xvEPA

EPA Document# EPA-740-R-25-008
January 2025

United States	Office of Chemical Safety and

Environmental Protection Agency	Pollution Prevention

Cancer Human Health Hazard Assessment for Diisononyl

Phthalate (DINP)

Technical Support Document for the Risk Evaluation

CASRNs: 28553-12-0 and 68515-48-0

(Representative Structure)

January 2025


-------
TABLE OF CONTENTS	

1	INTRODUCTION	5

2	GENOTOXICITY AND MUTAGENICITY	6

3	CANCER HAZARD IDENTIFICATION AND CHARACTERIZATION	9

3.1	Human Evidence	9

3.2	Animal Evidence	9

3.2.1	Liver Tumors	9

3.2.1.1 Conclusions on Liver Tumors	10

3.2.2	Mononuclear Cell Leukemia	13

3.2.2.1 Conclusions on Mononuclear Cell Leukemia	14

3.2.3	Kidney Tumors	17

3.2.3.1 Conclusions on Kidney Tumors	18

3.2.4	Other Tumors	20

4	POSTULATED MODE OF ACTION FOR LIVER TUMORS IN RATS AND MICE	23

4.1	Postulated Mode of Action in Rats and Mice	23

4.1.1	Key Event 1: PPARa Activation	24

4.1.2	Key Event 2: Alterations in Cell Growth Pathways	26

4.1.3	Key Event 3: Perturbation of Cell Growth and Survival	26

4.1.4	Key Event 4: Selective Clonal Expansion of Preneoplastic Foci	27

4.1.5	Modulating Factors	27

4.2	Dose-Response Concordance of Key Events with Tumor Response	28

4.3	Temporal Association of Key Events with Tumor Response	32

4.4	Strength, Consistency, and Specificity of Association of Tumor Response with Key Events.... 32

4.5	Biological Plausibility and Coherence	32

4.6	Other Modes of Carcinogenic Action	33

4.7	Uncertainties and Limitations	35

4.8	Weight of Scientific Evidence: Cancer Classification	36

4.9	Human Relevancy	36

5	CONCLUSIONS	38

REFERENCES	39

APPENDICES	46

Appendix A PATHOLOGY WORKING GROUP REVIEW FOR SPONGIOSIS HEPATIS

AND MNCL (EPL, 1999)	46

LIST OF TABLES	

Table 2-1. Summary of Genotoxicity Studies of DINP	6

Table 3-1. Incidences of Neoplastic Lesions in the Livers of Male and Female F344 Rats Exposed to

DINP for 24 Months (Lington et al., 1997; Bio/dynamics, 1986)	 11

Table 3-2. Incidence of Liver Tumors in Male and Female F344 Rats Exposed to DINP in the Diet for 2

Years (Covance Labs, 1998b)	11

Table 3-3. Incidence of Neoplastic Lesions in the Liver of Male and Female SD Rats Exposed to DINP
in the Diet for 2 Years (Bio/dynamics, 1987)	 12

Page 2 of 49


-------
Table 3-4. Incidence of Liver Tumors in Male and Female B6C3F1 Mice Exposed to DINP in the Diet

for 2 Years (Covance Labs, 1998a)	12

Table 3-5. Incidence of MNCL in F344 Rats Exposed to DINP for 2 Years (Lington et al., 1997;

Bio/dynamics, 1986)	 13

Table 3-6. Incidence of MNCL in F344 Rats Exposed to DINP in the Diet for 2 Years (Covance Labs,

1998b)	14

Table 3-7. MNCL as a Cause of Unscheduled Death in F344 Rats Exposed to DINP in the Diet

(Covance Labs, 1998b)	14

Table 3-8. Incidence of Kidney Tumors in Male F344 Rats Exposed to DINP in the Diet for 2 Years

(Covance Labs, 1998b)	17

Table 3-9. Incidence of Kidney Tumors in F344 Rats Exposed to DINP for 2 Years (Lington et al.,

1997; Bio/dynamics, 1986)	 18

Table 3-10. Incidence of Tumors in Pancreas, Testes, and Uterus in SD Rats Exposed to DINP for 2

Years (Bio/dynamics, 1987)	22

Table 4-1. Dose-Response Concordance for PPARa MOA in Rats	30

Table 4-2. Dose-Response Concordance for PPARa MOA in Mice	31

Table 4-3. Summary of Active ToxCast Assays for DINP	34

LIST OF APPENDIX TABLES	

TableApx A-l. Incidence of MNCL and Selected Hepatic Lesions at Terminal Sacrifice (104 Weeks)
in the Lington et al. (1997) Study in F344 Rats as Determined by the PWG (EPL, 1999)
	47

Table Apx A-2. Incidence of MNCL and Selected Hepatic Lesions at Terminal Sacrifice (104 Weeks)

in the Covance Labs (1998b) Study in F344 Rats as Determined by the PWG (EPL, 1999)
	48

Table Apx A-3. Comparison of Spongiosis Hepatis with MNCL as Determined by the PWG (EPL,

1999)	49

KEY ABBREVIATIONS AND ACRONYMS	

a2u-globulin Alpha 2u-globulin

AhR	Aryl hydrocarbon receptor

ALP	Alkaline phosphatase

ALT	Alanine aminotransferase

AST	Aspartate aminotransferase

BrdU	Bromodeoxyuridine

CAR	Constitutive androstane receptor

CASRN	Chemical abstracts service registry number

CPSC	Consumer Product Safety Commission (U.S.)

DINP	Diisononyl phthalate

DNA	Deoxyribonucleic acid

ECB	European Chemicals Bureau

ECHA	European Chemicals Agency

EFSA	European Food Safety Authority

EPA	Environmental Protection Agency (U.S.) (or the Agency)

F344	Fischer 344 (rats)

GJIC	Gap junctional intercellular communication

GLP	Good Laboratory Practice

GSH	Glutathione

Page 3 of 49


-------
I ARC

International Agency for Research on Cancer

IL

Interleukin

KE

Key event

LOAEL

Lowest-ob served-adverse-effect level

LOEC

Lowest-ob served-effect concentration

MINP

Monoisononyl phthalate

MNCL

Mononuclear cell leukemia

MOA

Mode of action

NF-kB

Nuclear factor kappa B

NICNAS

National Industrial Chemicals Notification and Assessment Scheme

NOAEL

No-observed-adverse-effect level

NTP

National Toxicology Program

OCSPP

Office of Chemical Safety and Pollution Prevention

OEHHA

Office of Environmental Health Hazard Assessment (California)

OPPT

Office of Pollution Prevention and Toxics

PBOX

Peroxisomal beta oxidation

POD

Point of departure

PPARa

Peroxisome proliferator-activated receptor alpha

PWG

Pathology Working Group

PXR

Pregnane X receptor

ROS

Reactive oxygen species

SACC

Science Advisory Committee on Chemicals

SD

Sprague-Dawley (rats)

Srts

Sirtuins

TSCA

Toxic Substances Control Act

U.S.

United States

Page 4 of 49


-------
1 INTRODUCTION

On May 24, 2019, the United States Environmental Protection Agency (EPA or the Agency) received a
request, pursuant to 40 CFR 702.37, from ExxonMobil Chemical Company, through the American
Chemistry Council's High Phthalates Panel (ACC HPP. 2019). to conduct a risk evaluation for
diisononyl phthalate (DINP) (CASRNs 28553-12-0 and 68515-48-0) (Docket ID: EPA-HO-QPPT-2018-
0436). EPA determined that these two CASRNs should be treated as a category of chemical substances
as defined in 15 U.S.C. section 2625(c). On August 19, 2019, EPA opened a 45-day public comment
period to gather information relevant to the requested risk evaluation. The Agency reviewed the request
(along with additional information received during the public comment period) and assessed (1) whether
the circumstances identified in the request constitute "conditions of use" under 40 CFR 702.33, and (2)
whether those conditions of use warrant inclusion within the scope of a risk evaluation for DINP. EPA
determined that the request meets the applicable regulatory criteria and requirements, as prescribed
under 40 CFR 702.37. The Agency granted the request on December 2, 2019, and published the draft
and final scope documents for DINP in August 2020 and 2021, respectively (U.S. EPA 2021. 2020).

Following publication of the final scope document, one of the next steps in the Toxic Substances
Control Act (TSCA) risk evaluation process is to identify and characterize the human health hazards of
DINP and conduct a dose-response assessment to determine the toxicity values to be used to estimate
risks from DINP exposures. This technical support document summarizes the cancer hazards associated
with exposure to DINP. Non-cancer hazards associated with exposure to DINP are summarized in a
separate technical support document, the Non-cancer Raman Health Hazard Assessment for Diisononyl
Phthalate (DINP) (U.S. EPA. 2025).

The carcinogenicity of DINP has been evaluated in existing assessments by Health Canada, U.S.
Consumer Product Safety Commission (U.S. CPSC), European Chemicals Agency (ECHA), Australia
National Industrial Chemicals Notification and Assessment Scheme (NICNAS), and California's Office
of Environmental Health Hazard Assessment (OEHHA) (ECCC/HC. 2020; EC/HC. 2015; ECHA. 2013;
Tomar et al.. 2013; NICNAS. 2012; U.S. CPSC. 2010; ECB. 2003; U.S. CPSC. 2001). To date, DINP
has been classified as a carcinogen by California OEHHA and is listed under California's Proposition 65
as a carcinogen (OEHHA. 2013; Tomar et al.. 2013). Other authoritative agencies have not classified
DINP as a carcinogen or evaluated DINP quantitatively for carcinogenic risk to human health.

This technical support document summarizes the available evidence for the carcinogenicity of DINP, the
majority of which comes from experimental animal models. The remainder of this document is
organized as follows:

•	Section 2 summarizes available genotoxicity data for DINP.

•	Section 3 summarizes available human and animal evidence for the carcinogenicity of DINP.

•	Section 4 summarizes available liver tumor data and postulated mode of action (MOA) for liver
tumors in rodents.

•	Section 5 summarizes EPA's conclusions and next steps.

•	Appendix A summarizes the results of a Pathology Working Group's review for spongiosis
hepatis and mononuclear cell leukemia.

Page 5 of 49


-------
2 GENOTOXICITY AND MUTAGENICITY

The genotoxicity of DINP has been evaluated in several existing assessments, which have consistently
concluded that DINP is not genotoxic nor is it likely to be genotoxic (ECCC/HC. 2020; EC/HC. 2015;
ECHA. 2013; NICNAS. 2012; U.S. CPSC. 2010; EFSA. 2005; ECB. 2003; U.S. CPSC. 2001). EPA
reviewed available genotoxicity studies of DINP that were cited in existing assessments (Table 2-1) and
considered newer studies published between 2014 and 2024. No new genotoxicity studies of DINP were
identified.

The mutagenic and genotoxic potential of DINP has been evaluated in 20 studies (Table 2-1). Available
studies include two in vivo micronucleus tests in rodents, one in vitro chromosomal aberration assay,
two in vitro mouse lymphoma assays, five bacterial reverse mutation assays, one in vitro unscheduled
DNA synthesis assay, and nine in vitro cell transformation assays. No evidence of mutagenic activity
was observed in five bacterial reverse mutation assays or two in vitro mouse lymphoma assays (with or
without metabolic activation). DINP did not induce chromosomal aberrations in Chinese hamster ovary
cells in vitro, cause unscheduled DNA synthesis in primary rat hepatocytes, or induce clastogenic effects
or micronuclei formation in vivo in studies of mice or rats. Of the nine available in vitro transformation
assays, only one study reported a positive result for transformation in Balb/c-3T3 A31 mouse cells in the
absence of metabolic activation (Microbiological Associates. 1982c).

Consistent with the conclusions of existing assessments of DINP, available studies that evaluated the
mutagenic and genotoxic potential of DINP are consistently negative. Therefore, EPA considers the
weight of scientific evidence to indicate that DINP is not likely to be genotoxic or mutagenic.

Table 2-1. Summary of Genotoxicity Studies of DINP

Test
Type

Test System
(Species/Strain/Sex)

Dose/Duration

Metabolic
Activation

Result

Reference(s)

Chromosomal aberrations - in vivo

Micronucleus
(bone marrow)
(Adhered to
OECD 474)

Male CD-I mice

Oral (gavage) doses of
0,500, 1,000, or 2,000
mg/kg-day for 2 days;
sacrificed on day 3

Not applicable

Negative for
micronuclei

(McKee et al..
2000)

Chromosomal
aberrations in
femoral bone
marrow cells

Male F344 rats

Oral (gavage) doses of
0, 0.5, 1.7, or 5.0
mL/kg-day for 5 days

Not applicable

Negative for
micronuclei

(Microbiological
Associates. 1982b)

Chromosomal aberrations - in vitro

Chromosomal
aberrations

Chinese hamster
ovary cells

0, 40, 80, or 160 ng/mL
for 3 hours (with
activation) or 20 hours
(without activation)

± Aroclor-
induced rat
liver S9

Negative for

chromosomal

aberrations

(McKee et al..
2000)

Gene mutations - in vitro

Mouse
lymphoma
mutation assay

L5178Y+/— mouse
lymphoma cells

0, 0.001,0.01,0.1, 1.0,
10, 100 nL/mL (±S9)

± Aroclor-
induced rat
liver S9

Negative for
mutagenicity

(EG&G Mason
Research Institute.
1982a)

Mouse
lymphoma
mutation assay

L5178Y+/— mouse
lymphoma cells

1.5-8 jil/ml (-S9);
0.05-0.6 nL/mL (+S9)

± Aroclor-
induced rat
liver S9

Negative for
mutagenicity

(Barber et al.. 2000)

Bacterial
reverse

mutation assay

S. typhimurium strains
TA 98, TA 100,

0.1,0.5,2.5,5, 10
|u L/plate

± Aroclor-
induced rat
liver S9

Negative for
mutagenicity

(EG&G Mason
Research Institute.
1982b)

Page 6 of 49


-------
Test
Type

Test System
(Species/Strain/Sex)

Dose/Duration

Metabolic
Activation

Result

Reference(s)



TA 1535, TA 1537,











and TA 1538









Bacterial
reverse

S. typhimurium strains
TA 98, TA 100,

0, 100,333, 1,000,
3,333, 10,000 (ig/plate

± Aroclor
1254-induced

Negative for
mutagenicity

(Zeiaer et al.. 1985)

mutation assay

TA 1535, and TA
1537



rat or hamster
liver S9





Bacterial
reverse

mutation assay

S. typhimurium strains
TA 98, TA 100,
TA 1535, and TA
1537

20-5,000 (ig/plate

± Aroclor-
induced rat
liver S9

Negative for
mutagenicity

IYBASF. 1995.
1986) as reported
by ECB (2003)]17

Bacterial

S. typhimurium strains

0.5-5,000 (ig/plate

± Aroclor-

Negative for

(McKee et al..

reverse

mutation assay

TA 98, TA 100,
TA 1535, TA 1537,



induced rat
liver S9

mutagenicity

2000)

(plate

and TA 1538









incorporation

assay)











Bacterial
reverse

S. typhimurium strains
TA 98, TA 100,

20-5,000 (ig/plate

± Aroclor-
induced rat

Negative for
mutagenicity

(McKee et al..
2000)

mutation assay

TA 1535, and TA



liver S9





(pre-incubation

1537









assay)











Other genotoxicity assays

Unscheduled

Rat hepatocyte

0, 0.625, 1.25,2.5,5.0,

No

No increase in

(Litton Bionetics.

DNA synthesis

primary culture

10.0 |iL/mL



unscheduled

DNA
synthesis

1982b)

In vitro cell

Balb/c-3T3 A31

125-3,750 nL/niL

No

No significant

(Litton Bionetics.

transformation

mouse cells





increase in

transformed

foci

1985)

In vitro cell

Balb/c-3T3 A31

2.5-254.5 |ig/mL

No

No significant

(Litton Bionetics.

transformation

mouse cells





increase in

transformed

foci

1981)

In vitro cell

Balb/c-3T3 A31

0.0326-3,260 yg/mL

No

No significant

(Litton Bionetics.

transformation

mouse cells





increase in

transformed

foci

1982a)

In vitro cell

Balb/c-3T3 A31

0.125-3.750 nL/mL

No

No significant

(Barber et al.. 2000)

transformation

mouse cells





increase in

transformed

foci



In vitro cell

Balb/c-3T3 A31

0.1-1 nL/mL

± rat liver S9

No significant

(Microbiological

transformation

mouse cells





increase in

transformed

foci

Associates. 1982a)

In vitro cell
transformation

Balb/c-3T3 A31
mouse cells

0.03-1 |iL/mL

No

No significant
increase in
transformed
foci

(Microbiological
Associates. 1982c)

In vitro cell
transformation

Balb/c-3T3 A31
mouse cells

0.01-1.0 jiL/mL

No

No significant
increase in

(Microbiological
Associates. 1981)

Page 7 of 49


-------
Test
Type

Test System
(Species/Strain/Sex)

Dose/Duration

Metabolic
Activation

Result

Reference(s)









transformed
foci



In vitro cell
transformation

Balb/c-3T3 A31
mouse cells

0.03-1 |iL/mL

No

Positive
(significant
increase in
transformed
foci)

(Microbiological
Associates. 1982d)

In vitro cell
transformation

Balb/c- 3T3 mouse
cells co-cultured with
transformed cloned
cells (strain 4-1-1)

5-5,000 ng/mL

No

No increase in
proliferation
rate of Balb/c
3T3 cells

(Fushiwaki et al..
2003)

" Studv reports were not reasonably available to EPA. Information is as reported by ECB (2003).

Page 8 of 49


-------
3 CANCER HAZARD IDENTIFICATION AND
CHARACTERIZATION

This section summarizes available human (Section 3.1) and animal evidence (Section 3.2) for the
carcinogenicity of DINP. Section 3.2 discusses evidence for mononuclear cell leukemia (MNCL),
kidney tumors, and other tumors observed in experimental animal models. Evidence for liver tumors in
rodents and EPA's MO A analysis for liver tumors is provided in Section 4.

3.1	Human Evidence

EPA reviewed conclusions from previous assessments conducted by Health Canada (2018) that
investigated the association between exposure to DINP and cancer. Additionally, EPA also evaluated
new epidemiologic studies published after the Health Canada (2018) assessment (i.e., published 2018-
2019) to determine if newer epidemiologic studies would provide useful information for evaluating
exposure-response relationship. The Agency identified two new medium quality studies that evaluated
exposure to DINP and cancer. The first medium quality study, a case-control analysis by Parada et al.
(2018) with a mortality follow-up component among women in the Long Island Breast Cancer Study
Project, evaluated breast cancer mortality among cases with spot urine sample collected 3 months after
breast cancer diagnosis. Inverse associations were observed between urine levels of two DINP
metabolites (i.e., mono-(carboxynonyl) phthalate [MCNP] and mono-(carboxyoctyl) phthalate [MCOP])
and breast cancer for single quintiles, but the associations were not statistically significant.

The second medium quality study, a nested case-control study by Reeves et al. (2019) of the Women's
Health Initiative prospective cohort, investigated the association between incident breast cancer cases in
postmenopausal women and DINP. The authors found no significant association with one urinary DINP
metabolite (i.e., MCOP) and breast cancer in analysis using either ln-transformed or quartile exposure
variables (adjusted odds ratio in models using ln-MCOP = 1.02; 95% confidence interval: 0.90-1.16).
Findings were similar in models stratified by estrogen/progesterone receptor status and body mass index.

Health Canada (2018) also evaluated the relationship between phthalates and breast cancer. However, no
epidemiologic studies were identified by Health Canada that examined the association between DINP
and its metabolites and biomarkers of breast cancer.

3.2	Animal Evidence

Four 2-year dietary studies evaluating the carcinogenicity of DINP in rodent models are available,
including three studies of male and female Fischer 344 (F344) and Sprague-Dawley (SD) rats (Covance
Labs. 1998b; Lington et al.. 1997; Bio/dynamics. 1987) and one study of male and female B6C3F1 mice
(Covance Labs. 1998a). Available studies have been discussed extensively in existing assessments of
DINP. No new carcinogenicity studies of DINP with experimental laboratory animals were identified by
EPA.

Across available studies, statistically significant increases in liver tumors, MNCL, and kidney tumors
have been reported. Statistically non-significant increases in tumors in the testes, uterus, and pancreas
have also been reported. Evidence for liver tumors, MNCL, kidney tumors, and other tumors is
discussed in Section 3.2.1 through Section 3.2.4.

3.2.1 Liver Tumors

The Non-cancer Raman Health Hazard Assessment for Diisononyl Phthalate (DINP) (U.S. EPA. 2025)
describes the non-cancer liver effects observed following exposure to DINP in experimental animal
models. Notably, many of the non-cancer liver effects observed in rodents following oral exposure to

Page 9 of 49


-------
DINP comprise a suite of effects that may represent a progression from non-cancer to cancer (e.g.,
increased liver weight, increased serum levels of alanine aminotransferase [ALT], AST, and ALP,
histopathologic lesions such as hepatocellular hypertrophy and focal necrosis).

DINP has been evaluated for carcinogenicity in two 2-year dietary studies of F344 rats (Covance Labs.
1998b; Lington et al.. 19971 one 2-year dietary study of SD rats (Bio/dynamics. 1987). and one 2-year
dietary study of B6C3F1 mice (Covance Labs. 1998a). Statistically significant increased incidences of
tumors in the liver were reported in three out of four of the chronic 2-year studies (see Table 3-1 through
Table 3-4). In one study, no statistically significant increases in neoplastic nodules and/or hepatocellular
carcinomas were observed in male or female F344 rats treated with up to 307 to 375 mg/kg-day DINP
for 2 years (Table 3-1)—although hepatocellular cancer was observed in 3 out of 80 males from the
high-dose groups compared to 0 out of 80 in controls (Lington et al.. 1997; Bio/dynamics. 1986).

Two other studies of F344 and SD rats by Covance Labs (1998b) and Bio/dynamics (1987).
respectively, included higher doses than Lington et al. (1997). and reported significant increases in
hepatocellular adenoma and/or carcinoma (Table 3-2 and Table 3-3). Increased incidence of
hepatocellular carcinoma (males only), and adenomas or carcinomas combined (both sexes) were
observed in male and female F344 rats treated with up to 733 to 885 mg/kg-day DINP for 2 years
(Covance Labs. 1998b) (Table 3-2). In the second study, hepatocellular carcinomas were significantly
increased in high-dose female SD rats treated with 672 mg/kg-day DINP for 2 years, while no
significant increase in neoplastic nodules or hepatocellular carcinomas were observed in male SD rats
treated with up to 553 mg/kg-day DINP for 2 years (Table 3-3) (Bio/dynamics. 1987).

Finally, in a 2-year chronic study of DINP with B6C3F1 mice, the incidence of carcinomas was
significantly increased in males at 1,560 mg/kg-day and females at 910 mg/kg-day and above, while the
combined incidence of hepatocellular adenomas and carcinomas were significantly increased in both
males (>742 mg/kg-day) and females (>336 mg/kg-day) (Table 3-4) (Covance Labs. 1998a).

3.2.1.1 Conclusions on Liver Tumors

Collectively, available studies provide consistent evidence that chronic oral exposure to DINP can cause
treatment-related liver tumors in both sexes of several strains of rats (i.e., F344 and SD) and mice
(B6C3F1). EPA further considers the weight of evidence for liver carcinogenesis and its underlying
MOA in Section 4.

Page 10 of 49


-------
Table 3-1. Incidences of Neoplastic Lesions in the Livers of Male and Female F344 Rats Exposed

to DINP for 24 Months (Lington et a

., 1997; Bio/dvnamics, 1986)

Lesion

Dose Group
mg/kg-day (ppm)

Control

15M/18F
(300)

152 M /184 F
(3,000)

307 M / 375 F
(6,000)

Males17

Neoplastic nodules

3/81 (3.7%)

1/80(1.3%)

1/80(1.3%)

1/80(1.3%)

Hepatocellular cancer

0/81 (0%)

0/80 (0%)

0/80 (0%)

3/80 (3.8%)

Neoplastic nodules or cancer (combined)

3/81(3.7%)

1/80(1.3%)

1/80(1.3%)

4/80 (5.0%)

Females'1

Neoplastic nodules

0/81 (0%)

2/81 (2.5%)

0/80 (0%)

1/80(1.3%)

Hepatocellular cancer

1/81 (1.2%)

0/81 (0%)

0/80 (0%)

1/80(1.3%)

Neoplastic nodules or cancer (combined)

1/81 (1.2%)

2/81 (2.5%)

0/80 (0%)

2/80 (2.5%)

Source: Table 8 in Lington et al. (1997)

M = male; F = female

" Number of animals with lesion/ total number of animals examined. Percent lesion incidence in parentheses. No statistically
significant increases in hepatocellular nodules and/or cancer were observed in either sex.

Table 3-2. Incidence of Liver Tumors in Male and Female F344 Rats Exposed to DINP in the Diet
for 2 Years (Covance Labs, 1998b)a&	

Lesion

Dose Group
mg/kg-day (ppm)

Control

29 M / 36 F
(500)

88 M /109 F
(1,500)

359 M / 442 F
(6,000)

733 M / 885 F
(12,000)

Males

Hepatocellular adenoma

4/65& (6%)

3/50 (6%)

2/50 (4%)

6/65 (9%)

10/65 (15%)

Hepatocellular carcinoma

1/65 (2%)

0/50 (0%)

0/50 (0%)

1/65 (2%)

12/65* (18%)

Adenoma or carcinoma (combined)

5/65 (8%)

3/50 (6%)

2/50 (4%)

7/65 (11%)

18/65* (28%)

Females

Hepatocellular adenoma

0/65 (0%)

1/49 (2%)

0/50 (0%)

1/65 (2%)

3/65 (5%)

Hepatocellular carcinoma

1/65 (2%)

0/49 (0%)

0/50 (0%)

1/65 (2%)

5/65 (8%)

Adenoma or carcinoma (combined)

1/65 (2%)

1/49 (2%)

0/50 (0%)

2/65 (3%)

8/65* (12%)

Source: U.S. CPSC (2001); Table IX-1 (t>g. 68); oo. 68-71 and AoDcndix B.

M = male; F = female

* = statistically significant at p < 0.05 by one or more of the following: Fisher's Exact test. Poly-3, Logistic Regression, or
Life Table analysis.

" Where results are of borderline significance or greater, level of statistical significance computed by logistic regression is
given. Significance value for trend is given in the column for the control group. Significance values for these findings
calculated using different statistical tests are given in Appendix B, section A. Analysis of individual animal data as
Dcrformcd b\ the National Toxicology Program (NTP) (U.S. CPSC. 2001).

h Number of animals with neoplasm/ total number of animals examined. Percent tumor incidence in parentheses. Based on
extraction and analysis of individual animal data as reported in U.S. CPSC (2001). Overall incidence for control. 6.000 DDin
and 12,000 ppm groups (n = 65) includes incidence data for unscheduled deaths, interim sacrifice at week 78, and terminal
sacrifice. Overall incidence for the remaining groups includes incidence data for unscheduled deaths and terminal sacrifice.

Page 11 of 49


-------
Table 3-3. Incidence of Neoplastic Lesions in the Liver of Male and Female SD Rats Exposed to
DINP in the Diet for 2 Years (Bio/dynamics, 1987)"	

Lesion

Dose Group
mg/kg-day (ppm)

Control

27 M / 33 F
(500)

271 M/331 F
(5,000)

553 M / 672 F
(10,000)

Males

Hepatocellular carcinoma

2/70 (2.9%)

2/69 (2.9%)

6/69 (8.7%)

4/70 (5.7%)

Neoplastic nodule(s) h

2/70 (2.9%)

5/69 (7.2%)

6/69 (8.7%)

5/70 (7.1%)

Females

Hepatocellular carcinoma

0/70 (0%)t

0/70 (0%)

5/70 (7.1%)

7/70 (10%)*

Neoplastic nodule(s)

1/70 (1.4%)

1/70 (1.4%)

5/70 (7.1%)

2/70 (2.9%)

Source: Appendix K, Figure 1, pp. 11 (pp. 426 of the study report PDF) (Bio/dynamics, 1987).

* Statistically significant (p < 0.05) from the control group by a two-tailed Fisher's exact test
f Statistically significant trend (p < 0.05) based on a Chi-square contingency trend test calculated for this review.

"Data in this table indicate all animals assessed for histopathology throughout the study; i.e., including the interim
sacrifice, the terminal sacrifice, and unscheduled deaths. For late-developing tumors (hepatocellular carcinoma, pancreatic
islet cell tumors, testicular interstitial cell tumors), statistical analysis was performed excluding animals that died or were
sacrificed up to 12 months, leaving n = 57, 57, 59, 59 in males and n = 59, 56, 60, 59 in females in the control, low-, mid-
and high-dose groups, respectively.

b Pathology report does not define this lesion further, which is a reporting deficiency that reduces the ability to compare
results of Bio/dynamics (1987) to those of other studies which report incidences of hepatocellular adenomas, carcinomas,
and adenomas or carcinomas, combined.

Table 3-4. Incidence of Liver Tumors in Male and Female B6C3F1 Mice Exposed to DINP in the
Diet for 2 Years (Covance Labs, 1998a)	

Lesion

Dose Group
mg/kg-day (ppm)

Control

(%)

90M/112F
(500)

276 M / 336 F
(1,500)

742 M / 910 F
(6,000)

1,560 M /1,888 F
(12,000)

Males

Hepatocellular adenoma

10/706 (14%)

7/67 (10%)

8/66 (12%)

15/65 (23%)

13/70 (19%)

Hepatocellular carcinoma

10/70 (14%)

8/67 (12%)

10/66 (15%)

17/65 (26%)

20/70* (29%)

Adenoma or carcinoma (combined)

16/70 (23%)

13/67 (19%)

18/66 (27%)

28/65* (43%)

31/70* (44%)

Females

Hepatocellular adenoma

2/70 (3%)

4/68 (6%)

5/68 (7%)

4/67 (6%)

18/70* (26%)

Hepatocellular carcinoma

1/70 (1%)

2/68 (3%)

5/68 (7%)

7/67* (10%)

19/70* (27%)

Adenoma or carcinoma (combined)

3/70 (4%)

5/68 (7%)

10/68* (15%)

11/67* (16%)

33/70* (47%)

Source: U.S. CPSC (2001) Table IX-6 (d. 73) and AoDcndix B.

M = male; F = female

* = significant from the control at p < 0.05 by logistic regression analysis

" Where results are of borderline significance or greater, level of statistical significance computed by logistic regression is
given. Significance value for trend is given in the column for the control group. Significance values for these findings
calculated usins different statistical tests are siven in Aroendix B. section B (U.S. CPSC. 2001).
b Number of animals with tumor/total number of animals examined. Percent tumor incidence in parentheses.

Page 12 of 49


-------
3.2.2 Mononuclear Cell Leukemia

MNCL has been observed in F344 rats in two 2-year dietary studies (Covance Labs. 1998b; Lington et
al.. 1997; Bio/dynamics. 1986). In contrast, MNCL has not been observed in SD rats in a 104-week
study (Bio/dynamics. 1987) nor in B6C3F1 mice exposed to DINP for at least 104 weeks (Covance
Labs. 1998a).

Lington et al. (1997) reported the incidence data for MNCL. The incidence of MNCL was statistically
significantly increased in the mid- and high-dose groups for both sexes when compared with the
concurrent control groups (Table 3-5). MNCL was detected in 41, 35, 60, and 64 percent of males and
27, 25, 38, and 54 percent of females in the control, low-, mid-, and high-dose groups, respectively. As
reported by the study authors, MNCL has a significant increasing trend over time and was the most
common cause of unscheduled deaths and/or morbidity. In many of the treated rats, MNCL was detected
at a very early stage and was limited to an increase in the mononuclear cells in the hepatic sinusoids.

Table 3-5. Incidence of MNCL in F344 Rats Exposed to DINP for 2 Years (Lington et al., 1997;
Bio/dynamics, 1986)	

Lesion

Dose Group
(mg/kg-day) (ppm)

Control

15M/18F
(300)

152 M /184
(3,000)

307 M / 375 F
(6,000)

Males17

33/81 (41%)

28/80 (35%)

48/80* (60%)

51/80* (64%)

Females'1

22/81 (27%)

20/81 (25%)

30/80* (38%)

43/80* (54%)

Source: Table 8 in Lington et al. (1997)

M = male; F = female

" Number of animals with lesion/ total number of animals examined. Percent lesion incidence in parentheses.
* Statistically significant at p < 0.05 when compared to the control incidence using Fisher's Exact test; statistical analysis
performed by Lington et al. (1997).

In a study by Covance Labs (1998b). the incidences of MNCL in male and female rats receiving the
6,000 and 12,000 ppm concentrations of DINP in the diet were significantly increased with statistically
significant dose-related trends (Table 3-6). The incidences of MNCL in the recovery groups were also
significantly greater than in the controls. There is some evidence that the onset of MNCL was earlier in
treated males. MNCL was first detected in the 6,000 ppm group via an unscheduled death at study day
352. In comparison, MNCL was first detected in the control group at an interim sacrifice at day 549.
Decreases in hemoglobin concentration and red blood cell numbers and a statistically significant
increase in mean spleen weight in both male and female rats were correlated with the incidence of
MNCL. Between 31 and 60 percent of unscheduled deaths in the study were attributable to MNCL
(Table 3-7), demonstrating that this lesion is life-threatening in rats treated with DINP.

A Histopathology Peer Review and a Pathology Working Group (PWG) review (EPL. 1999) was
conducted on selected lesions of the liver and spleen observed in F344 rats in the 2-year bioassays
reported by Lington et al. (1997) and Covance Labs (1998b). The PWG review evaluated the
significance of spongiosis hepatis, foci of cellular alteration, primary hepatocellular neoplasms in the
liver, and the significance of MNCL. Notably, the results of the PWG (EPL. 1999) generally confirmed
the original findings of the study pathologist(s), including incidence of MNCL in F344 rats in both
studies. PWG findings are further discussed in Appendix A.

Page 13 of 49


-------
Table 3-6. Incidence of MNCL in F344 Rats Exposed to DINP in the Diet for 2 Years (Covance
Labs. 1998b) "bc	

Sex

Dose Group
mg/kg-day (ppm)

Control

29 M / 36 F
(500)

88 M /109 F
(1,500)

359 M / 442 F
(6,000)

733 M / 885 F
(12,000)

High-Dose /
Recovery6
637 M/ 774 F (12,000)

Males

22/65 (34%)

23/50 (46%)

21/50 (42%)

32/65* (49%)

30/65* (46%)

31/50*'# (62%)

Females

17/65 (26%)

16/49 (33%)

9/50(18%)

30/65* (46%)

29/65* (45%)

24/50*'# (48%)

Source: U.S. CPSC (2001) text vv. 68-71 and Aoocndix B.

M = male; F = female

* = statistically significant at p < 0.05 by one or more of the following: Fisher's Exact test, Poly-3, Logistic
Regression, or Life Table analysis.

11 Analysis of individual animal data as performed by NTP and reported in the text and Appendix B of U.S CPSC
(2001).

h The high-dose/recovery group received 12,000 ppm for 78 weeks, followed by a 26-week recovery period during
which the animals received basal diet alone.

c Number of animals with neoplasm/ total number of animals examined. Percent tumor incidence in parentheses. Based
on extraction and analysis of individual animal data as reported in U.S. CPSC (2001). Overall incidence for control
(6,000 ppm and 12,000 ppm) groups (n = 65) includes incidence data for unscheduled deaths, interim sacrifice at week
78, and terminal sacrifice. Overall incidence for the remaining groups includes incidence data for unscheduled deaths
and terminal sacrifice.

d Statistically significant at p < 0.05 by Fisher's Exact test conducted by Syracuse Research Corporation.

Table 3-7. MNCL as a Cause of Unscheduled Death in F344 Rats Exposed to DINP in the Diet
Covance Labs, 1998b)	

Sex

Dose Group
mg/kg-day (ppm)

Control

29 M / 36 F
(500)

88 M /109 F
(1,500)

359 M / 442 F
(6,000)

733 M / 885 F
(12,000)

Recovery "
637 M / 774 F
(12,000)

Males

H22b (32%)

8/23 (35%)

7/21 (33%)

16/32 (50%)

18/30 (60%)

14/31 (45%)

Females

7/17(41%)

5/16(31%)

3/9 (33%)

12/29 (41%)

13/30 (43%)

12/24 (50%)

Source: Compiled from incidence data and death comments in Table 10E (pp. 365 and 381) in Covance Labs (1998b).
M = male; F = female

11 The high-dose/recovery group received 12,000 ppm for 78 weeks, followed by a 26-week recovery period during
which test animals received basal diet alone.

h Number of deaths attributed to MNCL/total number of deaths; percentage of deaths attributable to MNCL in
parentheses.

3.2.2.1 Conclusions on Mononuclear Cell Leukemia

The incidence of MNCL was significantly elevated in male and female F344 rats exposed to DINP in
the diet when compared to study control animals in two independent carcinogenicity studies (Covance
Labs. 1998b; Lington et al.. 1997). In Lington et al. (1997). incidences of MNCL were statistically
significantly increased at 152 and 307 mg/kg-day in the males (60-64% in treated rats vs. 41% in
concurrent controls) as well as in the females at 184 and 375 mg/kg-day (38-54% in treated rats vs. 27%
in concurrent controls). In the 2-year study in F344 rats conducted by Covance Labs (1998b). incidences
of MNCL were significantly increased at 359 and 733 mg/kg-day in the treated males (46-62%

Page 14 of 49


-------
incidence) compared to concurrent controls (34% incidence) as well as in the treated females at 442 and
885 mg/kg-day (45-48%) compared to concurrent controls (26%). Inconsistent with findings from the
two chronic studies of F344 rats, MNCL was not observed in male or female SD rats treated with up to
553 to 672 mg/kg-day DINP for 2 years (Bio/dynamics. 1987) or male and female B6C3F1 mice treated
with up to 1,560 to 1,888 mg/kg-day DINP for 2 years (Covance Labs. 1998a).

MNCL is a spontaneously occurring neoplasm of the hematopoietic system that reduces lifespan and is
one of the most common tumor types occurring at a high background rate in the F344 strain of rat (also
referred to as Fisher rat leukemia because it is so common) (Thomas et al.. 2007). Historical control data
from NTP have demonstrated an increase in the spontaneous background incidence of MNCL in
untreated male and female F344 rats from 7.9 and 2.1 percent in males and females, respectively, in
1971 to 52.5 and 24.2 percent in males and females, respectively, from 1995 through 1998 (Thomas et
al.. 2007). Spontaneous incidence of MNCL in other strains of rat appear to be rare. Brix et al. (2005)
report the incidence of MNCL in female Harlan SD rats to be 0.5 percent in NTP 2-year studies. Further,
MNCL does not appear to occur naturally in mice (Thomas et al.. 2007). The F344/N strain of rat was
used in NTP 2-year chronic and carcinogenicity bioassays for nearly 30 years (King-Herbert et al.. 2010;
King-Herbert and Thayer. 2006). However, in the early 2000s NTP stopped using the F344/N strain of
rat in large part because of high background incidence of MNCL and testicular Ley dig cell tumors that
confounded bioassay interpretation. NTP subsequently replaced the F344 strain of rats with the Harlan
SD strain (King-Herbert et al.. 2010; King-Herbert and Thayer. 2006).

Given the high and variable background rate of MNCL in F344 rats, it is important to consider
concurrent control data, historical control data, and time to onset of MNCL to assist in determining
whether observed increases in MNCL are treatment-related.

EPA acknowledges that MNCL has a high background incidence in F344 rats as is noted by concurrent
control incidence of 26 to 41 percent in the two studies described above (Covance Labs. 1998b; Lington
et al.. 1997). The incidence of MNCL was significantly elevated in male and female rats exposed to
DINP in the diet when compared to concurrent controls in these studies; however, no historical control
data from the performing laboratories were provided. EPA's Guidelines for Carcinogen Risk Assessment
(2005) state that the most relevant historical control data comes from the same laboratory and supplier
and are within two to three years of the study under review, and that other historical control data should
be used with extreme caution. Lack of relevant laboratory historical control data for incidence and time
to onset of MNCL make it challenging to determine if the increase in MNCL observed in high-dose
F344 rats treated with DINP, which was statistically significant compared to concurrent controls, is
treatment-related and is a source of uncertainty.

The limited information available indicates that time to onset of MNCL was shorter in DINP-treated
animals compared to concurrent controls. In Lington et al. (1997). the study authors reported that MNCL
has a significant increasing trend over time and was the most common cause of unscheduled deaths
and/or morbidity. In many of the treated rats, MNCL was detected at a very early stage but was limited
to an increase in the mononuclear cells in the hepatic sinusoids. Similar to the Lington study, in the 2-
year study in rats conducted by Covance Labs (1998b). there is some evidence that the onset of MNCL
was earlier in treated males, with the first detected in the 359 mg/kg-day group via an unscheduled death
at study day 352 compared to the first detected in the control group at an interim sacrifice at day 549.

Another source of uncertainty is lack of MO A information for induction of MNCL in F344 rats. The
MO A for induction of MNCL in F344 rats is unknown. Lack of MO A information makes it difficult to
determine human relevancy. There is additional uncertainty related to the human correlate to MNCL in

Page 15 of 49


-------
F344 rats. Some researchers have suggested that based on the biological and functional features in the
F344 rat, MNCL is analogous to large granular lymphocyte (LGL) in humans (Caldwell et al.. 1999;
Caldwell 1999; Reynolds and Foon. 1984). There are two major human LGL leukemias, including
CD3+ LGL leukemia and CD3- LGL leukemia with natural killer cell activity (reviewed in (Maronpot
et al.. 2016; Thomas et al.. 2007)). Thomas et al. (2007) contend that MNCL in F344 rats shares some
characteristics in common with aggressive natural killer cell leukemia (ANKCL) in humans, and that
ANKCL may be a human correlate. However, Maronpot et al. (2016) point out that ANKCL is
extremely rare with less than 98 cases reported worldwide, and its etiology is related to infection with
Epstein-Barr virus, not chemical exposure. This is in contrast to MNCL in F344 rats, which is a more
common form of leukemia and is not associated with a viral etiology. However, under EPA's Guidelines
for Carcinogen Risk Assessment (U.S. EPA. 2005). site concordance is not always assumed between
animals and humans.

EPA considers the available data inadequate for delineation of a plausible sequence of events leading to
development of MNCL in rats exposed to DINP. Therefore, the significance of MNCL and its biological
relevance for human cancer risk remains uncertain. Other regulatory agencies have also considered the
human relevance of MNCL. Generally, other agencies such as Australia NICNAS (2012)1 Health
Canada (EC/HC. 2015).2 U.S. CPSC (2010),3 and ECHA (2013)4 have concluded that MNCL observed
in F344 rats is not human relevant or has unclear human relevance and refrained from using MNCL to
predict cancer risk in humans. In contrast, California OEHHA (Tomar et al.. 2013) lists MNCL in F344
rats as one of the tumor types to support the Proposition 65 listing of DINP; however, OEHHA does not
appear to draw any specific conclusions related to the MOA underlying MNCL or its human relevance.

Overall, the Science Advisory Committee on Chemicals (SACC) recommended that "the observation of
an increased incidence of MNCL in a chronic bioassay employing the Fisher 344 rat should not be
considered a factor in the determination of the cancer classification..." and "Most Committee members
agreed that given the material presented in a retrospective review, MNCL and Ley dig Cell Tumors,
among other tumor responses in F344 rat carcinogenicity studies lack relevance in predicting human
carcinogenicity (Maronpot et al., 2016)." (U.S. EPA. 2024). Consistent with the recommendations of the

1	Australia NICNAS concluded, "In rat carcinogenicity studies, increased incidences of MCL, kidney and liver neoplasia
were observed. MCL was observed in DINP toxicological studies with Fischer 344 rats but not with Sprague Dawley rats.
MCL is a common neoplasm in Fischer 344 rats with no comparable tumour type in humans and its increased incidence after
chronic exposure to some substances is a strain-specific effect (Caldwell. 1999). Therefore, MCL observed in Fischer 344
rats is not regarded as relevant to humans" (p. 49 of (NICNAS. 2012)).

2	Health Canada concluded, "Mononuclear cell leukemia of the spleen was also reported in Fischer rats. However, this type
of lesion is likely specific to aging rats of this strain and is unlikely to be relevant to humans (Health Canada 2015d)." (p. 95
of (Health Canada. 2015)).

3	U.S. CPSC concluded, "Elevated incidence of MNCL is a common finding in chronic studies in Fischer rats. Due to its high
background rate, MNCL is often considered to be of uncertain relevance in the evaluation of the cancer hazard in humans.
Furthermore, no hematopoietic neoplasms were found in Sprague-Dawley CD rats treated with DINP-A (Bio/dynamics.
1986) or in mice treated with DINP-1 (Caldwell. 1999). Therefore, MNCL will not be used to predict cancer risk in humans"
(p. 82 of (U.S. CPSC. 2010)).

4	ECHA concluded, "With regard to MNCL, the review by (Thomas et al„ 2007) suggests that unlike previously thought
there might be a human counterpart to MNCL in rats. The probability that the MNCL seen in the Exxon and Aristech studies
would be a result of chance findings seems low. Nevertheless, the increased incidences of MNCL remain difficult to interpret
in the light of the high and variable background incidences and the unclear relevance to humans. DINP is not genotoxic, and
it is argued (Caldwell. 1999) that MNCL follows a threshold mode of action. The available information does not allow to
draw definite conclusions on the matter. However, as a reasonable approach it would be possible to conclude that the MNCL
findings further strengthen the selected NOAELs for repeated dose toxicity (15 and 88 mg/kg bw/day). Since such conclusion
would not influence the outcome of the current risk assessment, the endpoint is not taken further to the risk characterization
step" (p. 98 of (ECHA 2013)).

Page 16 of 49


-------
SACC, EPA is not further considering MNCL as a factor in the determination of the cancer
classification for DINP.

3.2.3 Kidney Tumors	

Statistically significant increased incidence of kidney tumors have been observed in one 2-year dietary
study of F344 rats (Covance Labs. 1998b). Malignant renal tubule cell carcinomas were detected in two
high-dose (733 mg/kg-day) male rats and four males treated with 637 mg/kg-day DINP for 78 weeks
followed by a 26-week recovery period (Table 3-8). However, incidence of renal tubular carcinomas
only reached statistical significance in the recovery group.

Table 3-8. Incidence of Kidney Tumors in Male F344 Rats Exposed to DINP in the Diet for 2
Years (Covance Labs, 1998b) ahc	

Lesion

Dose Group
mg/kg-day (ppm)

Control

29 M / 36 F
(500)

88 M /109 F
(1,500)

359 M / 442 F
(6,000)

733 M / 885 F
(12,000)

High-Dose/ Recovery
637 M / 774 F
(12,000)

Renal tubular
carcinoma

0/65
(0%)

0/55
(0%)

0/55
(0%)

0/65
(0%)

2/65
(3.1%)

4/50*
(8.0%)

Source: U.S. CPSC (2001) text vv. 68-71 and AtroendixB.

* = statistically significant at p < 0.05 by one or more of the following: Fisher's Exact test, Poly-3, LogisticRegression,
or Life Table analysis.

11 Analysis of individual animal data as performed by NTP and reported in the textand Appendix B of U.S. CPSC
(2001).

h The high-dose/recovery group received 12,000 ppm for 78 weeks, followed by a 26-week recovery periodduring
which they received basal diet alone.

c Number of animals with neoplasm/ total number of animals examined. Percent tumor incidence in parentheses. Based
on extraction and analysis of individual animal data as reported in U.S. CPSC (2001)

Overall incidence for control (6,000 ppm and 12,000 ppm) groups (n = 65) includes incidence data for unscheduled
deaths, interim sacrifice at week 78 and terminal sacrifice. Overall incidence for the remaining groups includes
incidence data for unscheduled deaths and terminal sacrifice.

Lington et al. (1997) reported the incidence data for selected transitional cell carcinomas, transitional
cell adenomas, and tubular cell carcinomas and adenomas in the kidney (Table 3-9). Renal tubular cell
carcinomas were observed in one male in the low-dose group and two males in the high-dose group and
renal transitional cell carcinoma was observed in three male rats in the mid-dose group. However,
neither tumor type was statistically significantly increased. Further, no preneoplastic renal lesions were
detected in rats of either sex and no neoplastic lesions were detected in the kidneys of female rats.

Kidney tumors have not been observed in male or female SD rats treated with up to 553 to 672 mg/kg-
day DINP for 2 years (Bio/dynamics. 1987) or male and female B6C3F1 mice treated with up to 1,560
to 1,888 mg/kg-day DINP for 2 years (Covance Labs. 1998a).

Page 17 of 49


-------
Table 3-9. Incidence of Kidney Tumors in F344 Rats Exposed to DINP for 2 Years (Lington et al.,
1997; Bio/dynamics, 1986) 	

Lesion

Dose Group
mg/kg-day (ppm)

Control

15M/18F
(300)

152 M /184 F
(3,000)

307 M / 375 F
(6,000)

Males17

Transitional cell carcinoma

0/81 (0%)

0/80 (0%)

3/80 (3.8%)

0/80 (0%)

Transitional cell adenoma

0/81 (0%)

0/80 (0%)

0/80 (0%)

0/80 (0%)

Tubular cell carcinoma

0/81 (0%)

1/80(1.3%)

0/80 (0%)

2/80 (2.5%)

Tubular cell adenoma

0/81 (0%)

0/80 (0%)

0/80 (0%)

0/80 (0%)

Females'1

Transitional cell carcinoma

0/81 (0%)

0/81 (0%)

0/80 (0%)

0/80 (0%)

Transitional cell adenoma

0/81 (0%)

0/81 (0%)

0/80 (0%)

0/80 (0%)

Tubular cell carcinoma

0/81 (0%)

0/81 (0%)

0/80 (0%)

0/80 (0%)

Tubular cell adenoma

0/81 (0%)

0/81 (0%)

0/80 (0%)

0/80 (0%)

Source: Table 8 in Lington et al. (1997)

M = male; F = female

11 Number of animals with lesion/ total number of animals examined. Percent lesion incidence in parentheses.
h Statistically significant at p < 0.05 when compared to the control incidence using Fisher's Exact test; statistical
analysis performed by Lington et al. (1997).

3.2.3.1 Conclusions on Kidney Tumors

Two tumor types have been reported in the kidneys of male F344 rats following chronic oral exposure to
DINP, including renal transitional cell carcinomas and renal tubule cell carcinomas.

Renal transitional cell carcinoma, an uncommon tumor type in rats, has been reported in two out of four
rodent carcinogenicity studies. Lington et al. (1997) reported transitional cell carcinoma in 3 out of 80
mid-dose (151 mg/kg-day) male F344 rats. However, the response was not statistically significant and
did not occur in a dose-dependent manner (not observed in high-dose males [307 mg/kg-day]).

Similarly, in a study conducted by Covance Labs (1998b). transitional cell carcinoma was detected in 1
out of 65 male F344 rats treated with 359 mg/kg-day DINP; however, the response was not statistically
significant and did not occur in high-dose (733 mg/kg-day) or high-dose recovery (637 mg/kg-day)
males. Renal transitional cell carcinoma was not reported in male SD rats treated with up to 553 mg/kg-
day DINP (Bio/dynamics. 1987) or male B6C3F1 mice treated with up to 1,560 mg/kg-day DINP
(Covance Labs. 1998a) and has not been reported in female mice or rats at any dose. Given the lack of
dose-response and statistical significance across available studies, the low incidence of renal transitional
cell carcinomas observed in male F344 rats is considered to be of uncertain toxicological significance.

Renal tubule cell carcinomas have also been reported in two of four rodent carcinogenicity studies. In
the study conducted with F344 rats by Covance Labs (1998b). renal tubule cell carcinoma was observed
in 2 out of 65 high-dose (733 mg/kg-day) males and 4/50 recovery high-dose (637 mg/kg-day) males
compared to 0 out of 65 in the control group. The response in recovery males was statistically significant
relative to the control group. In the Lington et al. (1997) study, a statistically non-significant increase in
renal tubule cell carcinoma was observed in 1 out of 80 low-dose (15 mg/kg-day), 0 out of 80 mid-dose

Page 18 of 49


-------
(152 mg/kg-day), and 2 out of 80 high-dose (307 mg/kg-day) male F344 rats. Renal tubule cell
carcinomas were not observed in SD rats treated with up to 533 mg/kg-day DINP (Bio/dynamics. 1987)
or in male B6C3F1 mice treated with up to 1,560 mg/kg-day DINP (Covance Labs. 1998a). No
preneoplastic or neoplastic lesions were observed in female rats or mice at any dose.

The male rat specific alpha 2u-globulin (a2U-globulin) MOA has been implicated as being causative of
renal tubule cell carcinomas. U.S. EPA (1991)5 and the International Agency for Research on Cancer
(IARC) (1995)6 have published related criteria for establishing an a2u-globulin MOA for this tumor type.
EPA does not consider kidney tumors arising through a a2u-globulin MOA to be human relevant (U.S.
EPA. 1991). Data are available to support many, but not all of, the EPA and IARC criteria for an a2u-
globulin MOA. The three specific criteria for establishing an a2u-globulin MOA include demonstration
(1) that renal tubule cell carcinomas only occur in male rats, (2) immunohistochemical evidence, and (3)
histological evidence. In the case of DINP, these three requisites have been met across four chronic
studies: kidney tumors were only observed in male rats, and the weight of evidence indicates that DINP
is not genotoxic. Much of the additional evidence supporting a a2U-globulin MOA comes from Caldwell
et al.'s (1999) retrospective evaluation of archived kidney tissue taken from the 12-month interim
sacrifice from the chronic rat study conducted by Lington et al. (1997). Caldwell et al. report a dose-
dependent increase in the accumulation of a2U-globulin and increased droplet size in the kidneys of high-
dose male (but not female) rats. Cell proliferation measured via immunohistochemical staining for
proliferating cell nuclear antigen in kidney sections was not statistically significantly elevated in high-
dose males (125% of controls) or females (112% of control).

Photomicrographs for proliferating cell nuclear antigen and a2u-globulin staining showed foci of
proliferating cells and a2U-globulin accumulating in proximal tubule cells of the P2 segment; however,
some cell proliferation was also observed in PI and P3 cells. Histopathologic re-analysis of kidney
sections showed a dose-dependent increase in minimal tubular repair (incidence: 6/9, 10/10, 9/10, and
10/10 in control, low-, mid-, and high-dose males, respectively) and minimal tubular epithelial
hypertrophy (0/9, 0/10, 10/10, and 9/10 in control, low-, mid-, and high-dose males, respectively).
Tubular epithelial hypertrophy was not observed in control or high-dose females; however, minimal
tubular repair was observed in 1 out of 10 high-dose females. Collectively, Caldwell et al. concluded
that findings were consistent with an a2u-globulin MOA.

Additional histopathological findings consistent with an a2u-globulin MOA have been noted. For
example, a dose-related increase in incidence of mineralization of renal papilla was reported in the
kidneys of male, but not female, F344 rats in the chronic study conducted by Covance Labs (1998a).

Generally, EPA's three primary criteria for establishing an a2u-globulin MOA have been met. However,
data are not available to inform all of the IARC criteria and several findings raise uncertainty. First,
reversible binding of DINP to a2u-globulin has not been demonstrated. Additionally, chronic exposure to

5	EPA criteria include (1) an increase in number and size of hyaline (protein) droplets in kidney proximal tubule cells of
treated male rats; (2) immunohistochemical evidence of Obu-globulin accumulating protein in the hyaline droplets; and (3)
histopathological evidence of kidney lesions associated with obu-globulin nephropathology. The Agency also acknowledges
additional information that may be useful for the analysis that are consistent with IARC criteria (e.g., chemical is negative for
genotoxicity, reversible binding of chemical to Obu-globulin, sustained cell division in the proximal tubule of the male rat).

6	IARC criteria include (1) tumors occur only in male rats, (2) acute exposure exacerbates hyaline droplet formation, (3) ctu-
globulin accumulates in hyaline droplets, (4) intermediate lesions include granular casts and linear papillary mineralization,
(5) absence of hyaline droplets and other histopathological changes in female rats and mice, and (6) negative for
genotoxicity. Additional supporting evidence includes (1) reversible binding of chemical to Obu-globulin, (2) increased
sustained cell proliferation in proximal tubule (P2 segment), and (3) dose-response relationship between hyaline droplet
severity and renal tumor incidence.

Page 19 of 49


-------
DINP has been shown to increase absolute and relative kidney weight in both male and female rats
(Covance Labs. 1998b; Lington et al.. 1997; Bio/dynamics. 1987) as well as cause a significant dose-
related increase in chronic progressive nephropathy in female mice (Covance Labs. 1998a); however,
this lesion was not elevated in the high-dose recovery group females, indicating its reversibility. These
kidney effects cannot be explained by an a2U-globulin MOA.

Other agencies have evaluated the renal tubule cell carcinoma MOA. The U.S. CPSC (2010),7 Australia
NICNAS (2012),8 and ECHA (2013)9 have all concluded that the renal tubule cell carcinomas observed
in male rats occur through an a2u-globulin MOA that is not relevant for use in human health risk
assessment. Although Health Canada (EC/HC. 2Q15)10 concluded that certain effects observed in the
kidneys of female rats and mice cannot be explained by an a2U-globulin MOA, Health Canada
considered the kidney tumors in rodents to be of little or unclear relevance to humans. In contrast,
California OEHHA concluded that "a2U-globulin accumulation in the renal tubules of male rats do not
explain the renal tubule carcinomas observed in DINP-exposed rats" and that renal tubule cell
carcinomas were one of the tumor types listed to support the Proposition 65 listing of DINP (Tomaret
al.. 2013).

Although some uncertainty remains, much of the available literature supports an a2u-globulin MOA to
explain the incidences of renal tubule cell carcinomas observed in male rats exposed to DINP. EPA does
not consider kidney tumors arising through a a2u-globulin MOA to be human relevant (U.S. EPA. 1991).
Therefore, EPA did not consider it appropriate to derive quantitative estimates of cancer hazardfor
data on kidney tumors observed in these studies and did not farther consider kidney tumors as a factor
in the determination of the cancer classification for DINP. This conclusion was supported by the SACC.
In its final report to EPA, the SACC states "The Agency has provided substantial evidence that the
kidney tumors produced by DINP are due to a 2u-globulin MOA and correctly classified them as not
relevant to humans" (U.S. EPA. 2024).

3.2.4 Other Tumors	

The carcinogenicity of DINP was investigated in a Good Laboratory Practice (GLP)-compliant 2-year
dietary study in SD rats by Bio/dynamics (1987). Incidence data for select histopathological
observations and results from statistical analyses are provided in Table 3-10. In addition to findings in
the liver and kidney previously discussed, tumors were noted in the pancreas, testes, and uterus.
However, for these organs histopathologic examination was only conducted on control and high-dose
rats.

7	The U.S. CPSC concluded, "A small number of renal tubular cell carcinomas were observed only in males exposed to 1.2
percent DINP. Furthermore, there is experimental evidence that these tumors arose by a mechanism involving the
accumulation of a2u-globulin (Caldwell et al. 1999). a2u-Globulin is a protein that is specific to the male rat. Renal tubular
cell tumors induced by this mechanism are not considered relevant to human risk assessment (Schaeffer 1991)" (p.81 of (U.S.
CPSC. 2010))

8	Australia NICNAS concluded, "kidney tumours in male rats appear consistent with a specific gender- and species-specific
alpha 2|i-globiilin accumulation mechanism that is not regarded as relevant to humans" (p. 49 of (NICNAS. 2012)).

9	ECHA concluded, "The available new information on the carcinogenicity of DINP further supports the conclusions of the
EU Risk Assessment concerning renal tumors (EC 2003a). These neoplasms are assumed to have modes of actions which are
not considered to be relevant for humans (alpha-2u-globulin)" (p. 98 of (ECHA. 2013)).

111 Health Canada concluded "Renal tubular cell carcinomas were also reported in one chronic study in rats. It has been
suggested that the mechanism responsible for these tumours was related to accumulation of a2u-globulin, a protein specific to
the male rat (Health Canada 2015d). While this type of neoplastic lesion has not been observed in female rats, increased
kidney weights accompanied by histopathological changes were noted in female rats exposed for 2 years (Covance Labs.
1998b) and treatment-related nephropathy was noted in female mice in another chronic study conducted by the same author
(Covance Labs. 1998a). Those kidney effects cannot be explained by an a2u-globulin mode of action. Overall, findings in the
kidneys of rodents could be considered of little or unclear relevance to humans" (p. 95 of (EC/HC. 2015)).

Page 20 of 49


-------
Pancreatic islet cell adenomas (8/70 treated vs. 6/70 controls) and carcinomas (4/70 treated vs. 1/70
controls) were observed at a slightly higher incidence in the high-dose males compared to controls, and
the nonsignificant incidences of pancreatic tumors were considered to be within the range of normal
biological variation. Furthermore, in the females, pancreatic islet cell adenomas were only observed in
one high-dose and one control animals, and no pancreatic islet cell carcinomas were noted in females.

In the testes of SD rats, incidences of interstitial cell hyperplasia were significantly increased at the
high-dose (22/70) compared to controls (4/70) and were also reported to exceed historical controls.
Testicular interstitial cell tumors was increased at the high-dose (7/70) compared to controls (2/70);
however, the increase in tumors was not statistically significant and was reported to be within the range
of historical controls.

Similarly, in the uterus, incidence of endometrial hyperplasia was significantly increased at the high-
dose (13/69) compared to controls (2/70). Endometrial adenocarcinoma was observed in 2 out of 69
females at the high-dose compared to 0 out of 70 controls; however, the increase in tumors was not
statistically significant.

It is plausible that the significantly increased incidences of hyperplasia noted in the testes and uterus at
the high-dose are proliferative responses that can lead to the slight (not significant) increases in
testicular and uterine tumors. However, the fact that the incidences of these tumors is low and, for the
testes data, within the range of historical controls, there is not strong evidence of a carcinogenic
response. Furthermore, the lack of examination of the low- and mid-dose groups limits the examination
of dose-dependency for the cancer incidence in these organs and may miss low-dose effects on any
hormonally-influenced tumors or receptor-mediated carcinogenicity. Finally, tumors in the testes and
uterus were not noted in other chronic studies of DINP in rodents; however, the background rate of
interstitial cell tumors is high in male F344 rats making it difficult to detect treatment-related increases
in this tumor type in this strain of rat (King-Herbert et al.. 2010; King-Herbert and Thayer. 2006).
Overall, there is too much uncertainty for EPA to consider using these data to derive quantitative
estimates of cancer risk.

Page 21 of 49


-------
Table 3-10. Incidence of Tumors in Pancreas, Testes, and Uterus in SD Rats Exposed to DINP for 2 Years (Bio/dynamics, 1987)"

Observation

Dose Group
mg/kg-day (ppm)

Males

Females

0

27 (500)

271 (5,000)

553 (10,000)

0

33 (500)

331 (5,000)

672 (10,000)

Pancreas

No. examined

70

0

0

70

69

0

0

70

Pancreatic islet cell adenoma

-

6

-

-

8

1

-

-

1

Pancreatic islet cell carcinoma

-

1

-

-

4

0

-

-

0

Testes

No. examined

69

0

0

70

N/A

N/A

N/A

N/A

Interstitial cell hyperplasia

Total

4

-

-

22*

-

-

-

-

Unilateral

3

-

-

9

-

-

-

-

Bilateral

1

-

-

13

-

-

-

-

Interstitial cell tumors

Total

2

-

-

7

-

-

-

-

Unilateral

2

-

-

6

-

-

-

-

Bilateral

0

-

-

1

-

-

-

-

Uterus

No. examined

N/A

N/A

N/A

N/A

70

0

0

69

Endometrial hyperplasia

-

-

-

-

-

2

-

-

13*

Endometrial adenocarcinoma

-

-

-

-

-

0

-

-

2

* p < 0.05 based on a two-tailed Fisher's exact test calculated for this review.

" Data in this table indicate all animals assessed for histopathology throughout the study; that is, including the interim sacrifice, the terminal sacrifice, and unscheduled
deaths. For late-developing tumors (pancreatic islet cell tumors, testicular interstitial cell tumors), statistical analysis was performed excluding animals that died or were
sacrificed up to 12 months, leaving n = 57, 57, 59, and 59 in males and n = 59, 56, 60, and 59 in females in the control, low-, mid- and high-dose groups, respectively.
Data from Appendix K of (Bio/dynamics, 1987).

Page 22 of 49


-------
4 POSTULATED MODE OF ACTION FOR LIVER TUMORS IN
RATS AND MICE

As described in Section 3.2.1, available studies provide consistent evidence that chronic oral exposure to
DINP can cause treatment-related hepatocellular adenomas and/or carcinomas in male and female F344
and SD rats and male and female B6C3F1 mice. EPA further considers the weight of evidence for liver
carcinogenesis and its underlying MOA in Sections 4.1 through 4.9.

4.1 Postulated Mode of Action in Rats and Mice	

Studies have demonstrated that DINP and its metabolite monoisononyl phthalate (MINP), can activate
peroxisome proliferator-activated receptor alpha (PPARa) in hepatocytes and cause hepatocellular
adenomas and carcinomas in mice and rats. Existing assessments of DINP by U.S. CPSC (2014. 2010),
Health Canada (ECCC/HC. 2020; EC/HC. 2015; Health Canada. 2015). ECHA (2013). and NICNAS
(2012) have postulated that DINP causes liver tumors in rats and mice through a PPARa MOA. In
contrast, California OEHHA has concluded that "PPARa activation may not be causally related to
DINP-induced liver tumors in rats and mice" (OEHHA. 2013; Tomar et al.. 2013). PPARa is a nuclear
receptor that controls transcription of genes involved in fatty acid P-oxidation and peroxisome
proliferation. PPARa activation in hepatocytes in rodent models can cause hepatocellular cancer through
a non-genotoxic MOA that involves activation of Kupfer cells. Activated Kupfer cells secrete cytokines
such as TNFa, IL-la, and IL-ip that influence hepatocyte growth and fate. As discussed by Corton et al.
(2018; 2014) and Klaunig et al. (2003). studies have demonstrated that Kupffer cell activation following
PPARa activation plays a crucial role in several tumor precursor effects. These effects include increased
DNA synthesis and cell proliferation in both normal and preneoplastic hepatocytes, as well as
suppression of apoptosis. Altered cell growth and survival can facilitate clonal expansion of initiated
cells leading to the selective clonal expansion of preneoplastic foci cells and ultimately tumor formation.

The PPARa MOA for liver tumorigenesis considered by EPA is described further by Corton et al.
(2018; 2014) and Klaunig et al. (2003). Consistent with EPA's Guidelines for Carcinogen Risk
Assessment (U.S. EPA. 2005) and the IPCS' Mode of Action Framework (IPCS. 2007). EPA further
evaluated the postulated PPARa MOA for liver tumors, as well as evidence for other plausible MO As
for DINP and its metabolite MINP.

The PPARa MOA includes the following sequence of key events (KEs):

•	KE1: activation of PPARa in hepatocytes;

•	KE2: alterations in cell growth pathways (e.g., Kupfer cell activation leading to increased
cytokine (e.g., TNFa, IL-la, IL-ip) secretion;

•	KE3: perturbation of cell growth and survival (i.e., increased cell proliferation and inhibition of
apoptosis); and

•	KE4: selective clonal expansion of preneoplastic foci cells leading to the apical outcome,
hepatocellular adenomas, and carcinomas.

Several modulating factors associated with the PPARa MOA have also been proposed, including
increases in reactive oxygen species (ROS) and activation of nuclear factor kappa B (NF-kB) (Corton et
al.. 2018). These modulating factors are not considered necessary to induce liver tumorigenesis but
might modulate the dose-response behavior or the probability of inducing one or more KEs (Corton et
al.. 2014V

Evidence for each KE (Sections 4.1.1 to 4.1.4) and EPA's analyses of dose-response (Section 4.1.5);
temporality (Section 4.3); strength, consistency, and specificity (Section 4.4); biological plausibility and

Page 23 of 49


-------
coherence (Section 4.5); other carcinogenic MOAs (Section 4.6); uncertainties and limitations (Section
4.7); weight of scientific evidence for liver tumors (Section 4.8) are presented below.

4.1.1 Key Event 1: PPARa Activation

PPARa activation can be assessed using trans-activation assays or by measuring specific events
associated with PPARa activation, such as increased expression of genes involved in beta oxidation or
peroxisome proliferation, increased activity of palmitoyl-CoA oxidase, increased peroxisomal beta
oxidation (PBOX), and/or peroxisome proliferation in hepatocytes. Activation of PPARa in hepatic cells
by DINP has been consistently demonstrated in six in vivo studies of mice and four in vivo studies of
rats. No evidence of PPARa activation in hepatic cells was observed in two in vivo studies of monkeys.
Additionally, six in vitro studies investigating PPARa activation are available. Available data for KE1
are discussed further below.

Evidence from In Vitro Studies

Four in vitro studies of DINP and two in vitro studies of MINP, a metabolite of DINP, are available that
consistently demonstrate that rat and mouse hepatocytes are more sensitive to PPARa activation
compared to human and monkey hepatocytes. Bendford et al. (1986) demonstrated that in vitro
treatment of primary rat hepatocytes isolated from adult Wistar albino rats with concentrations of MINP
ranging from 0.1 to 0.5 mM for 3 days caused large (up to -750%) dose-dependent increases in
palmitoyl-CoA oxidation and laurate hydroxylation activity. Comparatively, smaller (-200-300%)
increases in palmitoyl-CoA oxidation and laurate hydroxylation activity were observed in primary
hepatocytes from marmoset monkeys under similar experimental conditions. Hasmall et al. (1999)
demonstrated that treatment of primary rat hepatocytes isolated from male F344 rats with 250 and 500
|iM (but not 750 |iM) DINP can induce increases in PBOX activity. In contrast, no increase in PBOX
was noted in primary human hepatocytes treated with up to 750 |iM DINP under similar experimental
conditions.

Similarly, Shaw et al. (2002) report dose-related induction of PBOX activity in primary rat hepatocytes
isolated from male F344 rats treated with 150 to 250 |iM MINP, however, PBOX activity was not
increased in primary human hepatocytes treated with up to 250 |iM MINP under similar experimental
conditions. Bility et al. (2004) demonstrated that mouse PPARa is more inducible and activated at lower
concentrations compared to human PPARa in mouse 3T3-L1 fibroblasts transfected with a plasmid
encoding mouse or human PPARa luciferase reporter (lowest activation concentration: 3 and 10 |iM for
mouse and human, respectively; maximal fold-induction: 27.1 and 5.8 for mouse and human,
respectively). Laurenzana et al. (2016) demonstrated that MINP can activate human PPARa in COS-1
cells transfected with a human PPARa luciferase reporter. Briefly, transfected cells were treated with 0,
0.1, 1, 10, or 100 |iM MINP for 24 hours. An approximate two-fold increase in MINP stimulated
receptor activation was observed at the highest concentration. Finally, Kamenduliz et al. (2002)
demonstrated that three different isomers of MINP (MINP-1, MINP-2, MINP-3) can increase PBOX in
primary hepatocytes isolated from female B6C3F1 mice and male F344 rats treated with concentrations
of each isomer ranging from 10 to 300 |iM for 72-hours, but not in primary hepatocytes isolated from
male Syrian golden hamsters, male cynomolgus monkeys, or human donor liver tissue.

Evidence from In Vivo Studies of Rats

Three studies of rats provide consistent evidence of treatment-related increases in PPARa activation
following oral exposure to DINP. Smith et al. (2000) reported treatment-related increases in hepatic
PBOX in male F344 rats fed diets containing up to 12,000 ppm DINP (-1,200 mg/kg-day) for 2 or 4
weeks; however, no change was observed in the low-dose group (-100 mg/kg-day). Similarly, BIBRA
(1986) reported increased hepatic cyanide-insensitive palmitoyl-CoA oxidation levels and hepatic lauric

Page 24 of 49


-------
acid 11- and 12-hydroxylase activities, as well as marked to very marked increases in hepatic
peroxisomes (evaluated via transmission electron microscopy), in male and female F344 rats treated
with high-doses of DINP for 21-days (biomarkers of PPARa activation increased in males and females
starting at 639 and 1,198 mg/kg-day, respectively). Finally, cyanide-insensitive palmitoyl-CoA oxidase
activity was increased in the livers of male and female F344 rats treated with 733 (males) to 885
(females) mg/kg-day DINP after 1,2, 13, and 104 weeks of exposure to DINP, as well as for females
treated with 442 mg/kg-day DINP for 104 weeks (Covance Labs. 1998b). In contrast, no evidence of
peroxisome proliferation (evaluated via transmission electron microscopy) was reported in hepatocytes
from male or female F344 rats treated with up to 307 (males) or 375 mg/kg-day DINP (females) for 2
years (Lington et al.. 1997).

Evidence from In Vivo Studies of Mice

Six studies of mice provide consistent evidence of treatment-related increases in PPARa activation
following oral exposure to DINP. Smith et al. (2000) reported treatment-related increases in hepatic
PBOX in male B6C3F1 mice fed diets containing up to 6,000 ppm DINP (-900 mg/kg-day) for 2 or 4
weeks; however, no change was observed in the low-dose group at either timepoint (-75 mg/kg-day). In
a second study, Kaufmann et al. (2002) reported dose-related increases in the number and volume of
peroxisomes and hepatic cyanide-insensitive palmitoyl-CoA oxidation activity in male B6C3F1 mice
after 4 weeks at doses as low as 117 mg/kg-day, while similar changes were observed in female mice
starting at 546 mg/kg-day DINP. Similarly, Valles et al. (2003) reported treatment related increases in
hepatic palmitoyl-CoA oxidase activity in male and female B6C3F1 mice treated with diets containing
4,000 to 8,000 ppm DINP for 2 weeks (equivalent to approximately 600 to 1,200 mg/kg-day). In a study
by Hazleton Labs (1992). large (albeit not always statistically significant), dose-related, increases in
hepatic cyanide-insensitive palmitoyl CoA oxidation were observed in male and female B6C3F1 mice
treated with 365 and 2,600 mg/kg-day DINP for 4, 31, and 91 days. Similarly, large increases in hepatic
cyanide-insensitive palmitoyl-CoA oxidation activity were observed in male and female B6C3F1 mice
treated with 1,560 (males) to 1,888 (females) mg/kg-day DINP for 79 and 105 weeks (Covance Labs.
1998a).

ToxStrategies (2024) reported microarray data from archived mouse liver tissue samples from a study
originally conducted in 1999 at the Chemical Industry Institute of Toxicology (CUT). Briefly, wild-type
male mice (12-weeks of age) were orally gavaged with either vehicle control, 50 mg/kg-day WY14643
(WY), 1000 mg/kg-day DINP (low-dose), or 2000 mg/kg-day DINP (high-dose) for 1 or 3 weeks. RNA
was isolated from archived samples and analyzed by Affymetrix GeneTitan array analyzer. A total of
33, 43, and 330 genes were differentially expressed in low-dose, high-dose, and WY groups of
compared to controls. In the low- and high-dose DINP groups, the top up-regulated gene was Cyp4al4,
which is a target gene of PPARa, and is involved in fatty acid metabolism. Other significantly up-
regulated genes in both DINP treatment groups were involved in lipid and fatty acid metabolism, and
PPARa activation, including Cyp4alO, Cyp4a31, Vcminl, Acotl/2, and Accial/2. Only five gene sets
were enriched in both DINP dose groups, with the top gene sets enriched in both DINP treatment groups
being related to fatty acid metabolism, protein localization, and organic hydroxy compound metabolic
process. Comparatively, 58 gene sets were enriched in the WY group, with gene sets mostly related to
fatty acid and lipid metabolism, peroxisome activity, and PPAR signaling. Overall, PPARa was the top
(WY and high-dose DINP groups) or only (low-dose DINP group) transcriptional regulator predicted for
all three treatment groups. However, there were some sources of uncertainty that may impact the
interpretation of the results. For instance, the ToxStrategies report noted uncertainty pertaining to the
duration of exposure; ToxStrategies indicated that conflicting durations of 1 and 3 weeks were listed in
the original study. Additionally, a considerable length of time had elapsed between the original study
and the microarray analysis by ToxStrategies, although the authors did evaluate mRNA quality and

Page 25 of 49


-------
excluded low quality samples. Despite these and other uncertainties in study design, this study provides
evidence that the primary response in the liver following 1 or 3 weeks of exposure to high doses of
DINP (i.e., 1,000 and 2,000 mg/kg-day) are related to PPARa activation. Consistent with DINP being a
weaker PPARa activator compared to WY, the transcriptomic response in the livers of DINP treated
mice was lower than that of the potent PPARa activator, WY.

Evidence from In Vivo Studies of Monkeys

Two studies have evaluated biomarkers of PPARa activation in monkeys. Oral (gavage) exposure to
DINP had no effect on PBOX in male cynomolgus monkeys treated with 500 mg/kg-day DINP for 14-
days (Pugh et al.. 2000). Similarly, no effect on cyanide-insensitive palmitoyl CoA oxidase activity or
cytochrome P450 concentration and lauric acid 11- and 12-hydroxylase activities in hepatic microsomes
were observed in male and female marmosets gavaged with up to 2,500 mg/kg-day DINP for 13 weeks
(Hall et al.. 1999).

4.1.2	Key Event 2: Alterations in Cell Growth Pathways

EPA identified one in vivo study of mice investigating alterations in cell growth pathways. No in vivo
studies of rats or monkeys for KE2 were identified. Ma et al. (2014a) administered DINP via oral
gavage to male Kunming mice at 0, 0.2, 2, 20, and 200 mg/kg-day DINP daily for 14 days and then
determined TNFa and IL-1 in liver homogenates. IL-1 and TNFa content were significantly increased at
20 and 200 mg/kg-day. However, the study authors do not identify the specific IL-1 subtypes evaluated
(e.g., IL-la vs. IL-ip), which is an important consideration when interpreting these results.

4.1.3	Key Event 3: Perturbation of Cell Growth and Survival

Evidence of increased cell proliferation comes from five in vivo studies of mice, two in vivo studies of
rats, one in vivo study of monkeys, and two in vitro studies of primary rat and human hepatocytes.
Across in vivo studies of mice and rats, an acute cell proliferative response in the liver has been
consistently observed. In contrast, cellular proliferation in the liver is not sustained chronically in either
species. However, as discussed by Corton et al. (2018). weak PPARa activators tend to "produce
transient increases in replicative DNA synthesis during the first few days or weeks of exposure followed
by a return to baseline levels." Therefore, lack of a sustained proliferative response is consistent with the
proposed MOA. No evidence of replicative DNA synthesis was observed in one in vivo study of
monkeys. In the two in vitro studies, DINP consistently suppressed apoptosis and increased replicative
DNA synthesis in rat, but not human hepatocytes. Available data for KE3 are discussed further below.

Evidence from In Vitro Studies

Two in vitro studies are available that consistently demonstrate that DINP can suppress apoptosis and
increase replicative DNA synthesis in rat but not human hepatocytes. Hasmall et al. (1999) treated
primary rat hepatocytes obtained from male F344 rats and primary human hepatocytes with 250 to
750 |iM DINP. Treatment with DINP increased replicative DNA synthesis, suppressed apoptosis, and
suppressed TGFpi-induced apoptosis in rat but not human hepatocytes. Similarly, Shaw et al. (2002)
treated primary rat hepatocytes obtained from male F344 rats and primary human hepatocytes with 150
to 250 |iM DINP and observed treatment-related suppression of apoptosis and increased replicative
DNA synthesis in rat but not human hepatocytes.

Evidence from In Vivo Studies of Rats

Two studies of rats have evaluated cell proliferation in the liver following oral exposure to DINP. In
both studies, bromodeoxyuridine (BrdU) was administered to rats via osmotic minipumps and cell
proliferation was evaluated via BrdU labeling. No in vivo studies of rats have evaluated effects on
hepatocyte apoptosis. Smith et al. (2000) reported treatment-related increases in hepatocellular

Page 26 of 49


-------
replicative DNA synthesis in male F344 rats fed diets containing 12,000 ppm DINP (-1,200 mg/kg-day)
for 2 or 4 weeks; however, no change was observed in the low-dose group (-100 mg/kg-day). In the
second study, increased hepatocellular replicative DNA synthesis was observed in male and female
F344 rats after 1 week of dietary exposure to 733 (males) or 885 (females) mg/kg-day DINP, but not
after 2, 13, or 104 weeks of exposure (Covance Labs. 1998b).

Evidence from In Vivo Studies of Mice

Five studies have evaluated cell proliferation (measured via BrdU labeling in all five studies) and/or
apoptosis in the liver following oral exposure to DINP. Valles et al. (2003) fed female B6C3F1, SV129,
and Ppara-null mice diets containing 8,000 ppm DINP (-1,200 mg/kg-day) for 1 week and observed
increased hepatocellular replicative DNA synthesis in B6C3F1 and SV129 mice, but not Ppara-null
mice. Smith et al. (2000) report treatment-related increases in hepatocellular replicative DNA synthesis
in male B6C3F1 mice fed diets containing up to 6,000 ppm DINP (-900 mg/kg-day) for 2 but not 4
weeks. Further, no change in replicative DNA synthesis was observed in the low-dose group at either
timepoint (-75 mg/kg-day). Two other studies reported no increase in hepatocellular replicative DNA
synthesis in the livers of male or female B6C3F1 mice dosed with 2,600 mg/kg-day DINP for 4, 31, and
91 days (Hazleton Labs. 1992) or 1,560 (males) to 1,888 (females) mg/kg-day DINP for 79 and 105
weeks (Covance Labs. 1998a).

In another study, Kaufmann et al. (2002) evaluated hepatocellular replicative DNA synthesis and
apoptosis (via TUNEL staining) in male and female B6C3F1 mice administered 117 to 2,806 mg/kg-day
DINP for 1 or 4 weeks. Dose-related increases in hepatocellular replicative DNA synthesis were
observed in male and female mice after 1 week at doses as low as 116 (male) to 1,272 (female) mg/kg-
day; however, no significant changes in females were noted after 4 weeks at doses as high as 2,806
mg/kg-day, while significant increases in males after 4 weeks were observed at doses as low as 117
mg/kg-day but without a clear dose-response relationship. In males, apoptosis was increased after
1 week in the high-dose group (1,860 mg/kg-day). At 4 weeks, apoptosis appeared reduced in all
treatment groups for males; however, the effect was not statistically significant. No clear treatment-
related effects on apoptosis were observed for females at either timepoint.

Evidence from In Vivo Studies of Monkeys

Treatment with DINP had no effect on replicative DNA synthesis (measured via proliferating cell
nuclear antigen [PCNA] immunohistochemistry) in male cynomolgus monkeys treated with 500 mg/kg-
day DINP for 14 days (Pugh et al.. 2000).

4.1.4	Key Event 4: Selective Clonal Expansion of Preneoplastic Foci	

EPA identified no in vitro or in vivo studies of DINP that evaluated KE4. Further, hepatocellular
hyperplasia, which may provide some evidence of expansion of preneoplastic foci, has not been reported
in any short-term, subchronic, or chronic studies of DINP.

4.1.5	Modulating Factors

EPA identified no studies evaluating activation of NF-kB in the liver.

Three studies provide data on the relationship between oxidative stress and DINP following in vivo
exposures in male Kunming mice (Ma et al.. 2014b) and male Balc/c mice (Liang and Yan. 2020) or in
vitro investigations in human hepatic cell-types (Gutierrez-Garcia et al.. 2019). Available studies
provide evidence that DINP can induce ROS in the liver.

Page 27 of 49


-------
Ma et al. (2014b) exposed male Kunming mice to DINP via oral gavage daily for 14 days and evaluated
several endpoints related to oxidative stress in homogenized hepatic tissue. Indices of oxidative stress
were generally observed at the same doses that resulted in histopathological lesions of the liver, although
quantification of the tissue sections was not performed. Dose-dependent increases in ROS and increases
in malondialdehyde were observed, reaching significance at 200 mg/kg-day. In parallel, decreases in
glutathione content occurred at 20 mg/kg-day DINP and above, indicative of oxidative stress. The
authors also reported DNA-protein-crosslinks and increases in 8-hydroxydeoxyguanosine at 200 mg/kg-
day, which indicate oxidative damage to DNA.

Liang and Yan (2020) exposed male Balb/c mice dermally to 0, 0.02, 0.2, 2, 20, and 200 mg/kg-day
DINP for 28 days and then evaluated several endpoints related to oxidative stress in homogenized liver
tissue. Dose-dependent increases in ROS and increases in malondialdehyde were observed at 20 mg/kg-
day DINP and above, while levels of glutathione (GSH) decreased at 20 mg/kg-day DINP and above.
Additionally, increased DNA-protein-crosslinks were observed at 200 mg/kg-day, which indicate
oxidative damage to DNA. These data are consistent with increases in oxidative stress at doses of 20
mg/kg-day and above. In parallel, significant increases in relative liver weight were observed at 20
mg/kg-day DINP and above, which may have been associated with activation of PPARa; however, no
specific biomarkers of PPARa activation were evaluated in this study.

As discussed in the Non-cancer Raman Health Hazard Assessment for Diisononyl Phthalate (DINP)
(U.S. EPA. 2025). dermal absorption of DINP is low (i.e., 2-4% over 7 days), which suggests that the
dose of absorbed DINP that caused oxidative stress in the study by Liang and Yan (2020) is much lower
than the dermally applied dose of 20 mg/kg-day. However, there are several sources of uncertainty
associated with the study by Liang and Yan (2020) with regard to the actual received doses in the study,
as only nominal doses are provided. Liang and Yan state that 20 |iL of test solution (concentration of
applied test solution not provided) was applied evenly to a 2 cm2 area of exposed skin on the center of
the back of the mouse, however, additionally methodological details pertaining to how DINP was
dermally administered were not provided. For example, study authors do not provide information
relating to how hair was removed from the backs of mice and whether or not care was taken to avoid
applying solutions of DINP to abraded skin, which would be expected to increase dermal absorption;
how frequently DINP solutions were applied and whether DINP was washed from the skin at the
application site between dermal applications; and whether skin was covered with a bandage to help limit
evaporation, as well as oral ingestion of DINP through grooming.

An in vitro study in HepG2 cells by Gutierrez-Garcia et al. (2019) evaluated the potential for DINP to
elicit oxidative stress and investigated a mechanism involving sirtuins (srts), which are a group of
mitochondrial NAD+-dependent histone deacetylases. Increases in ROS were observed at the highest
concentration tested in parallel with increases in lysine acetylation and dose-dependent reductions in
expression of several sirtuin genes (i.e., Sirtl, Sirt2, Sirt3, SirtS) as well as decreases in sirtuin protein
levels. Although the data does not directly provide evidence that ROS is a modulating factor within the
PPARa activation MOA for hepatic tumors, considered more broadly, it does suggest that DINP can
induce ROS in hepatocytes.

4.2 Dose-Response Concordance of Key Events with Tumor Response

Dose-Response Concordance: Rats

As discussed in Sections 4.1.1 through 4.1.4, data from in vivo rat studies is limited to KE1, KE3, and
the apical outcome, hepatocellular adenomas and/or carcinomas. No data is available for KE2 or KE4.
Available data used by EPA for its dose-response concordance analysis of the PPARa MOA in rats is
presented in Table 4-1.

Page 28 of 49


-------
Although limited, there is some evidence to demonstrate that KE1 occurs at lower doses than KE2 and
the apical outcome, liver tumors. For KE1, three studies reported consistent dose-related increases in
several biomarkers of PPARa activation (i.e., increased PBOX, lauric acid 11- and 12-hydroxylase, and
palmitoyl-CoA oxidase activity) (Smith et al.. 2000; Covance Labs. 1998b; BIBRA. 1986). The lowest
dose at which PPARa activation was reported in rats is 442 mg/kg-day, following 104 weeks of
exposure to DINP (Covance Labs. 1998b). For KE3, one study reported a dose-related increased in
hepatocellular replicative DNA synthesis at very high doses of DINP (i.e., 1,200 mg/kg-day) after 2 and
4 weeks of exposure (Smith et al.. 2000). A second study, which only evaluated hepatocellular
replicative DNA synthesis at a single dose (i.e., 733 and 885 mg/kg-day in males and females,
respectively), reported increased hepatocellular replicative DNA synthesis and palmitoyl-CoA oxidase
activity after 1 week of exposure (Covance Labs. 1998b). Statistically significant dose-related increases
in hepatocellular carcinomas and/or combined adenomas and carcinomas have been observed in two
studies of rats at doses at low as 672 to 885 mg/kg-day (Covance Labs. 1998b; Bio/dynamics. 1987). In
the study of F344 rats by Covance Labs (1998b). increased hepatic palmitoyl-CoA oxidase activity
(KE1) was observed in female (but not male) rats at lower doses than which adenomas and carcinomas
were observed after 104 weeks of treatment (i.e., 442 vs. 885 mg/kg-day for tumors), providing evidence
of concordance.

Overall, there is some evidence to support dose-response concordance for KE1, KE3, and hepatocellular
adenomas and/or carcinomas. However, no data are available for KE2, KE4, or apoptosis (part of KE3)
in rat hepatocytes, which prevents a complete analysis of dose-response concordance across all KEs in
the postulated MOA.

Dose-Response Concordance: Mice

As discussed in Sections 4.1.1 through 4.1.4, data from in vivo mouse studies is limited to KE1, KE2,
KE3, and the apical outcome, hepatocellular adenomas and/or carcinomas. No data is available for KE4.
Available data considered by EPA for its dose-response concordance analysis of the PPARa MOA in
mice is presented in Table 4-2.

Although limited, available data indicate the KE1, KE2, and KE3 occur in mice at lower doses than
hepatocellular adenomas and/or carcinomas, providing some evidence of concordance. However,
concordance across KE1, KE2, and KE3 is less apparent. As can be seen from Table 4-2, the lowest
dose at which biomarkers of PPARa activation were increased was 117 mg/kg-day for male mice after 4
weeks of exposure (Kaufmann et al.. 2002); for KE2 increased TNFa and IL-1 in liver homogenate has
been observed at doses as low as 20 mg/kg-day (Ma et al.. 2014a); for KE3 increased DNA synthesis
has been reported at doses as low as 116 mg/kg-day in male mice (Kaufmann et al.. 2002); and
hepatocellular adenomas and carcinomas have been observed at doses as low as 336 mg/kg-day in
female mice. However, there are several sources of uncertainty related to KE2 data from Ma et al.
(2014a). First, Ma et al. evaluated DINP exposure with Kunming mice, while other studies of DINP
were performed with B6C3F1 mice, and it is unclear if there is a strain difference in sensitivity or if
studies testing lower doses of DINP with B6C3F1 mice would produce similar results. Additionally, Ma
et al. report increased IL-1 in liver homogenate, but do not differentiate between cytokine subtypes (e.g.,
IL-1 a, IL-ip). Another limitation of the available data set is that PBOX is generally not considered as
sensitive of a biomarker as other measures of PPARa activation—especially compared to measures of
PPARa-inducible genes.

Page 29 of 49


-------
Table 4-1.

)ose-Response Concordance for PPARa M<

3 A in Rats

Dose
(mg/kg-
day)

KE 1

(Sex; Dose in mg/kg-day; Timepoint)

KE 2

KE3

(Sex; Dose in mg/kg-day; Timepoint)

KE 4

Hepatocellular Tumors

1-200

NC - PBOX (M; 120; 2, 4 weeks)a



NC - DNA synthesis (M; 120; 2, 4
weeks)"



NC - Neoplastic nodules, hepatocellular cancer,
or combined (M/F; 15-184; 104 weeks) d
NC - Adenomas, carcinomas, combined (M/F,
29-109; 104 weeks)b

NC - Neoplastic nodules, carcinoma (M/F; 27-
33; 104 weeks)e

201-400

NC - PP (M/F; 307-375; 2 years) d







NC - Neoplastic nodules, hepatocellular cancer,
or combined (M/F; 307-375; 104 weeks) d
NC - Adenomas, carcinomas, combined (M/F,
359-442; 104 weeks)b

NC - Neoplastic nodules, carcinoma (M/F; 271-
331; 104 weeks)e

401-600

| Palm CoA (F (not M); 442; 104 (but not

1,2, or 13) weeks)b

-

-

-

-

601-1,000

NC - Palm CoA (M/F;607-639; 3 weeks)c
t 11/12 H-lase (M, not F); 639; 3 weeks)c
t Palm CoA (M/F; 733-885; 1, 2, 13, 104
weeks)*



t DNA synthesis (M/F; 733-885; 1
week (but not 2, 13, 104 weeks)b



t Carcinoma (F (not M); 672; 104 weeks)e
t Carcinoma (M (not F); 733-885; 104 weeks)b
t Combined adenoma and carcinoma (M/F); 733—
885; 104 weeks)b

1,001-1,400

t Palm CoA (M/F; 1,192-1,198; 3 weeks)c
t 11/12 H-lase (M, not F); 1,192; 3 weeks)c
t PBOX (M; 1,200; 2, 4 weeks)a



| DNA synthesis (M; 1,200; 2, 4
weeks)"





1,401-2,000



-

-

-

-

2,001-2500

t 11/12 H-lase (M/F; 2,195-2,289; 3 weeks)c
t PP (M/F; 2,195-2,289; 3 weeks)c

-

-

-

-

11/12 H-lase = lauric acid 11- and 12-hydroxylase; F = female; M = male; NC = no significant change; Palm CoA: cyanide-insensitive palmitoyl-CoA oxidation; PBOX
= peroxisomal beta-oxidation; PP = peroxisomal proliferation

indicates no experimental evidence is available
° (Smith et al.. 2000)
h (Covance Labs. 1998b)

<¦ (BIBRA. 1986)
d (Lineton et al.. 1997)
e (Bio/dvnamics. 1987)

Page 30 of 49


-------
Table 4-2. Dose-Response Concordance for PPARg MOA in Mice

Dose
(mg/kg-day)

KE 1

(Sex; Dose in mg/kg-day; Timepoint)

KE 2

KE3

(Sex; dose in mg/kg-day; timepoint)

KE 4

Hepatocellular Tumors

1-200

NC - PBOX (M; 75; 2,4 weeks)"
t PP & Palm CoA (M (but not F); 117; 4
weeks)b

NC - TNFa (M,
0.2-2, 2 weeks)f
t TNFa (M, 20-
200, 2 weeks)f

NC - DNA synthesis (M; 75; 2, 4 weeks) °
t DNA synthesis (M (but not F); 116-167; 1, 4 weeks)

b

NC - Apoptosis (M/F; 116-167; 1, 4 weeks)b



NC - Adenomas or
carcinomas (M/F; 90-112,
2 years) d

201-400

t PP & Palm CoA (M; 350; 4 weeks)6
t Palm CoA (M/F; 365; 4, 31, 91 days)c



| DNA synthesis (M; 337-350; 1, 4 weeks)b
NC - Apoptosis (M; 337-350; 1, 4 weeks)b



t Combined adenomas &
carcinomas (F (but not M);
336, 2 years) d

401-600

| PP & Palm CoA (F; 546; 4 weeks)b
| Palm CoA (M/F; 600; 2 weeks)e

-

NC - DNA synthesis (F; 520-546; 1, 4 weeks)b
NC - Apoptosis (F; 520-546; 1, 4 weeks)b

-

-

601-800









t Combined adenomas &
carcinomas (M; 742, 2
years) d

801-1,000

t PBOX (M; 900; 2,4 weeks)a
t PP & Palm CoA (M; 913; 4 weeks)b
t gene expression (M; 1,000; 1 or 3
weeks)g



t DNA synthesis (M; 75; 2 (not 4) weeks) °
| DNA synthesis (M; 901-913; 1,4 weeks)b
NC - Apoptosis (M; 901-913; 1,4 weeks)b



t Carcinomas and
combined adenomas &
carcinomas (F; 910, 2
years) d

1,001-1,400

| Palm CoA (M/F; 1,200; 2 weeks)e
t PP & Palm CoA (F; 1,272; 4 weeks)b



| DNA synthesis (F; 1200; 1 week)e
t DNA synthesis (F; 1272-1278; 1 (but not 4) weeks)b
NC - Apoptosis (F; 1272-1278; 1, 4 weeks)b





1,401-2,000

t PP & Palm CoA (M; 1,860; 4 weeks)b
t Palm CoA (M/F; 1,560-1,888; 79, 105
weeks) d

t gene expression (M; 1,000; 1 or 3
weeks)g



| DNA synthesis (M; 1766-1860; 1, 4 weeks)b
NC - DNA synthesis (M/F; 1,560-1,888; 79, 105
weeks) d

t Apoptosis (M; 1,766-1,860; 1 (but not 4) weeks)b



t Adenomas and/or
carcinomas (M/F; 1,560-
1,888, 2 years) d

2,001-3,000

t Palm CoA (M/F; 2600; 4, 31, 91 days)

c

t PP & Palm CoA (F; 2806; 4 weeks)6



| DNA synthesis (F; 2593-2806; 1 (but not 4) weeks)b
NC - DNA synthesis (M/F; 2,600; 4, 41, 91 days)c
NC - Apoptosis (F; 2,593-2,806; 1, 4 weeks)b





indicates no experimental evidence is available; t = significant increase; j = significant decrease; 11/12 H-lase = lauric acid 11- and 12-hydroxylase; F = female;
M = male; NC = no significant change; Palm CoA: cyanide-insensitive palmitoyl-CoA oxidation; PBOX = peroxisomal beta-oxidation; PP = peroxisomal proliferation
° (Smith et al.. 2000)
b (Kaufmann et al.. 2002)
c (Hazleton Labs. 1992)
d (Covance Labs. 1998a)
e (Valles etal.. 2003)

^(Ma et al.. 2014a)
g(ToxStrateeies. 2024)

Page 31 of 49


-------
4.3 Temporal Association of Key Events with Tumor Response

In rats, it is clear that KE1 and KE3 precede tumor formation, however, the temporal sequence of KE1
and KE3 cannot be established (Table 4-1). Biomarkers of PPARa activation (KE1) and hepatic cell
proliferation (KE3) are both increased as early as one week following oral exposure to DINP (Covance
Labs. 1998b); however, no studies are available that evaluate either KE at early timepoints.
Comparatively, liver neoplasms were first detected during an interim sacrifice on study week 79 in a
study of F344 rats by Covance Labs (1998b) (albeit without a clear dose-relationship; adenomas
detected in one control male and one high-dose female; carcinoma detected in one high-dose male).

In mice, it is clear that KE1, KE2, and KE3 precede tumor formation; however, the temporal sequence
of KE1, KE2, and KE3 cannot be established (Table 4-2). Biomarkers of PPARa activation (KE1) are
significantly increased in one study as early as 4 days after oral exposure (Hazleton Labs. 1992). while
KE2 is measured in only a single study that reports increases in TNFa and IL-1 in liver homogenate
after 14 days (Ma et al.. 2014a). and hepatic cell proliferation (KE3) is increased after 1 week of oral
exposure to DINP (Kaufmann et al.. 2002). However, no studies are available that evaluate any of these
KEs at earlier timepoints. Comparatively, in the available 2-year bioassay of mice (Covance Labs.
1998a). hepatocellular adenomas and carcinomas were first detected on study days 167 and 366,
respectively, in a single high-dose male at each timepoint (as reported by (U.S. CPSC. 2001)).

4.4	Strength, Consistency, and Specificity of Association of Tumor
Response with Key Events	

Available in vivo studies of mice and rats and in vitro studies of rat and mouse hepatocytes provide
remarkably consistent evidence that DINP can activate PPARa (KE1).There is also consistent evidence
that DINP can cause acute proliferative cellular responses in the livers of rats and mice in vivo and rat
hepatocytes in vitro (KE3). In contrast, cellular proliferation in the liver is not sustained chronically in
either species. As discussed by Corton et al. (2018). PPARa activators tend to "produce transient
increases in replicative DNA synthesis during the first few days or weeks of exposure followed by a
return to baseline levels." Chronic or sustained proliferative responses for potent PPARa activators tend
to be much lower compared to acute proliferative responses. Comparatively, DINP is a relatively weak
PPARa activator and low levels of chronic hepatic cell proliferation may be difficult to detect over
variable background levels. Therefore, lack of a detectable sustained proliferative response is consistent
with the proposed MOA for a weak PPARa activator such as DINP. Further adding to the strength of
evidence, KE1 and KE3 have been observed in studies of differing design and originating from different
laboratories with hepatic effects such as increases in relative liver weight and hepatocellular hypertrophy
observed in short-term, subchronic, and chronic studies of rats and mice. These effects, although not
KEs in the PPARa MOA, are frequently observed following PPARa activation and subsequent
peroxisome proliferation.

4.5	Biological Plausibility and Coherence	

Extensive evidence exists to support the hypothesis that chronic PPARa activation can lead to
alterations in cell growth pathways, perturbations of cell growth and survival, and selective clonal
expansion of preneoplastic foci cells leading to hepatocellular tumorigenesis in rodents (reviewed in
(Corton et al.. 2018; Corton et al.. 2014)). This proposed MOA for DINP-induced liver tumors in rats
and mice is consistent with available data, indicating biological plausibility. Available data from mice
and rats demonstrate PPARa activation after short-term (several days to weeks) oral exposure to DINP
that can be sustained with chronic exposure (Covance Labs. 1998a. b). Although studies also
demonstrate that oral exposure to DINP can cause acute hepatic cell proliferative responses, other

Page 32 of 49


-------
studies demonstrate that oral exposure to DINP does not cause chronic proliferative response in the liver
of mice or rats. As discussed by Corton et al. (2018) chronic or sustained proliferative responses for
potent PPARa activator are much lower compared to acute proliferative responses. Comparatively,

DINP is a relatively weak PPARa activator and low levels of chronic hepatic cell proliferation may be
difficult to detect over variable background levels.

4.6 Other Modes of Carcinogenic Action

This section summarizes evidence for other modes of carcinogenic action in the liver for DINP.
Ppara-NuU Mice

Valles et al. (2003) conducted a series of short-term (1- to 3-week) studies in which male and female
B6C3F1, wild-type SV129, and Ppara-mxW mice were exposed to DINP. Repeated dose studies well-
established that in response to exposure to DINP, male and female B6C3F1 wild-type show
hepatotoxicity. Across these studies, dose-dependent increases in relative liver weight that were
dependent on PPARa were generally observed; however, in one study of older (30-week) female Ppara-
mxW mice, PPARa-independent increases in relative liver weight has also been observed, (these increases
were specific for older female mice; younger female or older male Ppara-mxW mice did not exhibit any
changes in liver to body weight ratios after exposure to DINP), thereby hinting at the possibility of
PPARa-independent mechanisms being at play in the liver under certain conditions. Unique gene
expression changes in older Ppara-mxW female mice have been identified in expression arrays, like
testosterone hydroxylase (Cyp2d9). Cyp2d9 is down-regulated by DINP in wild-type mice, but Cyp2d9
was up-regulated in Ppara-mxW mice. The relevance of these subtle PPARa-independent effects to
hepatocarcinogenesis is not known, but Ppara-mxW mice are resistant to the carcinogenicity of a
prototypical PPARa activator (Peters et al.. 1997). It is important to note that most of the studies
conducted by Valles et al. support the hypothesis that PPARa plays a dominant role in mediating the
carcinogenic effects of DINP in the liver.

Other Nuclear Receptors

Constitutive androstane receptor (CAR), pregnane X receptor (PXR), and aryl hydrocarbon receptor
(AhR) are known to play a role in liver homeostasis and disease. Although their precise role, if any, in
liver tumorigenesis in response to chronic exposure to DINP has not yet been established. In addition to
PPARa, DINP has been shown to activate multiple nuclear receptors that may play a role in liver
tumorigenesis. Several studies have demonstrated that DINP can activate CAR, which is a nuclear
receptor with an adverse outcome pathway with KEs like those of PPARa and has been implicated in
hepatic carcinogenesis in rodents (Felter et al.. 2018). DeKeyser et al. (2011) used transactivation and
mammalian two-hybrid assays in COS-1 cells to demonstrate that DINP is a strong activator of human
CAR variant 2 (hCAR2). Furthermore, DINP induced expression of CYP2B6, one of the primary target
genes of CAR, in primary human hepatocytes. In a subsequent study by the same research group,
Laurenzana et al. demonstrates that MINP, a metabolite of DINP, can also activate hCAR2 (Laurenzana
et al.. 2016). Additionally, in vitro studies have also shown that DINP and MINP can activate human
PXR (Laurenzana et al.. 2016; Dekevser et al.. 2011) as well as mouse and human PPAR gamma,
although the degree of PPAR gamma activation was greater for the mouse receptor than for the human
receptor under the conditions of the study (Bilitv et al.. 2004). DINP has also been shown to promote
and induce tumorigenesis in a variety of cell types through AhR-mediated genomic and nongenomic
pathways (Wang et al.. 2012). DINP induces several changes in rodent liver consistent with PPARa
activation (Laurenzana et al.. 2016); notably, DINP induces some of these liver changes independently
of PPARa activation as shown in Ppara-mxW mice (Valles et al.. 2003).

Page 33 of 49


-------
DINP has also been evaluated in 442 high-throughput assays as part of EPA's Toxicity ForeCaster
(ToxCast) program. Curated high-throughput screening data for DINP accessed through the National
Toxicology Program (NTP) Integrated Chemical Environment (ICE) indicated that DINP was inactive in
the majority of tested assays and active in only seven assays (Table 4-3). Consistent with available
literature, DINP was active in two assays for PXR activation. However, DINP was inactive in assays for
other nuclear receptors (i.e., CAR, AhR, PPARa, and PPARy) and other assays of PXR (i.e.,
TOX21_PXR_Agonist, TOX21_PXR_viability) and these results are inconsistent with available
literature.

Table 4-3. Summary of Active ToxCast Assays for DINP

ToxCast Assay

Mode of Action

AC50fl /
LOEC (jiM)

BSK_SAg_Eselectin_up

Cancer - KCC6: Chronic Inflammation,
CardioTox - Endothelial Injury/Coagulation

0.2

BSK_CASM3C_TissueFactor_down

AcuteTox - Immune and Inflammatory Response,
CardioTox - Endothelial Injury/Coagulation

0.2

ATGPXREC ISup

Cancer - KCC8: Receptor Mediated Effects

1.2

ATG PXR TRANS up

Cancer - KCC8: Receptor Mediated Effects

1.7

BSKKF3 CTIL1 adown



4

NV SENZhBACE



8.7

ACEA ER AUC viability

AcuteTox - Cytotoxicity, Cancer - KCC10: Cell
Proliferation/Death/Energetics

38.8

AC50 = concentration at which 50% maximum activity is observed; LOEC = lowest-observable-effect

concentration; LOEC = lowest-observed-effect concentration

11 Data accessed through NTP's Integrated Chemical Environment in February 2024.

Gap Junction Intercellular Communication

Gap junctional intercellular communication (GJIC) is the only portal by which multicellular organisms
mediate the intercellular exchange of cellular signal factors from the interior of one cell to that of
neighboring cells (Loewenstein. 1987; Pitts and Finbow. 1986). GJIC is considered to play a crucial role
in the maintenance of homeostasis, and in turn, aberrant GJIC is likely to be involved in carcinogenesis,
given that cancer cells do indeed behave as if they have dysfunctional GJIC and are dissociated from the
homeostasis maintained by the organism. Inhibition of GJIC has been proposed as a non-genotoxic
carcinogenic mechanism (Yamasaki et al.. 1995; Yamasaki. 1995). Aberrant GJIC has been known as a
non-genotoxic event that is important for carcinogenesis. This is based on the observation that many
non-genotoxic, tumor-promoting agents inhibit GJIC (Klaunig et al.. 2003). Several tumor types,
including hepatocellular carcinomas, have been shown to demonstrate inhibited GJIC (Trosko et al..
1990c; Trosko et al.. 1990a. b; Trosko and Chang. 1989). DINP is shown to inhibit hepatic GJIC, and
the inhibition of GJIC has been proposed as a non-genotoxic carcinogenic mechanism in rodents
exposed to DINP for 2 or 4 weeks (Smith et al.. 2000; Trosko et al.. 1990c; Trosko et al.. 1990b).
Additionally, DINP has been shown to inhibit GJIC in vitro. Kamenduliz et al. (2002) demonstrate that
three different MINP isomers (MINP-1, MINP-2, MINP-3) can inhibit GJIC in primary hepatocytes
isolated from female B6C3F1 mice and male F344 rats treated with concentrations of each isomer
ranging from 50 to 300 |iM for 72 hours—but not in primary hepatocytes isolated from male Syrian
golden hamsters, male cynomolgus monkeys, or human donor liver tissue.

Page 34 of 49


-------
Cytotoxicity and Regenerative Proliferation

Cytotoxicity followed by regenerative proliferation is an established nongenotoxic MOA (Felter et al..
2018). There is some limited evidence that DINP may act through a cytotoxic MOA. The KEs for
establishing a cytotoxic MOA are (1) the chemical is not DNA reactive; (2) evidence of cytotoxicity by
histopathology (e.g., the presence of necrosis and/or increased apoptosis); (3) evidence of toxicity by
increased serum enzymes indicative of cellular damage that are relevant to humans; (4) presence of
increased cell proliferation as evidenced by increased labeling index and/or increased number of
hepatocytes; (5) demonstration of a parallel dose response for cytotoxicity and formation of tumors; and
(6) reversibility upon cessation of exposure (Felter et al.. 2018). As discussed in Section 2 as well as
below in the genotoxicity section, EPA considers DINP not likely to be genotoxic or mutagenic. Four
studies have provided quantitative liver histopathology with clear evidence of lesions consistent with
cytotoxicity, namely focal necrosis, including three 2-year bioassay studies in rats (Covance Labs.
1998b; Lington et al.. 1997; Bio/dynamics. 1986). one 13-week study in mice (Hazleton Labs. 1992).
and one 4-week study in mice (Hazleton Labs. 1991). In Lington et al (1997). a significant dose-related
increased incidence of focal necrosis was observed in male rats while the Bio/dynamics study (1987)
reported increased incidence of focal necrosis in males of the mid-dose group, with no clear dose-
response. In the rat study by Covance Labs (1998b). individual cell degeneration/necrosis was
significantly increased in males of the high-dose group. However, not all chronic studies reported this
lesion. The 2-year study in mice by Covance Labs (1998a) did not observe focal necrosis or apoptosis,
even with a study design that included higher doses.

As mentioned above in Section 4.1.3, DINP has been shown to elicit acute proliferative responses in
mouse hepatocytes in vivo and in vitro. Hyperplasia has not been observed in hepatic tissues, suggesting
against regenerative proliferation. Increases in periportal hepatocellular replicative DNA synthesis have
been reported in mice and rats following exposure to 12,000 ppm DINP for 2 or 4 weeks (Smith et al..
2000). consistent with increases in hepatocyte proliferation observed in two other mouse studies at doses
ranging from 150 to 8,000 ppm for 1 to 4 weeks (Valles et al.. 2003; Kaufmann et al.. 2002) or in rats up
to 855 mg/kg-day DINP for up to 104 weeks (Covance Labs. 1998b). Two in vitro studies (Shaw et al..
2002; Hasmall et al.. 1999) reported increased replicative DNA synthesis and suppressed apoptosis in rat
hepatocytes at doses of DINP ranging from 150 to 750 |iM, The available data do not consistently
support the various KEs in the MOA for cytotoxicity, suggesting other MO As are at play.

4.7 Uncertainties and Limitations	

There are several limitations and uncertainties associated with the available data set for the postulated
PPARa MOA. First, no data is available for KE2 and KE4 for rats or mice, with the exception of a
single study of mice that reported increased TNFa and IL-1 (KE2) in liver homogenate (Ma et al..
2014a). However, that study is limited in that it evaluated a single duration of exposure (14 days) and
did not distinguish between IL-1 subtypes (i.e., IL-la and IL-ip). Lack of data for KE2 and KE4 is a
data gap, which reduces EPA's confidence in the postulated PPARa MOA.

For KE3, only one in vivo study of mice (and none of rats) is available that examined apoptosis in the
liver (Kaufmann et al.. 2002). In the available study, apoptosis was significantly increased after 1 week
of exposure to DINP and was unaffected after 4 weeks. This is inconsistent with the postulated MOA, in
which suppression of apoptosis is anticipated. However, this uncertainty is somewhat addressed by the
two available in vitro studies of rat hepatocytes that report consistent, dose-related, increases in PPARa
activation (KE1), increases in replicative DNA synthesis (KE3), and suppression of apoptosis (KE3) in
hepatocytes following exposure to DINP (Shaw et al.. 2002; Hasmall et al.. 1999).

Page 35 of 49


-------
Most of the available data for KE1 and KE3 comes from in vivo studies of rats and mice; however,
available studies are of variable design and in some instances employ large dose spacing, which makes
comparisons across studies difficult. Although it is clear that KE1 and KE2 occur at lower doses and
earlier than the apical outcome—liver tumors—providing some evidence of dose-response and temporal
concordance, concordance between KEs could not be established, which reduces EPA's confidence in
the postulated PPARa MOA.

Despite remaining uncertainties, there is strong evidence to support the postulated PPARa MOA.
Available evidence indicates that DINP is not genotoxic (Section 2). Furthermore, other potential modes
of carcinogenic action, such as activation of CAR, PXR, and AhR, as well as cytotoxicity and
regenerative proliferation are also non-genotoxic threshold MO As. Finally, as discussed further below in
Section 4.8, the chronic non-cancer point of departure (POD) identified in EPA's Non-cancer Raman
Health Hazard Assessment for Diisononyl Phthalate (DINP) (U.S. EPA. 2025) will adequately account
for all chronic toxicity, including carcinogenicity and activation of PPARa (KE1), which could
potentially result from exposure to DINP.

4.8	Weight of Scientific Evidence: Cancer Classification	

Under the Guidelines for Carcinogen Risk Assessment (U.S. EPA. 2005). EPA reviewed the weight of
evidence and determined that DINP is Not Likely to be Carcinogenic to Humans at doses below levels
that do not result in PPARa activation (KE1). This classification was based on the following weight of
scientific evidence considerations:

•	DINP exposure resulted in treatment-related PPARa activation (KE1) in male mice (doses >117
mg/kg-day) (Kaufmann et al.. 2002) and female rats (doses >442 mg/kg-day (Covance Labs.
1998b).

•	DINP exposure resulted in treatment-related liver tumors (adenomas and/or carcinomas
combined) in female mice (doses >336 mg/kg-day DINP) (Covance Labs. 1998a) and female rats
(doses >672 mg/kg-day DINP) (Bio/dynamics. 1987).

•	Available MOA data for liver tumors in mice and rats support the proposed PPARa MOA.

•	Limited data are available that indicate a role for other non-genotoxic, threshold MO As,
including activation of other nuclear receptors (e.g., CAR, PXR, AhR, PPARy), inhibition of
GJIC, and cytotoxicity and regenerative proliferation.

•	There is no evidence for mutagenicity.

Further, the non-cancer chronic POD (NOAEL/LOAEL of 15/152 mg/kg-day) based on non-cancer liver
effects (see Non-cancer Human Health Hazard Assessment for Diisononyl Phthalate (DINP) (U.S. EPA.
2025)) will adequately account for all chronic toxicity, including carcinogenicity, which could
potentially result from exposure to DINP. In one study of male mice (Kaufmann et al.. 2002).
biomarkers of PPARa activation were significantly increased at 117 mg/kg-day, which is less than the
chronic LOAEL of 152 mg/kg-day based on non-cancer liver effects. Although, the study by Kaufman
et al. did not test sufficiently low doses to establish a NOAEL for PPARa activation, other studies of
mice have established a NOAEL of 75 mg/kg-day for PPARa activation (Smith et al.. 2000). Therefore,
the non-cancer chronic POD of 15 mg/kg-day is considered protective of PPARa activation.

4.9	Human Relevancy

Several panels have been convened to address the human relevancy of liver tumors in rodents occurring
through a PPARa MOA (Felter et al.. 2018; Corton et al.. 2014). These panels have generally concluded
that the PPARa MOA is not relevant to humans or unlikely to be relevant to humans based on
qualitative and quantitative differences between species. Nevertheless, uncertainty and differing

Page 36 of 49


-------
scientific opinions on the human relevance of the PPARa MOA for liver tumorigenesis remain, despite
the related efforts of previous panels and workshops.

Several authoritative agencies have evaluated the role of PPARa and peroxisome proliferation in
inducing hepatocellular tumors in rodents following chronic exposure to DINP. Australia NICNAS
(2012) and U.S. CPSC (2010) concluded that liver tumors in rodents observed following exposure to
DINP are not likely to be human-relevant, while ECHA (2013) and Health Canada (EC/HC. 2015)
concluded that liver tumors in rats are of unclear human relevance. However, none of these agencies
quantitatively evaluated DINP for carcinogenic risk to humans.

As discussed further in EPA's Non-cancer Raman Health Hazard Assessment for Diisononyl Phthalate
(DINP) (U.S. EPA 2025). not all of the non-cancer liver effects observed in rodents are consistent with
PPARa activation (e.g., spongiosis hepatis). Furthermore, the non-cancer chronic POD
(NOAEL/LOAEL of 15/152 mg/kg-day) that is based on non-cancer liver toxicity will adequately
account for all chronic toxicity, including carcinogenicity, which could potentially result from exposure
to DINP.

Page 37 of 49


-------
5 CONCLUSIONS

DINP has been evaluated for carcinogenicity in two 2-year dietary studies of F344 rats (Covance Labs.
1998b; Lington et al.. 19971 one 2-year dietary study of SD rats (Bio/dynamics. 1987). and one 2-year
dietary study of B6C3F1 mice (Covance Labs. 1998a). Across available studies, treatment-related
hepatocellular adenomas and carcinomas have consistently been observed in F344 and SD rats as well as
B6C3F1 mice. Existing assessments of DINP by U.S. CPSC (2014. 2010). Health Canada (ECCC/HC.
2020; EC/HC. 2015; Health Canada. 2015). ECHA (2013). and NICNAS (2012) have postulated that
DINP causes liver tumors in rats and mice through a PPARa MOA. Consistent with EPA Guidelines for
Carcinogen Risk Assessment (U.S. EPA. 2005) and the IPCS Mode of Action Framework (IPCS. 2007).
EPA further evaluated the postulated PPARa MOA for liver tumors, as well as evidence for other
plausible MO As for DINP.

Although some uncertainties remain, there is strong evidence to support the postulated, non-genotoxic,
PPARa MOA. Under the Guidelines for Carcinogen Risk Assessment (U.S. EPA. 2005). EPA
determined that DINP is Not Likely to be Carcinogenic to Ramans at doses below levels that do not
result in PPARa activation (KE1). Further, the non-cancer chronic POD (NOAEL/LOAEL of 15/152
mg/kg-day) based on non-cancer liver effects; see EPA's Non-cancer Raman Health Hazard Assessment
for DiisononylPhthalate (DINP) (U.S. EPA. 2025)) will adequately account for all chronic toxicity—
including carcinogenicity—which could potentially result from exposure to DINP. Therefore, the non-
cancer chronic POD of 15 mg/kg-day is considered protective of PPARa activation and carcinogenicity.

This cancer human health hazard assessment for DINP was released for public comment and was peer-
reviewed by the Science Advisory Committee on Chemicals during the July 30 to August 1, 2024,
meeting of the SACC (U.S. EPA 2024).

Page 38 of 49


-------
REFERENCES

ACC HPP. (2019). Manufacturer Request for Risk Evaluation Di-isononyl Phthalate (DINP).
(730R19001). American Chemistry Council High Phthalates Panel :: ACC HPP
https://nepis. epa.gov/exe/ZyPURL. cgi?Dockev=P 100YEGF.txt
Barber. ED; Cifone. M; Rundell. J; Przygoda. R; Astill. BP; Moran. E; Mulholland. A; Robinson. E;
Schneider. B. (2000). Results of the L5178Y mouse lymphoma assay and the Balb/3t3 cell in
vitro transformation assay for eight phthalate esters. J Appl Toxicol 20: 69-80.
http://dx.doi.org/10.1002/(SICD1099-1263(200001/02)20: l<69::AID-JAT630>3.0.CQ:2-2
BASF. (1986). Report on the study of Palatinol N (ZNT test substance N° 85/513) in the Ames test
(standard plate test with Salmonella typhimurium) performed by BASF Aktiengesellschaft
department of toxicology FRG. Project N° 40/1M0513/85.

BASF. (1995). Report on the study of diisononylphthalat IGS 21002 (ZHT test substance N° 95/91) in
the Ames test performed by BASF Aktiengesellschaft department of toxicology FRG. Project N°
40M0091/954045.

Benford. DJ: Patel. S: Reavv. HJ; Mitchell. A: Sarginson. NJ. (1986). Species differences in the
response of cultured hepatocytes to phthalate esters. Food Chem Toxicol 24: 799-800.
http://dx.doi. org/10.1016/0278-6915(86)90189-4
BIBRA. (1986). A 21-day feeding study of di-isononyl phthalate to rats: Effects on the liver and liver
lipids [TSCA Submission], (3.0495.6. 0495/6/85. OTS0509544. 40+8626208. 42005 Gl-23.
TSCATS/201730). Chemical Manufacturers Association.

https://ntrl.ntis. gov/NTRL/dashboard/searchResults.xhtml?searchQuery=OTS0509544
Bilitv. MT; Thompson. JT; McKee. RH; David. RM; Butala. JH; Vanden Heuvel. JP; Peters. JM.

(2004). Activation of mouse and human peroxisome proliferator-activated receptors (PPARs) by
phthalate monoesters. Toxicol Sci 82: 170-182. http://dx.doi.org/10.1093/toxsci/kfh253
Bio/dynamics. (1986). Chronic toxicity/oncogenicity study in F-344 rats (final report) with cover letter
dated 042386 [TSCA Submission], (EPA/OTS Doc #868600062). Houston, TX: Exxon
Chemical Americas.

https://ntrl.ntis. gov/NTRL/dashboard/searchResults.xhtml?searchQuery=OTS0510211
Bio/dynamics. (1987). A chronic toxicity carcinogenicity feeding study in rats with Santicizer 900 with
cover letter dated 06/05/87 [TSCA Submission], (EPA/OTS Doc #86870000362). St. Louis,
MO: Monsanto Company.

https://ntrl.ntis.gov/NTRL/dashboard/searchResults/titleDetail/OTSQ513172.xhtml
Brix. AE; Nvska. A: Haseman. JK; Sells. DM: Jokinen. MP: Walker. NJ. (2005). Incidences of selected
lesions in control female Harlan Sprague-Dawley rats from two-year studies performed by the
National Toxicology Program. Toxicol Pathol 33: 477-483.
http://dx.doi.org/10.1080/0192623059Q961836
Caldwell. DJ. (1999). Review of mononuclear cell leukemia in F-344 rat bioassays and its significance
to human cancer risk: A case study using alkyl phthalates [Review], Regul Toxicol Pharmacol
30: 45-53. http://dx.doi.org/10.1006/rtph.1999.1305
Caldwell. DJ: Eldridge. SR; Lington. AW: McKee. RH. (1999). Retrospective evaluation of alpha 2u-
globulin accumulation in male rat kidneys following high doses of diisononyl phthalate. Toxicol
Sci 51: 153-160. http://dx.doi.org/10.1093/toxsci/51.1.153
Corton. JC: Cunningham. ML: Hummer. BT; Lau. C: Meek. B; Peters. JM: Popp. JA; Rhomberg. L;

Seed. J: Klaunig. JE. (2014). Mode of action framework analysis for receptor-mediated toxicity:
The peroxisome proliferator-activated receptor alpha (PPARa) as a case study [Review], Crit
Rev Toxicol 44: 1-49. http://dx.doi.org/10.3109/10408444.2Q13.835784
Corton. JC: Peters. JM: Klaunig. JE. (2018). The PPARa-dependent rodent liver tumor response is not
relevant to humans: addressing misconceptions [Review], Arch Toxicol 92: 83-119.
http://dx.doi.org/10.1007/s00204-017-2Q94-7

Page 39 of 49


-------
Covance Labs. (1998a). Support: oncogenicity study in mice with di(isononyl)phthalate including
ancillary hepatocellular proliferation and biochemical analyses with cover letter dated
11/18/1998 [2598-105] [TSCA Submission], (2598-105). Philadelphia, PA: Aristech Chem
Corp. https://ntrl.ntis.gov/NTRL/dashboard/searchResults/titleDetail/OTS05562833.xhtml
Covance Labs. (1998b). Support: Oncogenicity study in rats with di(isononyl) phthalate including

ancillary hepatocellular proliferation & biochemical analyses with cover [TSCA Submission],
(EPA/OTS Doc #89980000308). Philadelphia, PA: Aristech Chemical Corp.
https://ntrl.ntis.gov/NTRL/dashboard/searchResults/titleDetail/OTSQ5562832.xhtml
Dekevser. JG: Laurenzana. EM: Peterson. EC: Chen. T; Omiecinski. CJ. (2011). Selective phthalate

activation of naturally occurring human constitutive androstane receptor splice variants and the
pregnane X receptor. Toxicol Sci 120: 381-391. http://dx.doi.org/10.1093/toxsci/kfq394
EC/HC. (2015). State of the science report: Phthalate substance grouping 1,2-Benzenedicarboxylic acid,
diisononyl ester; 1,2-Benzenedicarboxylic acid, di-C8-10-branched alkyl esters, C9-rich
(Diisononyl Phthalate; DINP). Chemical Abstracts Service Registry Numbers: 28553-12-0 and
68515-48-0. Gatineau, Quebec, https://www.ec.gc.ca/ese-
ees/default. asp?lang=En&n=47F 5 8 A A5 -1
ECB. (2003). European union risk assessment report: DINP. European Commission.

http://publications.irc.ec.europa.eu/repository/handle/JRC25827
ECCC/HC. (2020). Screening assessment - Phthalate substance grouping. (Enl4-393/2019E-PDF).
Environment and Climate Change Canada, Health Canada.

https://www.canada.ca/en/environment-climate-change/services/evaluating-existing-
substances/screening-assessment-phthalate-substance-grouping.html
ECHA. (2013). Evaluation of new scientific evidence concerning DINP and DIDP in relation to entry 52
of Annex XVII to REACH Regulation (EC) No 1907/2006. Helsinki, Finland.
http://echa.europa.eu/documents/10162/31b4067e-de40-4044-93e8-9c9ffl960715
EFSA. (2005). Opinion of the scientific panel on food additives, flavourings, processing aids and
materials in contact with food (AFC) on a request from the commission related to di-
isononylphthalate (DINP) for use in food contact materials. Question N° EFSA-q-2003-194 (pp.
1-18). http://dx.doi.Org/10.2903/i.efsa.2005.244
EG&G Mason Research Institute. (1982a). Evaluation of test article R-1218 (MRI #549) for mutagenic
potential employing the L5178Y TK+/- mutagenesis assay [TSCA Submission], (EPA/OTS Doc
#878210230). Piscataway, NJ: Nuodex, Inc.

https://ntrl.ntis. gov/NTRL/dashboard/searchResults.xhtml?searchOuerv=OTS020626Q
EG&G Mason Research Institute. (1982b). Salmonella/mammalian-microsome plate incorporation

mutagenesis assay with index (pp. 286-320). (EPA/OTS Doc #878210228). EG and G MASON
RES INST. https://ntrl.ntis.gov/NTRL/dashboard/searchResults/titleDetail/OTS020626Q.xhtml
EPL. (1999). Histopathology Peer Review and Pathology Working Group. Review of selected lesions of
the liver and spleen in male and female F344 rats exposed to di(isononyl)phthalate. (WIL-
354002). Research Triangle Park, NC: Experimental Pathology Laboratories Inc.

Felter. SP; Foreman. JE; Boobis. A; Corton. JC; Doi. AM; Flowers. L; Goodman. J; Haber. LT; Jacobs.
A; Klaunig. JE; Lynch. AM; Moggs. J; Pandiri. A. (2018). Human relevance of rodent liver
tumors: Key insights from a toxicology forum workshop on nongenotoxic modes of action
[Review], Regul Toxicol Pharmacol 92: 1-7. http://dx.doi.Org/10.1016/i.yrtph.2017.l 1.003
Fushiwaki. Y; Niino. T; Ishibashi. T; Takeda. K; Onodera. S. (2003). Tumor-promoting activity of
phthalate esters estimated by in vitro transformation using Bhas cells. J Health Sci 49: 82-87.
http://dx.doi.org/10.1248/ihs.49.82
Gutierrez-Garcia. AK; Choudhury. M; De Leon-Rodriguez. A. (2019). Diisononyl phthalate

differentially affects sirtuin expression in the HepG2 cell line. Chem Res Toxicol 32: 1863-1870.
http://dx.doi.org/10.1021/acs.chemrestox.9b002Q6

Page 40 of 49


-------
Hall. M; Matthews. A; Weblev. L; Harling. R. (1999). Effects of di-isononyl phthalate (DINP) on

peroxisomal markers in the marmoset - DINP is not a peroxisome proliferator. J Toxicol Sci 24:
237-244. http://dx.doi.Org/10.2131/its.24.3 237
Hasmall. SC: James. NH; Macdonald. N: West. D; Chevalier. S: Cosulich. SC: Roberts. RA. (1999).
Suppression of apoptosis and induction of DNA synthesis in vitro by the phthalate plasticizers
monoethylhexylphthalate (MEHP) and diisononylphthalate (DINP): A comparison of rat and
human hepatocytes in vitro. Arch Toxicol 73: 451-456.
http://dx.doi.org/10.1007/s00204005Q634
Hazleton Labs. (1991). A subchronic (4-week) dietary oral toxicity study of di(isononyl)phthalate in
B6C3F1 mice (final report) with cover sheet dated 052991 [TSCA Submission], (HWA Study
No. 2598-100. OTS0529425. 86-910000793. TSCATS/415527). Vienna, VA: Aristech Chemical
Corporation.

https://ntrl.ntis. gov/NTRL/dashboard/searchResults.xhtml?searchQuery=OTS0529425
Hazleton Labs. (1992). A 13-week subchronic dietary oral toxicity study in mice with

di(isononyl)phthalate including ancillary hepatocellular proliferation and biochemical analyses.
(HWA 2598-103). Aristech Chemical Corporation.

Health Canada. (2015). Supporting documentation: Carcinogenicity of phthalates - mode of action and
human relevance. In Supporting documentation for Phthalate Substance Grouping. Ottawa, ON.
Health Canada. (2018). Supporting documentation: Evaluation of epidemiologic studies on phthalate
compounds and their metabolites for effects on behaviour and neurodevelopment, allergies,
cardiovascular function, oxidative stress, breast cancer, obesity, and metabolic disorders. Ottawa,
ON.

IARC. (1995). Peroxisome proliferation and its role in carcinogenesis: Views and expert opinions of an
IARC Working Group: Lyon, 7-11 December 1994 [IARC Monograph], (IARC Technical
Publication No. 24). Lyon, France.

http://monographs.iarc.fr/ENG/Publications/techrep24/IARCrep24.pdf
IPCS. (2007). Harmonization project document no. 4: Part 1: IPCS framework for analysing the

relevance of a cancer mode of action for humans and case-studies: Part 2: IPCS framework for
analysing the relevance of a non-cancer mode of action for humans. Geneva, Switzerland: World
Health Organization.

http://www.who.int/ipcs/methods/harmonization/areas/cancer mode.pdf?ua=l
Kamendulis. LM; Isenberg. JS: Smith. JH; Pugh. G. Jr.: Lington, AW; Klaunig, JE. (2002). Comparative
effects of phthalate monoesters on gap junctional intercellular communication and peroxisome
proliferation in rodent and primate hepatocytes. J Toxicol Environ Health A 65: 569-588.
http://dx.doi.org/10.1080/152873902317349736
Kaufmann. W: Deckardt. K; McKee. RH; Butala. JH: Bahnemann. R. (2002). Tumor induction in mouse
liver: Di-isononyl phthalate acts via peroxisome proliferation. Regul Toxicol Pharmacol 36: 175-
183. http://dx.doi.org/10.1006/rtph.2002.1575
King-Herbert. A: Thayer. K. (2006). NTP workshop: Animal models for the NTP rodent cancer
bioassay: Stocks and strains - Should we switch? Toxicol Pathol 34: 802-805.
http://dx.doi.org/10.1080/0192623060Q935938
King-Herbert. AP; Sills. RC: Bucher. JR. (2010). Commentary: update on animal models for NTP

studies. Toxicol Pathol 38: 180-181. http://dx.doi.org/10.1177/0192623309356450
Klaunig. JE: Babich. MA: Baetcke. KP; Cook. JC: Corton. JC: David. RM; Deluca. JG: Lai. DY;

McKee. RH: Peters. JM; Roberts. RA: Fenner-Crisp. PA. (2003). PPARalpha agonist-induced
rodent tumors: modes of action and human relevance [Review], Crit Rev Toxicol 33: 655-780.
http://dx.doi.org/10.1080/713608372

Page 41 of 49


-------
Laurenzana. EM; Coslo. DM; Vigilar. MY; Roman. AM; Omiecinski. CJ. (2016). Activation of the
Constitutive Androstane Receptor by Monophthalates. Chem Res Toxicol 29: 1651-1661.
http: //dx. doi. or g/10.1021 /acs. chemrestox. 6b 0018 6
Liang. F; Yan. B. (2020). Oxidative damage in the liver and kidney induced by dermal exposure to
diisononyl phthalate in Balb/c mice. Toxicol Ind Health 36: 30-40.
http://dx.doi.org/10.1177/0748233719900861
Lington. AW; Bird. MG; Plutnick. RT; Stubblefield. WA; Scala. RA. (1997). Chronic toxicity and
carcinogenic evaluation of diisononyl phthalate in rats. Fundam Appl Toxicol 36: 79-89.
http://dx.doi.Org/10.1093/toxsci/36.l.79
Litton Bionetics. (1981). Evaluation of R-1271 in the in vitro transformation of BALB/3T3 cells assay.
Final report. (EPA/OTS Doc #878210237). Dover, DE: Nuodex Inc.
https://ntrl.ntis. gov/NTRL/dashboard/searchResults.xhtml?searchQuery=OTS0206260
Litton Bionetics. (1982a). Evaluation of R-1218 in the in vitro transformation of BALB/3T3 cells assay
final report [TSCA Submission], (EPA/OTS Doc #878210233). Piscataway, NJ: Nuodex, Inc.
https://ntrl.ntis. gov/NTRL/dashboard/searchResults.xhtml?searchOuerv=OTS020626Q
Litton Bionetics. (1982b). Evaluation of R-1218 in the primary rat hepatocyte unscheduled DNA

synthesis assay final report [TSCA Submission], (EPA/OTS Doc #878210229). Piscataway, NJ:
Nuodex, Inc.

https://ntrl.ntis. gov/NTRL/dashboard/searchResults.xhtml?searchOuerv=OTS020626Q
Litton Bionetics. (1985). Evaluation of 1J di isononyl phthalate in the in vitro transformation of

BALB/3T3 cells assay final report. (EPA/OTS Doc #40-8526194). Washington, DC: Chemical
Manufacturers Association.

https://ntrl.ntis.gov/NTRL/dashboard/searchResults/titleDetail/OTS05085Q4.xhtml
Loewenstein. WR. (1987). The cell-to-cell channel of gap junctions. Cell 48: 725-726.

http://dx.doi.org/10.1016/0092-8674(87)90067-5
Ma. P; Yan. B; Xu. Y. (2014a). Dl-ISO-nonyl phthalate oral exposure of kunming mice induces hepatic
and renal tissue injury. 938-945.

https://heronet.epa.gov/heronet/index.cfm/reference/download/reference id/7978853
Ma. P; Yan. B; Zeng. O; Liu. X; Wu. Y; Jiao. M; Liu. C; Wu. J; Yang. X. (2014b). Oral exposure of
Kunming mice to diisononyl phthalate induces hepatic and renal tissue injury through the
accumulation of ROS. Protective effect of melatonin. Food Chem Toxicol 68: 247-256.
http://dx.doi.Org/10.1016/i.fct.2014.03.027
Maronpot. RR; Nvska. A; Foreman. JE; Ramot. Y. (2016). The legacy of the F344 rat as a cancer

bioassay model (a retrospective summary of three common F344 rat neoplasms) [Review], Crit
Rev Toxicol 46: 641-675. http://dx.doi.org/10.1080/10408444.2Q16.1174669
McKee. RH; Przygoda. RT; Chirdon. MA; Engelhardt. G; Stanley. M. (2000). Di(isononyl) phthalate
(DINP) and di(isodecyl) phthalate (DIDP) are not mutagenic. J Appl Toxicol 20: 491-497.
http://dx.doi.org/10.1002/1099-1263(200011/12)20:6<491::AID-JAT724>3.0.CO:2-H
Microbiological Associates. (1981). Activity of T1646 in the in vitro mammalian cell transformation
assay in the absence of exogenous metabolic activation. (EPA/OTS Doc #878210232). Dover,
DE: Nuodex Inc.

https://ntrl.ntis. gov/NTRL/dashboard/searchResults.xhtml?searchOuerv=OTS020626Q
Microbiological Associates. (1982a). Activity of T1646 in the in vitro mammalian cell transformation
assay in the presence of exogenous metabolic activation [TSCA Submission], (EPA/OTS Doc
#878210236). Piscataway, NJ: Nuodex, Inc.

https://ntrl.ntis. gov/NTRL/dashboard/searchResults.xhtml?searchOuerv=OTS020626Q
Microbiological Associates. (1982b). Activity of T1646 in the in vivo cytogenetics assay in rodents.
(EPA/OTS; Doc #878210231). Nuodex Inc.

Page 42 of 49


-------
Microbiological Associates. (1982c). Activity of T1674 in the in vitro mammalian cell transformation
assay in the absence of exogenous metabolic activation [TSCA Submission], (EPA/OTS Doc
#878210234). Piscataway, NJ: Nuodex, Inc.

https://ntrl.ntis. gov/NTRL/dashboard/searchResults.xhtml?searchOuerv=OTS0206260
Microbiological Associates. (1982d). Activity of T1677 in the in vitro mammalian cell transformation
assay in the absence of exogenous metabolic activation [TSCA Submission], (EPA/OTS Doc
#878210235). Piscataway, NJ: Nuodex, Inc.

https://ntrl.ntis. gov/NTRL/dashboard/searchResults.xhtml?searchQuery=OTS0206260
NICNAS. (2012). Priority existing chemical assessment report no. 35: Diisononyl phthalate. (PEC35).
Sydney, Australia: Australian Government Department of Health and Ageing.
https://www.industrialchemicals.gov.au/sites/default/files/PEC35-Piisononyl-phthalate-
DINP.pdf

OEHHA. (2013). Chemical listed effective December 20, 2013 as known to the state of California to
cause cancer: Diisononyl phthalate (DINP) [Website], https://oehha.ca.gov/proposition-
65/crnr/chemical-listed-effective-december-20-2013-known-state-california-cause-cancer
Parada. H; Gammon. MP: Chen. J: Calafat AM: Neugut AI; Santella. RM; Wolff MS: Teitelbaum.
SL. (2018). Urinary Phthalate Metabolite Concentrations and Breast Cancer Incidence and
Survival following Breast Cancer: The Long Island Breast Cancer Study Project. Environ Health
Perspect 126: 047013. http://dx.doi.org/10.1289/EHP2083
Peters. JM; Taubeneck. MW: Keen. CL; Gonzalez. FJ. (1997). Di(2-ethylhexyl) phthalate induces a

functional zinc deficiency during pregnancy and teratogenesis that is independent of peroxisome
proliferator-activated receptor-alpha. Teratology 56: 311-316.
http://dx.doi.org/10.1002/(sici) 1096-9926(199711)56:5<31 l::aid-tera4>3.0.co:2-#

Pitts. JD; Finbow. ME. (1986). The gap junction [Review], 4: 239-266.

http://dx.doi.org/10.1242/ics,1986.supplement 4.15
Pugh. G: Isenberg. J: Kamendulis. L; Acklev. D; Clare. L; Brown. R; Lington. A: Smith. J: Klaunig. J.
(2000). Effects of di-isononyl phthalate, di-2-ethylhexyl phthalate, and clofibrate in cynomolgus
monkeys. Toxicol Sci 56: 181-188. http://dx.doi.Org/10.1093/toxsci/56.l.181
Reeves. KW: Santana. MP: Manson. JE; Hankinson. SE; Zoeller. RT; Bigelow. C: Sturgeon. SR;

Spiegelman. D; Tinker. L; Luo. J: Chen. B; Meliker. J: Bonner. MR: Cote. ML: Cheng. TP:
Calafat. AM. (2019). Urinary phthalate biomarker concentrations and postmenopausal breast
cancer risk. J Natl Cancer Inst 111: 1059-1067. http://dx.doi.org/10.1093/inci/diz002
Reynolds. CW: Foon. KA. (1984). T gamma-lymphoproliferative disease and related disorders in
humans and experimental animals: A review of the clinical, cellular, and functional
characteristics [Review], Blood 64: 1146-1158.
http://dx.doi.Org/10.l 182/blood.V64.6.1146.1146
Shaw. P; Lee. R; Roberts. RA. (2002). Species differences in response to the phthalate plasticizer
monoisononylphthalate (MINP) in vitro: A comparison of rat and human hepatocytes. Arch
Toxicol 76: 344-350. http://dx.doi.org/10.1007/s00204-002-Q342-x
Smith. JH; Isenberg. JS: Pugh. G. Jr.: Kamendulis. LM; Acklev. P; Lington. AW: Klaunig. JE. (2000).
Comparative in vivo hepatic effects of Pi-isononyl phthalate (PINP) and related C7-C11 dialkyl
phthalates on gap junctional intercellular communication (GJIC), peroxisomal beta-oxidation
(PBOX), and PNA synthesis in rat and mouse liver. Toxicol Sci 54: 312-321.
http://dx.doi.Org/10.1093/toxsci/54.2.312
Thomas. J: Haseman. JK; Goodman. JI; Ward. JM: Loughran. TP. Jr.: Spencer, PJ. (2007). A review of
large granular lymphocytic leukemia in Fischer 344 rats as an initial step toward evaluating the
Implication of the endpoint to human cancer risk assessment [Review], Toxicol Sci 99: 3-19.
http ://dx. doi. org/10.1093/toxsci/kfm098

Page 43 of 49


-------
Toman RS; Budroe. JD; Cendak. R. (2013). Evidence of the carcinogenicity of diisononyl phthalate
(DINP). Sacramento, CA: California Environmental Protection Agency.
https://oehha.ca.gov/media/downloads/proposition-65/chemicals/dinphidl00413.pdf
ToxStrategies. (2024). Analysis of microarray data from wild-type mice exposed to diisononyl phthalate

(DINP). Annandale, NJ: ExxonMobil Biomedical Sciences, Inc.

Trosko. JE; Chang. CC. (1989). Stem cell theory of carcinogenesis [Review], Toxicol Lett 49: 283-295.

http://dx.doi. org/10.1016/03 78-4274(89)9003 8-6
Trosko. JE: Chang. CC: Madhukar. BY. (1990a). In vitro analysis of modulators of intercellular
communication: implications for biologically based risk assessment models for chemical
exposure. Toxicol In Vitro 4: 635-643. http://dx.doi.org/10.1016/0887-2333(90)90133-e
Trosko. JE: Chang. CC: Madhukar. BY. (1990b). Modulation of intercellular communication during

radiation and chemical carcinogenesis [Review], RadiatRes 123: 241-251.

Trosko. JE: Chang. CC: Madhukar. BY: Klaunig. JE. (1990c). Chemical, oncogene and growth factor
inhibition gap junctional intercellular communication: An integrative hypothesis of
carcinogenesis [Review], Pathobiology 58: 265-278. http://dx.doi.Org/10.l 159/000163596
U.S. CPSC. (2001). Report to the U.S. Consumer Product Safety Commission by the Chronic Hazard

Advisory Panel on diisononyl phthalate (DINP).

U.S. CPSC. (2010). Toxicity review of Diisononyl Phthalate (DINP). Bethesda, MD.

http://www.cpsc.gov/PageFiles/126539/toxicitvDINP.pdf
U.S. CPSC. (2014). Chronic Hazard Advisory Panel on Phthalates and Phthalate Alternatives (with

appendices). Bethesda, MD: U.S. Consumer Product Safety Commission, Directorate for Health
Sciences. https://www.cpsc.gov/s3fs-public/CHAP-REPORT-With-Appendices.pdf
U.S. EPA. (1991). Alpha-2u-globulin: Association with chemically induced renal toxicity and neoplasia
in the male rat [EPA Report], (EPA625391019F. PB92143668). Washington, DC: U.S.
Environmental Protection Agency, National Center for Environmental Assessment.
https://ntrl.ntis. gov/NTRL/dashboard/ searchResults.xhtml?searchOuery=PB92143668
U.S. EPA. (2005). Guidelines for carcinogen risk assessment [EPA Report], (EPA630P03001F).
Washington, DC. https://www.epa.gov/sites/production/files/2013-
09/documents/cancer guidelines final 3-25-05.pdf
U.S. EPA. (2020). Draft scope of the risk evaluation for di-isononyl phthalate (CASRNs 28553-12-0 and
68515-48-0) [EPA Report], (EPA Document No. EPA-740-D-20-033). Research Triangle Park,
NC: Office of Pollution Prevention and Toxics; U.S. Environmental Protection Agency.
U.S. EPA. (2021). Final scope of the risk evaluation for di-isononyl phthalate (DINP) (1,2-benzene-
dicarboxylic acid, 1,2-diisononyl ester, and 1,2-benzenedicarboxylic acid, di-C8-10-branched
alkyl esters, C9-rich); CASRNs 28553-12-0 and 68515-48-0 [EPA Report], (EPA-740-R-21-
002). Washington, DC: Office of Chemical Safety and Pollution Prevention.
https://www.epa.gov/svstem/files/documents/2021-08/casrn-28553-12-0-di-isononvl-phthalate-
final-scope.pdf

U.S. EPA. (2024). Science Advisory Committee on Chemicals Meeting Minutes and Final Report No.
2024-2, Docket ID: EPA-HQ-OPPT-2024-0073: For the Draft Risk Evaluation for Di-isodecyl
Phthalate (DIDP) and Draft Hazard Assessments for Di-isononyl Phthalate (DINP). Washington,
DC: U.S. Environmental Protection Agency, Science Advisory Committee on Chemicals.
U.S. EPA. (2025). Non-cancer Human Health Hazard Assessment for Diisononyl Phthalate (DINP)

Washington, DC: Office of Pollution Prevention and Toxics.

Valles. EG; Laughter. AR; Dunn. CS: Cannelle. S: Swanson. CL; Cattlev. RC: Corton. JC. (2003). Role
of the peroxisome proliferator-activated receptor alpha in responses to diisononyl phthalate.
Toxicology 191: 211-225. http://dx.doi.org/10.1016/S0300-483X(03)00260-9

Page 44 of 49


-------
Wang. YC; Chen. HS; Long. CY; Tsai. CF; Ftsieh. TH; Hsu. CY; Tsai. EM. (2012). Possible mechanism
of phthalates-induced tumorigenesis [Review], Kaohsiung J Med Sci 28: S22-S27.
http://dx.doi.Org/10.1016/i.kims.2012.05.006
Yamasaki. H. (1995). Non-genotoxic mechanisms of carcinogenesis: studies of cell transformation and
gap junctional intercellular communication [Review], Toxicol Lett 77: 55-61.
http://dx.doi.org/10.1016/0378-4274(95K)3272-x
Yamasaki. H; Mesnil. M; Omori. Y; Mironov. N: Krutovskikh. V. (1995). Intercellular communication
and carcinogenesis [Review], Mutat Res Fundam Mol Mech Mutagen 333: 181-188.
http://dx.doi. org/10.1016/0027-5107(95)00144-1
Zeiger. E; Haworth. S: Mortelmans. K; Speck. W. (1985). Mutagenicity testing of di(2-

ethylhexyl)phthalate and related chemicals in Salmonella. Environ Mol Mutagen 7: 213-232.
http://dx.doi.org/10.1002/em.28600702Q9

Page 45 of 49


-------
APPENDICES

Appendix A PATHOLOGY WORKING GROUP REVIEW FOR
	SPONGIOSIS HEPATIS AND MNCL (EPL. 1999)

A Histopathology Peer Review and a Pathology Working Group (PWG) review (EPL. 1999) was
conducted on selected lesions of the liver and spleen observed in F344 rats in the 2-year bioassays
reported by Lington et al. (1997) and Covance Labs (1998b). The PWG review evaluated the
significance of spongiosis hepatis, foci of cellular alteration, primary hepatocellular neoplasms in the
liver, and the significance of MNCL. The peer and PWG reviews were conducted in accordance with
EPA Pesticide Regulation Notice 94-5, which describes the procedure to be followed for submission of
pathology re-reads to the Agency (EPL. 1999).

Spongiosis Hepatis

Induction of spongiosis hepatis, also referred to as cystic degeneration by some authors, is of interest
because it appears to be the most sensitive non-neoplastic response in rats chronically exposed to DINP
(Covance Labs. 1998b; Lington et al.. 1997). However, questions have arisen regarding the relationship
of this lesion to other pathological processes occurring in animals treated with DINP that may not be
relevant to humans, including peroxisome proliferation and MNCL. Although a few differences were
noted, the Histology Peer Review and the PWG review of lesions in the liver and spleen generally
confirm the incidence data reported by the original study pathologists. The incidences of spongiosis
hepatis in the Lington et al. (1997) and Covance Labs (1998b) studies as determined by the PWG are
shown in TableApx A-l and TableApx A-2.

The PWG noted that spongiosis hepatis might be found as an independent lesion or within foci of
cellular alteration or hepatocellular neoplasms. In the reviewed studies, spongiosis hepatis was
diagnosed whenever it occurred, regardless of relationship to other hepatic changes that were also
present. This method of diagnosis differs from some standard pathology guidelines, which recommend
that spongiosis hepatis not be diagnosed separately when it occurs within foci or tumors. The PWG
concluded that the method of diagnosis used in the DINP rat studies made interpretation of spongiosis
hepatis as a treatment-related effect difficult. As noted in EPL (1999). some differences were noted in
the pathology protocols for the two studies which may have affected the reported incidences. These
differences include the number of sections taken from the liver in each study and the protocol for
examination of the spleen. These differences make the direct comparison of the results from Lington et
al. (1997) and Covance Labs (1998b) difficult and may account for the greater incidence of foci of
cellular alteration and foci of spongiosis hepatis observed by Lington et al. (1997).

Page 46 of 49


-------
TableApx A-l. Incidence of MNCL and Selected Hepatic Lesions at Terminal Sacrifice (104
Weeks) in the Lington et al. (1997) Study in F344 Rats as Determined by the PWG (EPL, 1999)

Lesion

Dose Group mg/kg-day (ppm)

Control

15M/18F
(300)

152 M/184 F
(3,000)

307 M / 375 F
(6,000)

Males

MNCL

32/81

27/80

48/80

49/80

Hepatocellular adenoma

3/81

1/80

2/80

1/80

Hepatocellular carcinoma

0/81

1/80

0/80

3/80

Eosinophilic foci

58/81

50/80

46/80

52/80

Basophilic foci

53/81

62/80

48/80

42/80

Spongiosis hepatis

22/81

24/80

51/80

62/80

Females

MNCL

22/81

21/81

29/80

41/80

Hepatocellular adenoma

0/81

4/81

0/80

2/80

Hepatocellular carcinoma

1/81

0/81

0/80

1/80

Eosinophilic foci

59/81

47/81

42/80

32/80

Basophilic foci

72/81

64/81

64/80

55/80

Spongiosis hepatis

4/81

1/81

3/80

4/80

Source: Modified from data in Table 6 in EPL (1999)
M = male; F = female

Page 47 of 49


-------
TableApx A-2. Incidence of MNCL and Selected Hepatic Lesions at Terminal Sacrifice (104
Weeks) in the Covance Labs (1998b) Study in F344 Rats as Determined by the PWG (EPL, 1999)



Dose Group mg/kg-day (ppm)

Lesion

Control

29 M / 36 F

88 M/109

359 M / 442 F

733 M / 885 F

Recovery
637 M / 773 F
(12,000)



(500)

F (1,500)

(6,000)

(12,000)

Males

MNCL

21/55

23/50

21/50

32/55

28/55

30/50

Hepatocellular
adenoma

2/55

4/50

1/50

4/55

7/55

6/50

Hepatocellular

1/55

0/50

0/50

3/55

11/55

3/50

carcinoma













Eosinophilic foci

22/55

14/50

16/50

15/55

10/55

12/50

Basophilic foci

40/55

34/50

33/50

28/55

27/55

25/50

Spongiosis hepatis

6/55

6/50

3/50

18/55

26/55

10/50

Females

MNCL

17/55

16/50

9/50

28/55

28/55

24/50

Hepatocellular
adenoma

1/55

1/50

0/50

1/55

1/55

1/50

Hepatocellular

0/55

0/50

0/50

1/55

6/55

2/50

carcinoma













Eosinophilic foci

10/55

5/50

7/50

7/55

0/55

4/50

Basophilic foci

37/55

32/50

31/50

18/55

5/55

13/50

Spongiosis hepatis

0/55

0/50

0/50

1/55

2/55

0/50

Source: Modified from data in Tables 9 and 10 in EPL (1999)
M = male; F = female

Examination of Co-occurrence of MNCL and Spongiosis Hepatis

It has been suggested that the occurrence of spongiosis hepatis in rats exposed to DINP is a consequence
of MNCL (EPL. 1999). To address this possibility, the PWG examined the co-occurrence of spongiosis
hepatis and MNCL in the study by Lington et al. (1997) and Covance Labs (1998b). A comparison of
the numbers of animals with spongiosis hepatis with and without MNCL diagnosed by the study
pathologist did not support the conclusion that spongiosis hepatis is a consequence of MNCL as shown
in Table Apx A-3. Although approximately half of the rats with spongiosis hepatis also had MNCL,
spongiosis hepatis was also observed in the absence of MNCL in the remainder of the affected animals.

Page 48 of 49


-------
TableApx A-3. Comparison of Spongiosis Hepatis with MNCL as Determined by the PWG (EPL,

1999)

Sex

Dose Group
(ppm)

Total with Spongiosis
Hepatis

Spongiosis Hepatis
without MNCL

Spongiosis Hepatis
with MNCL

Comparison of data from Lington et al. (1997)

F

0

4

1

3

F

300

1

1

0

F

3,000

3

0

3

F

6,000

4

1

3

M

0

24

16

8

M

300

24

12

12

M

3,000

54

17

37

M

6,000

66

27

39

Com

parison of data from Covance Labs (1998b)

F

0

0

0

0

F

500

0

0

0

F

1,500

0

0

0

F

6,000

1

0

1

F

12,000

2

0

2

F

12,000 recovery

0

0

0

M

0

5

1

4

M

500

5

4

1

M

1,500

2

1

1

M

6,000

14

8

6

M

12,000

21

11

10

M

12,000 recovery

9

5

4

Source: Modified from data in Tables 11 and 12 in EPL (1999)
M = male; F = female

Page 49 of 49


-------