitrosopyrrolidine and N-nitrosonornicotine in target rat tissues. Cane.
Res. 44:2924-2928.
Chung, F.L., M. Wang and S.S. Hecht. 1985. Effects of dietary indoles and isothiocyanates on
N-nitrosodimethylamine and 4-(methyl-nitrosamino)-l-(3-pyridyl)-l-butanone
alpha-hydroxylation and DNA methylation in rat liver. Carcinogenesis. 6(4):539-543.
DePaolo, L.V. and E.J. Masoro. 1989. Endocrine hormones in laboratory animals. In: The
Clinical Chemistry of Laboratory Animals, W.F. Loeb and F.W. Quimby, Ed. Pergamon Press,
New York. p. 279-308.
Eastman Kodak Co. 1979. Basic toxicity of phenyl isothiocyanate. TSCA 8e submission.
OTS0537528.
Fregert, S., I. Dahlquist and L. Trulsson. 1983. Sensitization capacity of diphenylthiourea and
phenyl isothiocyanate. Cont. Derm. 9(l):87-88.
IARC (International Agency for Research on Cancer). 2007. Search of IARC Monographs.
Online, http://monographs.iarc.fr/.
Langer, P. and M.A. Greer. 1968. Antithyroid action of some naturally occurring
isothiocyanates in vitro. Metabolism. 17:596-605.
Langer, P. and M.A. Greer. 1977. Antithyroid substances and naturally occurring goitrogens:
Chapter 7. Naturally occurring goitrogens. Basel, New York: Karger. p 77-96.
Leise, M.K. and R.A. Sibilia. 1993. Thyroid endocrinology. In: Clinical Chemistry Concepts
and Applications, S.C. Anderson and S. Cockayne, Ed. W.B. Saunders, Philadelphia,
p. 492-521.
Lui, V.W., A. Wentzel, D. Xiao, K. Lew, S. Singh and J. Grandis. 2003. Requirement of a
carbon spacer in benzyl isothiocyanate-mediated cytotoxicity and MAPK activation in head and
neck squamous cell carcinoma. Carcinogenesis. 24(10): 1705-1712.
Manesh, C., and G. Kuttan. 2003. Effect of naturally occurring allyl and phenyl isothiocyanates
in the inhibition of experimental pulmonary metastasis induced by B16F-10 melanoma cells.
Fitoterapia. 74(4):355-363.
11
-------
FINAL
9-8-2009
Morse, M., S. Amin, S. Hecht et al. 1989. Effects of aromatic isothiocyanates on
tumorigenicity, 06-methylguanine formation, and metabolism of the tobacco-specific
nitrosamine 4-(methylnitrosamino)-l-(3-pyridyl)-l-butanone in A/J Mouse Lung. Cane. Res.
49(11):2894-2897.
Morse, M.A., K.I. Eklind, S.S. Hecht and F.L. Chung. 1991. Inhibition of tobacco-specific
nitrosamine 4-(N-nitrosomethylamino)-l-(3-pyridyl)-l-butanone (NNK) tumorigenesis with
aromatic isothiocyanates. IARC Sci Publ. 1991. (105):529-534.
Musk, S.R.R. and I.T. Johnson. 1993. The clastogenic effects of isothiocyanates. Mutat. Res.
300(2): 111-117.
NIOSH (National Institute for Occupational Safety and Health). 2007. NIOSH Pocket Guide to
Chemical Hazards. Online, http://www.cdc.gov/niosh/npe/.
NTP (National Toxicology Program). 2005. 11th Report on Carcinogens. U.S. Department of
Health and Human Services, Public Health Service, National Institutes of Health, Research
Triangle Park, NC. Online, http://ntp.niehs.nih.gov/ntp/roc/tocl 1 .htm.
OSHA (Occupational Safety and Health Administration). 2007. OSHA Standard 1910.1000
Table Z-l. Part Z, Toxic and Hazardous Substances. Online.
https://www.osha.gov/pls/oshaweb/owadisp.show document?p table standards^p id=9992.
Rothkopf-Ischebeck, M. 1978. Phenyl isothiocyanate pleurisy in rats—a model for the
evaluation of anti-exudative substances. Agents Act. 8(6):610-617.
Schmidt, R.J. and L.Y. Chung. 1993. Perturbation of glutathione status and generation of
oxidative stress in mouse skin following application of contact allergenic sesquiterpene lactones
and isothiocyanates. Xenobiotica. 23(8):889-897.
Smith, T.J., Z. Guo, P. Thomas et al. 1990. Metabolism of 4-(methylnitrosamino)-l-(3-pyridyl)-
1-butanone in mouse lung microsomes and its inhibition by isothiocyanates. Cane. Res.
150(21):6817-6822.
Speijers, G.J., L.H. Danse, F.X. van Leeuwen et al. 1985. Four-week toxicity study of phenyl
isothiocyanate in rats. Food Chem. Toxicol. 23(11):1015—1017.
Talalay, P., J. W. Fahey, Z R. Healy, S. L. W eh age, A.L. Benedict, C. Min and
A.T. Dinkova-Kostova. 2007. Sulforaphane mobilizes cellular defenses that protect skin against
damage by UV radiation. PNAS 104(44): 7500-17505.
Thejass, P. and G. Kuttan. 2007. Allyl isothiocyanate (AITC) and phenyl isothiocyanate (C)
inhibit tumor-specific angiogenesis by downregulating nitric oxide (NO) and tumor necrosis
factor-a (TNF-a) production. Nitr. Oxide. 16(2):247-257.
12
-------
FINAL
9-8-2009
Thomson, S.J., K. Brown, J. Pullar et al. 2006. Phenethyl isothiocyanate triggers apoptosis in
jurkat cells made resistant by the over expression of bcl-2. Cane. Res. 66(13):6772-6777.
U.S. EPA. 1991. Chemical Assessments and Related Activities (CARA). Office of Health and
Environmental Assessment, Washington, D.C. April.
U.S. EPA. 1994. Chemical Assessments and Related Activities (CARA). Office of Health and
Environmental Assessment, Washington, D.C. December.
U.S. EPA. 1997. Health Effects Assessment Summary Tables (HEAST). FY-1997 Update.
Prepared by the Office of Research and Development, National Center for Environmental
Assessment, Cincinnati, OH for the Office of Emergency and Remedial Response, Washington,
D.C. July, 1997. EPA/540/R-97/036. NTIS PB97-921199.
U.S. EPA. 2000. Benchmark Dose Technical Guidance Document. EPA/630/R-00/001, October
2000 external review draft.
U.S. EPA. 2005. Guidelines for Cancer Risk Assessment. Risk Assessment Forum,
Washington, DC. EPA/630/P-03/001F. Online, http://www.epa.gov/raf/.
U.S. EPA. 2006. 2006 Edition of the Drinking Water Standards and Health Advisories. Office
of Water, Washington, D.C. Summer 2006. EPA/822/R-06/013. Online.
http://water.epa.eov/action/advisories/drinking/upload/2009 04 27 criteria drinking dwstandar
ds.pdf.
U.S. EPA. 2009. Integrated Risk Information System (IRIS). Online. Office of Research and
Development, National Center for Environmental Assessment, Washington, D.C.
http ://www. epa. gov/iris/.
Wattenberg, L. 1977. Inhibition of carcinogenic effects of polycyclic hydrocarbons by benzyl
isothiocyanate and related compounds. J. Natl. Cane. Inst. 58:395-398.
WHO (World Health Organization). 2007. Online catalogs for the Environmental Health
Criteria Series. Online, http://www.who.int/dsa/cat98/chemtox8.htm#.
Xiao D., C. Johnson, D. Trump et al. 2004. Proteasome-mediated degradation of cell division
cycle 25C and cyclin-dependent kinase 1 in phenethyl isothiocyanate-induced G2-M-phase cell
cycle arrest in PC-3 human prostate cancer cells. Mol. Cane. Ther. 3(5):567-575.
Yu, R., S. Mandlekar, K. Harvey et al. 1998. Chemopreventive isothiocyanates induce
apoptosis and caspase-3-like protease activity. Cane. Res. 58(3):402-408.
13
-------
FINAL
9-8-2009
APPENDIX A. DERIVATION OF A SCREENING VALUE FOR
PHENYL ISOTHIOCYANATE
For reasons noted in the main PPRTV document, it is inappropriate to derive provisional
toxicity values for phenyl isothiocyanate. However, information is available for this chemical
which, although insufficient to support derivation of a provisional toxicity value under current
guidelines, may be of limited use to risk assessors. In such cases, the Superfund Health Risk
Technical Support Center summarizes available information in an Appendix and develops a
"screening value." Appendices receive the same level of internal and external scientific peer
review as the PPRTV documents to ensure their appropriateness within the limitations detailed in
the document. Hazard identification and dose-response information contained in an Appendix
receives the same level of internal and external scientific peer review as the main body of
PPRTV documents, to ensure their appropriateness within the limitations detailed in the
document. In the OSRTI hierarchy, screening values are considered to be below Tier 3, "Other
(Peer-Reviewed) Toxicity Values."
Screening values are intended for use in limited circumstances when no Tier 1, 2, or 3
values are available. Screening values may be used, for example, to rank relative risks of
individual chemicals present at a site to determine if the risk developed from the associated
exposure at the specific site is likely to be a significant concern in the overall cleanup decision.
Screening values are not defensible as the primary drivers in making cleanup decisions because
they are based on limited (e.g., scope, depth, validity, etc.) information. Questions or concerns
about the appropriate use of screening values should be directed to the Superfund Health Risk
Technical Support Center.
Screening Chronic Oral Value
As noted earlier, there are no chronic oral studies available for use in developing a
chronic p-RfD for phenyl isothiocyanate. The longest duration study based on oral exposure to
phenyl isothiocyanate is the 4-week study by Speijers et al. (1985). Thus, in addition to the
aforementioned recommendations on the intended and appropriate use of screening values, it is
important to note that due to the less-than-subchronic 4-week exposure duration of the principal
study (Speijers et al., 1985), the certitude of the chronic oral screening value derived below is
particularly diminished.
The BMDLisd for the most sensitive endpoint from the Speijers et al. (1985) study,
2.33 mg/kg-day for free serum T4, was selected as the POD for derivation of the chronic oral
screening value. This study involved phenyl isothiocyanate exposure by oral gavage
5 days/week; therefore, the BMDLisd for free T4 was duration adjusted to 1.7 mg/kg-day. This
duration-adjusted BMDLisd (BMDLisd [adj]) was divided by a composite UF of 10,000 to derive
a screening chronic oral value for phenyl isothiocyanate, as follows:
Screening Chronic Oral Value = BMDLisd [adj] ^ UF
= 1.7 mg/kg-day ^ 10,000
= 0.0002 mg/kg-day or 2 x 10"4 mg/kg-day
14
-------
FINAL
9-8-2009
The UF of 10,000 was composed of the following UFs:
• A UFa of 10 is applied for interspecies extrapolation to account for potential
pharmacodynamic and pharmacokinetic differences between rodents and humans.
• A UFh of 10 is applied for extrapolation to potentially susceptible individuals in
the absence of quantitative information or information on the variability of
response in humans.
• A UFd of 10 is applied to account for database insufficiencies due to the lack of
multigenerational reproduction studies and developmental toxicity studies by oral
exposure.
• UFs of 10 is applied for using data from less than lifetime exposure to assess
potential effects from chronic exposure.
Confidence in the principal study (i.e., Speijers et al., 1985) is low because, despite
adequate investigation of endpoints at multiple dose levels, only a small number of male rats
were tested and the experiment only lasted 4 weeks. Confidence in the database is low because
the data set includes only short-term, subacute, and acute studies. No studies are available on the
potential of ingested phenyl isothiocyanate to induce developmental or reproductive effects.
Low confidence in the screening chronic oral value follows.
15
-------
FINAL
9-8-2009
APPENDIX B. DETAILS OF BENCHMARK DOSE MODELING
FOR THE PROVISIONAL RfDs
Model Fitting Procedure for Continuous Data
The BMD modeling for continuous data (i.e., reduced total and free serum T4 levels) was
conducted with the U.S. EPA's BMD software (BMDS version 2.1 beta). The original data were
modeled with all the continuous models available within the software employing a BMR of
1 SD. An adequate fit was judged based on three criteria: (1) the goodness-of-fitp-walue
(p > 0.1), (2) magnitude of scaled residuals in the vicinity of the BMR, and (3) visual inspection
of the model fit. In addition to the three criteria forjudging the adequate model fit, whether the
variance needed to be modeled, and if so, how it was modeled also determined final use of the
model results. If a constant variance model was deemed appropriate based on the statistical test
provided in the BMDS (i.e., Test 2), the final BMD results were estimated from a constant
variance model. If the test for constant variance was rejected (p< 0.1), the model was run again
while modeling the variance as a power function of the mean to account for this nonconstant
variance. If this nonconstant variance model did not adequately fit the data (i.e., Test 3;
p-w alue < 0.1), the data set was considered unsuitable for BMD modeling. Among all models
providing adequate fit, the lowest BMDL was selected if the BMDLs estimated from different
models varied >3-fold; otherwise, the BMDL from the model with the lowest Akaike's
Information Criterion (AIC) was selected as a potential POD from which to derive an RfD.
Model Predictions for Total and Free Serum T4 Levels in Male Rats Given Phenyl
Isothiocyanate by Gavage 5 Days/Week for 4 Weeks
Because total and free serum T4 levels were identified as sensitive effects in rats
administered phenyl isothiocyanate by gavage 5 days/week for 4 weeks (Speijers et al., 1985), all
available continuous models in the BMDS (version 2.1 beta) were fit to the total and free serum
T4 data from this study (see Table 1). BMD modeling has been performed using the doses
administered in the study before duration adjustment. A default BMR of 1 SD from the control
mean was used in the BMD modeling because no specific criteria on the magnitude of change of
total and free serum T4 levels that would be considered biologically significant could be
identified.
For total T4, the Linear, Polynomial, Power, and Hill models in the BMDS provide an
adequate fit to the data, and Test 2(p = 0.2363) also indicates that using a constant variance
model is appropriate for modeling the data. Thus, all of the BMD modeling results for total T4
shown in Table B-l were obtained from constant variance models. The Hill model failed to
estimate a goodness-of-fit p-w alue. Based on the goodness-of-fit ^-values and AICs from all
models, the Linear model provided the best to the data (see Table B-l and Figure B-l). The
estimated BMDisd and BMDLisd for total T4 was 16.66 and 11.68 mg/kg-day, respectively.
For free T4, the Linear, Polynomial, Power, and Hill models in the BMDS provided
adequate fit to the data, and Test 2 (p = 0.2134) also indicates that using a constant variance
model is appropriate for modeling the data. Thus, all of the BMD modeling results shown in
Table B-l for free T4 were obtained from constant variance models. Based on the
goodness-of-fit /^-values and AICs from all models, the Linear, Polynomial, and Power models
provided identical fits to the data (see Table B-l). However, because the BMDLisds estimated
16
-------
FINAL
9-8-2009
from the four different models varied >3-fold, the lowest BMDLisd calculated by the Hill model
was selected as the POD (see Table B-l and Figure B-2). Thus, the estimated BMDisd and
BMDLisd for free T4 was 6.46 and 2.33 mg/kg-day, respectively.
Table B-l BMD Modeling Results Based on Total and Free Serum T4 Levels in Rats
Given Phenyl Isothiocyanate by Gavage 5 Days/Week for 4 Weeks"
Model
Test 2
Test 3
Goodness-of-fit />-value
AIC
bmd1sd
BMDL1sd
Total T4
Linearb'c
0.2363
0.2363
0.9779
178.70
16.66
11.68
Polynomial13'0
0.2363
0.2363
0.8867
180.68
18.59
11.70
Power°'d
0.2363
0.2363
0.9066
180.67
18.45
11.70
Hilfd
0.2363
0.2363
NA
182.66
16.22
5.88
Free T4
Linearbc
0.2134
0.2134
0.4828
92.47
14.45
10.40
Polynomial13'0
0.2134
0.2134
0.4828
92.47
14.45
10.40
Power°'d
0.2134
0.2134
0.4828
92.47
14.45
10.40
TTillcd
0.2134
0.2134
0.8159
93.07
6.46
2.33
aSpeijers et al., 1985
bRestrict betas <0.
°Constant variance.
dRestrict power >1.
17
-------
FINAL
9-8-2009
Linear Model with 0.95 Confidence Level
Linear
BMD Lower Bound
160
140
120
100
80
60
40
BMDL
BMD
0
5
10
15
20
25
30
35
40
dose
10:24 05/22 2009
Figure B-l. Dose-Response Modeling of Total Serum T4 in Rats Given Phenyl
Isothiocyanate by Gavage 5 Days/Week for 4 Weeks (Speijers et al., 1985).
18
-------
FINAL
9-8-2009
The BMDs and BMDLs are associated with a change of 1 SD from the control and are in units of mg/kg-day.
Polynomial Model. (Version: 2.13; Date: 04/08/2008)
Input Data File: C:\USEPA\BMDS2lBeta\Data\2LinPITLin.(d)
Gnuplot Plotting File: C:\USEPA\BMDS21Beta\Data\2LinPITLin.plt
Tue May 26 11:34:03 2009
BMDS Model Run
The form of the response function is:
Y[dose] = beta_0 + beta_l*dose + beta_2*dose/s2 + ...
Dependent variable = Mean
Independent variable = Dose
rho is set to 0
The polynomial coefficients are restricted to be negative
A constant variance model is fit
Total number of dose groups = 4
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
Default Initial Parameter Values
alpha = 587.75
rho = 0 Specified
beta_0 = 125.452
beta 1 = -1.32965
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -rho
have been estimated at a boundary point, or have been specified by
the user,
and do not appear in the correlation matrix )
alpha beta_0 beta_l
alpha 1 -3.6e-009 8.2e-010
beta_0 -3.6e-009 1 -0.64
beta 1 8.2e-010 -0.64 1
Parameter Estimates
95.0% Wald Confidence
Interval
Variable Estimate Std. Err. Lower Conf. Limit Upper Conf.
Limit
alpha 490.705 141.654 213.068
768.342
19
-------
beta 0
136.93
beta 1
0.773893
125.452
-1.32965
5.85633
0.283553
113.973
-1.8854
Table of Data and Estimated Values of Interest
Dose
Obs Mean
Est Mean Obs Std Dev Est Std Dev Scaled Res.
0
2.5
10
40
124
123
113
72
125
122
112
72 .3
21
33
14
25
22 .2
22 .2
22 .2
22 .2
-0.161
0.0965
0.0934
-0.0294
Model Descriptions for likelihoods calculated
Model A1: Yij = Mu(i) + e(ij)
Var{e(ij)} = SigmaA2
Model A2: Yij = Mu(i) + e(ij)
Var{e(ij)} = Sigma(i)^2
Model A3: Yij = Mu(i) + e(ij)
Var{e(ij)} = Sigma^2
Model A3 uses any fixed variance parameters that
were specified by the user
Model R: Yi = Mu + e(i)
Var{e(i)} = Sigma^2
Likelihoods of Interest
Model
A1
A2
A3
fitted
R
Log(likelihood)
-86.327762
-84.205920
-86.327762
-86.350115
-94.154281
# Param's
5
8
5
3
2
AIC
182.655523
184.411840
182.655523
178.700229
192.308563
Explanation of Tests
Do responses and/or variances differ among Dose levels?
(A2 vs. R)
Are Variances Homogeneous? (A1 vs A2)
Are variances adeguately modeled? (A2 vs. A3)
Does the Model for the Mean Fit? (A3 vs. fitted)
(Note: When rho=0 the results of Test 3 and Test 2 will be the same.)
Test
1:
Test
2 :
Test
3:
Test
4 :
(Not
e:
Tests of Interest
Test
-2*log(Likelihood Ratio) Test df
p-value
Test
Test
Test
Test
19.8967
4.24368
4.24368
0.044706
0. 002889
0.2363
0.2363
0.9779
20
-------
FINAL
9-8-2009
The p-value for Test 1 is less than .05. There appears to be a
difference between response and/or variances among the dose levels
It seems appropriate to model the data
The p-value for Test 2 is greater than .1. A homogeneous variance
model appears to be appropriate here
The p-value for Test 3 is greater than .1. The modeled variance appears
to be appropriate here
The p-value for Test 4 is greater than .1. The model chosen seems
to adeguately describe the data
Benchmark Dose Computation
Specified effect
1
Risk Type
Estimated standard deviations from the control mean
Confidence level
0.95
BMD
16. 66
BMDL
11.6792
21
-------
FINAL
9-8-2009
Hill Model with 0.95 Confidence Level
Hill
30
25
20
15
BMDL
BMD
10
0
5
10
15
20
25
30
35
40
14:10 05/22 2009
Figure B-2. Dose-Response Modeling of Free Serum T4 in Rats Given Phenyl
Isothiocyanate by Gavage 5 Days/Week for 4 Weeks (Speijers et al., 1985).
22
-------
FINAL
9-8-2009
The BMDs and BMDLs are associated with a change of 1 SD from the control and are in units of mg/kg-day
Hill Model. (Version: 2.14; Date: 06/26/2008)
Input Data File: C:\USEPA\BMDS2lBeta\Data\lHilPITHil.(d)
Gnuplot Plotting File: C:\USEPA\BMDS21Beta\Data\lHilPITHil.plt
Tue May 26 11:35:53 2009
BMDS Model Run
The form of the response function is:
Y[dose] = intercept + v*dose^n/(k^n + dose^n)
Dependent variable = Mean
Independent variable = Dose
rho is set to 0
Power parameter restricted to be greater than 1
A constant variance model is fit
Total number of dose groups = 4
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008
Default Initial Parameter Values
alpha = 15.25
rho = 0 Specified
intercept = 27
v = -11
n = 0.875798
k = 12.5
Asymptotic Correlation Matrix of Parameter Estimates
( *** The model parameter(s) -rho -n
have been estimated at a boundary point, or have been specified by
and do not appear in the correlation matrix )
alpha intercept v k
alpha 1 2.4e-008 3.4e-008 -2.9e-008
intercept 2.4e-008 1 0.31 -0.59
v 3.4e-008 0.31 1 -0.92
k -2.9e-008 -0.59 -0.92 1
the user,
Parameter Estimates
95.0% Wald Confidence
Interval
Variable Estimate Std. Err. Lower Conf. Limit Upper Conf.
Limit
23
-------
alpha
19.9436
intercept
29.3967
v
2.1387
n
k
81.4567
12 .7371
26.8434
-17.7761
1
25.7117
3.67688
1.30271
7.97842
NA
28.4418
NA - Indicates that this parameter has hit a bound
implied by some inequality constraint and thus
has no standard error.
5 .53053
24.2901
-33.4135
-30.0332
FINAL
9-8-2009
Table of Data and Estimated Values of Interest
Dose N Obs Mean Est Mean Obs Std Dev Est Std Dev Scaled Res.
0
6
27
26.8
4
3.57
0.107
2.5
6
25
25 .3
5
3.57
-0.184
10
6
22
21.9
2
3.57
0.0922
40
6
16
16
4
3.57
-0.0156
Model Descriptions for likelihoods calculated
Model A1: Yij = Mu(i) + e(ij)
Var{e(ij)} = Sigma^2
Model A2: Yij = Mu(i) + e(ij)
Var{e(ij)} = Sigma(i)^2
Model A3: Yij = Mu(i) + e(ij)
Var{e(ij)} = Sigma^2
Model A3 uses any fixed variance parameters that
were specified by the user
Model R: Yi = Mu + e(i)
Var{e(i)} = Sigma^2
Likelihoods of Interest
Model
A1
A2
A3
fitted
R
Log(likelihood)
-42.507095
-40.263184
-42.507095
-42.534207
-52.797690
# Param's
5
8
5
4
2
AIC
95.014191
96.526368
95.014191
93.068414
109.595381
Explanation of Tests
Do responses and/or variances differ among Dose levels?
(A2 vs. R)
Are Variances Homogeneous? (A1 vs A2)
Are variances adequately modeled? (A2 vs. A3)
Does the Model for the Mean Fit? (A3 vs. fitted)
(Note: When rho=0 the results of Test 3 and Test 2 will be the same.)
Test
1:
Test
2 :
Test
3:
Test
4 :
(Not
e:
24
-------
FINAL
9-8-2009
Tests of Interest
Test -2*log(Likelihood Ratio) Test df
p-value
Test 1
Test 2
Test 3
Test 4
0.0542231
25.069
4.48782
4.48782
6
3
3
1
0.0003316
0.2134
0.2134
0.8159
The p-value for Test 1 is less than .05. There appears to be a
difference between response and/or variances among the dose levels
It seems appropriate to model the data
The p-value for Test 2 is greater than .1. A homogeneous variance
model appears to be appropriate here
The p-value for Test 3 is greater than .1. The modeled variance appears
to be appropriate here
The p-value for Test 4 is greater than .1. The model chosen seems
to adeguately describe the data
Benchmark Dose Computation
Specified effect
1
Risk Type
Estimated standard deviations from the control mean
Confidence level
0. 95
BMD
6.45889
BMDL
2.33191
25
------- |