Registration Dossier

Data platform availability banner - registered substances factsheets

Please be aware that this old REACH registration data factsheet is no longer maintained; it remains frozen as of 19th May 2023.

The new ECHA CHEM database has been released by ECHA, and it now contains all REACH registration data. There are more details on the transition of ECHA's published data to ECHA CHEM here.

Diss Factsheets

Toxicological information

Carcinogenicity

Currently viewing:

Administrative data

Description of key information

Dietary carcinogenicity studies in rats and mice including the recently published CLARITY Core study concluded that Bisphenol A was not carcinogenic in either species. No inhalation or dermal carcinogenicity studies were available, although in repeated exposure inhalation toxicity studies, Bisphenol A did not exhibit properties that raised concern for potential carcinogenicity. Overall, Bisphenol A has no carcinogenic potential. For Carcinogenicity via oral Route the NOAEL was taken from the CLARITY Core study as it is the largest-ever scientific investigation of Bisphenol A 

Key value for chemical safety assessment

Carcinogenicity: via oral route

Link to relevant study records

Referenceopen allclose all

Endpoint:
carcinogenicity: oral
Type of information:
experimental study
Adequacy of study:
key study
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
other: Acceptable, well-documented study report which meets basic scientific principles.
Qualifier:
no guideline followed
Principles of method if other than guideline:
50 rats and 50 mice per dose group consumed control diets or test diets containing either 1000 or 2000 ppm Bisphenol A (rats), 1000 or 5000 ppm Bisphenol A (male mice), or 5000 or 10000 ppm Bisphenol A (female mice) ad libitum for 103 weeks. Rats and mice were observed for five weeks or four weeks, respectively, after treatment ended. Clinical signs were recorded and all animals were necropsied and examined for neoplastic and non-neoplastic lesions on major tissues and organs at the end of the study.
GLP compliance:
no
Species:
other: rat and mouse
Strain:
other: Fischer 344 rats and B6C3F1 mice
Sex:
male/female
Details on test animals or test system and environmental conditions:
TEST ANIMALS
- Age at study initiation: Five weeks
- Housing: Animals were housed five per cage in polycarbonate cages covered with non-woven polyester filter sheets. Racks and filters were changed every two weeks. Bedding was Absorb-dri hardwood chips. Cages, bedding, and water bottles were changed twice per week. Food hoppers were changed once per week. Animals were housed in a room in which feeding studies on caprolactam (CAS 105-60-2), 11-aminoundecanoic acid (CAS 2432-99-7), and 2,6-dichloro-p-phenylenediamine (CAS 609-20-1) were also being conducted.
- Diet: Ad libitum. Purina Laboratory Chow (Ralston Purina Co., Richmond, Indiana, United States) in stainless steel food hoppers.
- Water: Ad libitum. Tap water acidified with hydrochloric acid to pH 2.5 in glass water bottles equipped with stainless steel sipper tubes.
- Acclimation period: Two weeks

ENVIRONMENTAL CONDITIONS
- Temperature: 22-26 degrees C
- Humidity: 30-70%
- Photoperiod: 12 hours light/12 hours dark
Route of administration:
oral: feed
Vehicle:
unchanged (no vehicle)
Details on exposure:
Test diets were formulated by mixing a small amount of Purina Lab Chow and the required amount of BPA with a mortar and pestle and then adding this premix to the required amount of animal meal and mixing in a blender for 20 minutes. Test diets were stored in the dark at 4 degrees C for no longer than two weeks. Control diets consisted of Purina Lab Chow.

Food consumption measurements were recorded every four weeks for the rats, but could not be precisely evaluated for the mice because of excessive spilling of feed.
Analytical verification of doses or concentrations:
yes
Details on analytical verification of doses or concentrations:
Analyses of methanol-extracted samples of test diets were performed by high-pressure liquid chromatography. The mean analytical concentrations of BPA in selected samples from test diets were usually within 10% of the theoretical level.
Duration of treatment / exposure:
103 weeks
Frequency of treatment:
Animals ate Bisphenol A-containing test diets ad libitum
Post exposure period:
Rats were observed for five weeks after exposure. Mice were observed for four weeks after exposure.
Remarks:
Doses / Concentrations:
0, 1000, and 2000 ppm for rats; 0, 1000, and 5000 ppm for male mice; 0, 5000, and 10000 ppm for female mice
Basis:
nominal in diet
No. of animals per sex per dose:
50
Control animals:
yes, plain diet
Details on study design:
50 rats and 50 mice per dose group consumed control diets or test diets containing either 1000 or 2000 ppm Bisphenol A (rats), 1000 or 5000 ppm Bisphenol A (male mice), or 5000 or 10000 ppm Bisphenol A (female mice) ad libitum for 103 weeks. Rats and mice were observed for five weeks or four weeks, respectively, after treatment ended. Clinical signs were recorded and all animals were necropsied and examined for neoplastic and non-neoplastic lesions on major tissues or organs at the end of the study.
Observations and examinations performed and frequency:
CAGE SIDE OBSERVATIONS: Yes, twice daily.

CLINICAL SIGNS: Yes, monthly.

BODY WEIGHT: Yes, every two weeks for the first 13 weeks, then monthly thereafter.

FOOD CONSUMPTION: Measurements were recorded every four weeks for the rats, but could not be precisely evaluated for the mice because of excessive spilling of feed.
Sacrifice and pathology:
SACRIFICE:
- Moribund animals and those that survived to the end of the study were killed with carbon dioxide and necropsied.

GROSS NECROPSY:
- Necropsies were performed on all moribund animals and those that survived until the end of the study.
- Necropsies were performed on all animals found dead, unless precluded in whole or in part by autolysis or cannibalisation. Thus, the number of animals from which particular organs or tissues were examined microscopically varied and did not necessarily represent the number of animals placed in the study group.
- Examinations for grossly visible lesions were performed on major tissues and organs.

HISTOPATHOLOGY:
- The following tissues underwent histopathological examination: tissue masses, lymph nodes, mammary gland, skin, salivary gland, thigh muscle, sciatic nerve, bone marrow, costochondral junction (rib), thymus, larynx, trachea, lungs and bronchi, heart, thyroid, parathyroid, esophagus, stomach, duodenum, jejunum, ileum, colon, liver, gallbladder (mice only), pancreas, spleen, kidneys, adrenals, bladder, seminal vesicles/prostate/testes, ovaries/uterus, nasal cavity, brain, pituitary, eyes, and spinal cord.
Statistics:
Probabilities of survival were estimated by the product-limit procedure of Kaplan and Meier (1958). Statistical analyses for possible dose-related effects on survival used the method of Cox (1972) for testing two groups for equality and Tarone's (1975) extensions of Cox's methods for testing for a dose-related trend. One-tailed P values were reported for all tests except the departure from linearity test, which was reported only when its two-tailed P value was less than 0.05.

The one-tailed Fisher exact test (Cox, 1970) was used to compare the tumour incidence of a control group with that of a group of treated animals at each dose level.

The Cochran-Armitage test for linear trend in proportions, with continuity correction (Armitage, 1971), was used. When a linear trend was assumed, this test determined if the slope of the dose-response curve was different from zero at the one-tailed 0.05 level of significance.

Life table methods were used to analyse the incidence of tumours. Curves of the proportions surviving without an observed tumour were computed as in Saffiotti et al. (1972). The methods of Cox and Tarone were used for the statistical tests of the groups, which were one-tailed.

A time-adjusted analysis was applied, and statistical tests were based on animals that survived at least 52 weeks, unless a tumour was found at the anatomic site of interest before this time.

The approximate 95% confidence interval for the relative risk of each dosed group compared with its control was calculated from the exact interval on the odds ratio (Gart, 1971).
Clinical signs:
no effects observed
Mortality:
no mortality observed
Body weight and weight changes:
effects observed, treatment-related
Food consumption and compound intake (if feeding study):
effects observed, treatment-related
Food efficiency:
not examined
Ophthalmological findings:
not examined
Haematological findings:
not examined
Clinical biochemistry findings:
not examined
Urinalysis findings:
not examined
Behaviour (functional findings):
not examined
Organ weight findings including organ / body weight ratios:
not examined
Gross pathological findings:
no effects observed
Histopathological findings: non-neoplastic:
effects observed, treatment-related
Histopathological findings: neoplastic:
effects observed, treatment-related
Details on results:
RATS:
- Body weight and food consumption decreased in BPA-treated males and females at both doses. No other clinical signs were observed in treated animals.
- A statistically significant increase in incidence of interstitial cell tumours of the testis was observed in males of both dose groups compared to controls. This type of lesion normally occurs in aging F344 rats at a high incidence and was not considered by the authors as convincing evidence of a BPA-related effect.

MICE:
- Body weight and food consumption decreased in BPA-treated males at the high dose and BPA-treated females at both doses. No other clinical signs were observed in treated animals.
- There were no statistically significant increases in tumour incidence at any site in BPA-treated animals compared to controls.
- There was a statistically significant increase in the incidence of multinucleated hepatocellular giant cells in males of both dose groups compared to controls.
Relevance of carcinogenic effects / potential:
The authors stated that the increased incidence of interstitial cell tumours in the testis of BPA-treated male rats is not convincing evidence of a BPA-related effect because this lesion normally occurs in aging F344 rats. There were no statistically significant increases in tumour incidence at any other site in rats or mice.
Dose descriptor:
LOAEL
Effect level:
1 000 ppm (nominal)
Sex:
male/female
Basis for effect level:
other: Based on body weight reduction in rats and female mice and increased incidence of multinucleated hepatocellular giant cells in male mice.
Remarks on result:
other: Effect type: toxicity (migrated information)
Dose descriptor:
NOAEL
Remarks:
rats
Effect level:
2 000 ppm (nominal)
Sex:
male
Basis for effect level:
other: see 'Remark'
Remarks on result:
other: Effect type: carcinogenicity (migrated information)
Dose descriptor:
NOAEL
Remarks:
mice
Effect level:
5 000 ppm (nominal)
Sex:
male/female
Basis for effect level:
other: There were no statistically significant increases in tumour incidence at any site in Bisphenol A-treated mice compared to controls.
Remarks on result:
other: Effect type: carcinogenicity (migrated information)
Conclusions:
The authors concluded that, under the conditions of this bioassay, there was no convincing evidence that bisphenol A was carcinogenic for F344 rats or B6C3F1 mice of either sex.
Executive summary:

50 rats and 50 mice per dose group consumed control diets or test diets containing either 1000 or 2000 ppm BPA (rats), 1000 or 5000 ppm BPA (male mice), or 5000 or 10000 ppm BPA (female mice) ad libitum for 103 weeks. Rats and mice were observed for five weeks or four weeks, respectively, after treatment ended. Clinical signs were recorded and all animals were necropsied and examined for neoplastic and non-neoplastic lesions on major tissues or organs at the end of the study. In rats, body weight and food consumption decreased in BPA-treated males and females at both doses. No other clinical signs were observed in treated rats. A statistically significant increase in incidence of interstitial cell tumours of the testis was observed in male rats of both dose groups compared to controls. This type of lesion normally occurs in aging F344 rats at a high incidence and was not considered by the authors as convincing evidence of a BPA-related effect. In mice, body weight and food consumption decreased in BPA-treated males at the high dose and BPA-treated females at both doses. No other clinical signs were observed in treated mice. There were no statistically significant increases in tumour incidence at any site in BPA-treated mice compared to controls. There was a statistically significant increase in the incidence of multinucleated hepatocellular giant cells in male mice of both dose groups compared to controls. The authors concluded that, under the conditions of this bioassay, there was no convincing evidence that bisphenol A was carcinogenic for F344 rats or B6C3F1 mice of either sex.

Endpoint:
carcinogenicity: oral
Type of information:
experimental study
Adequacy of study:
key study
Study period:
2014
Reliability:
1 (reliable without restriction)
Reason / purpose for cross-reference:
reference to same study
Qualifier:
no guideline followed
Principles of method if other than guideline:
Bisphenol A was administered by oral gavage from gestation day 6 through the start of labor and then directly to pups from postnatal day (PND) 1 (day of birth = PND 0) until termination at PND 90 ± 5 to Sprague-Dawley rats from the NCTR breeding colony (Sprague-Dawley/CD23/NCTR BR). Bisphenol A doses were 2.5, 8, 25, 80, 260, 840, 2,700, 100,000, and 300,000 μg/kg body weight (bw)/day.
Vehicle (0.3% carboxymethylcellulose) and naïve control groups were included to assess any effects of the gavage procedure on the endpoints measured. Two doses (0.5 and 5.0 μg/kg bw/day) of the synthetic estrogenic substance ethinyl estradiol (EE2) were also included.

The litter was the unit of statistical analysis and the target litter number was 20 per dose group (actual n = 18 – 23).

For further details on the study design see chapter toxicity to reproduction. Only investigations concerning mammary gland are discussed in this chapter.
On PND 21, one animal per sex per litter was euthanized, inguinal mammary glands were collected and fifth mammary glands were fixed in 10% NBF (neutral buffered formalin).
On PND 90 ± 5, one animal per sex per litter was euthanized. The left fifth (inguinal) mammary gland were removed and fixed in 10% NBF.
GLP compliance:
yes
Species:
rat
Strain:
other: Sprague-Dawley/CD23/NCTR BR
Sex:
male/female
Route of administration:
oral: gavage
Vehicle:
CMC (carboxymethyl cellulose)
Analytical verification of doses or concentrations:
yes
Duration of treatment / exposure:
Bisphenol A was administered by oral gavage from gestation day 6 through the start of labor and then directly to pups from postnatal day (PND) 1 (day of birth = PND 0) until termination at PND 90 ± 5
Frequency of treatment:
once a day
Post exposure period:
No
Remarks:
Doses / Concentrations:
2.5, 8, 25, 80, 260, 840, 2,700 µg/kg/day
Basis:
actual ingested
Remarks:
Doses / Concentrations:
100,000, and 300,000 µg/kg/day
Basis:
actual ingested
No. of animals per sex per dose:
The litter was the unit of statistical analysis and the target litter number was 20 per dose group (actual n = 18 – 23).
Control animals:
yes, concurrent no treatment
yes, concurrent vehicle
Dose descriptor:
other:
Effect level:
>= 2.7 - <= 300 mg/kg bw/day (actual dose received)
Based on:
test mat.
Sex:
male/female
Basis for effect level:
other: see 'Remark'
Remarks on result:
other:
Remarks:
Effect type: other: (migrated information)

↑ in the number (density) of branching ducts and alveolar buds per unit area of mammary gland in whole mount histopathology:

-      PND 21 in 2,7 and 100 mg/kg dose groups; no significant observation at 300 mg/kg.

-      PND 90 in 300 mg/kg dose group.

Memorandumdated August 2, 2013n byYan Gu:

“Taking the incidences, statistical testing results, and all pathologist and study author opinions together, we conclude that the evidence for duct hyperplasia in the mammary gland of females on either PND 21 or 90 is weak. We consider it to be an equivocal finding that may be a reflection of normal biological variability and/or a reflection of limits in tissue processing. We agree with the study authors that there were no clearly adverse effects in the low-dose range. However, we also believe that this conclusion will be tested by the results of the chronic toxicity study with BPA, for which the current study served primarily as a dose-range finding study.”

Memorandum dateJuly 19, 2013 by Steven Mog:

“Therefore, we agree with the toxicologist that there are no significant mammary gland effects at the low does BPA range including the 2700 dose group as stated in the overall summary conclusion of the toxicology report (reference 2, pg. 9): “In conclusion, BPA in the dose region between 2.5 and 2700 μg/kg bw/day did not exhibit clear adverse effects in either sex under the conditions of this study”.

NCTR GLP/NTP Technical Report by Barry Delclos – May 2013:

NCTR investigated also intraductular hyperplasia and carcinoma in sito following the method by Murray. 2007. Reprod Toxicol. 23:383-390.NCTR reported high variability of the method and “no apparent BPA treatment effects was discerned”. The data reported by the Soto Laboratory in rats were not confirmed by NCTR

No effect on mammary gland transcriptom;“The expression changes below the NOAEL[5 mg/kg-bw/day]were observed in single BPA doses and/or did not show a consistent dose response pattern, suggesting that BPA has limited effects in the gene expression of mammary glands in the low dose range.” Overall conclusion by Delclos et al 2014: "Taking the incidences, the statistical testing results, and all pathologists and study authors opinions together, the authors of the NTP report (Gu and Mitkus, 2013), concluded that the evidence for duct hyperplasia in the mammary gland of females on either PND 21 or PND 90 was weak. They considered it an equivocal finding that may be the reflection of normal variability and/or a reflection of limits in tissue processing. BPA did not cause duct hyperplasia in the mammary glands of male rats, while conversely the reference estrogen EE2 induced hyperplasia in the male but not the female mammary gland."
Executive summary:

Overall conclusion concerning mammary gland:"Taking the incidences, the statistical testing results, and all pathologists and study authors opinions together, the authors of the NTP report (Gu and Mitkus, 2013), concluded that the evidence for duct hyperplasia in the mammary gland of females on either PND 21 or PND 90 was weak. They considered it an equivocal finding that may be the reflection of normal variability and/or a reflection of limits in tissue processing. BPA did not cause duct hyperplasia in the mammary glands of male rats, while conversely the reference estrogen EE2 induced hyperplasia in the male but not the female mammary gland."

Endpoint:
carcinogenicity: oral
Type of information:
experimental study
Adequacy of study:
key study
Study period:
2012 -
Reliability:
1 (reliable without restriction)
Rationale for reliability incl. deficiencies:
test procedure in accordance with generally accepted scientific standards and described in sufficient detail
Qualifier:
according to guideline
Guideline:
other: Guideline-compliant research standard conducted at the FDA designed by NCTR and NIEHS scientists
Principles of method if other than guideline:
The core GLP study was designed to characterize and evaluate the toxicologic potential of BPA follow ing perinatal only (stop-dose study arm) or chronic exposure in rats under the conditions of a chronic, extended-dose response design. Sacrifice was conducted at one year of age (PND 365 ± 20) and the terminal sacrifice was conducted at two years of age (PND 730 ± 20). The interim (one-year) sacrifice group was included to allow evaluation of long-term exposure effects with less confounding due to background lesions of aging than would be expected at two years (for details see chapter 7.5.1). EE2 was included for this purpose rather than necessarily to compare effects of the two agents. PND 1 pups (in utero exposure is described in chapter 7.8.2) were weighed and daily dosed until weaning at PND 21. After weaning the pubs were assigned either to the chronic study (continuous dose arm and stop-dose study arm) or to the hypothesis-driven studies of academic investigators. In the chronic study postweaning animals were dosed daily throughout the study (termination at 1 or 2 years) = continuous-dose arm) or were held without further treatment until termination (1 or 2 years). This stop-dose study arm was included to assess any effects that were due to early exposure only.
GLP compliance:
yes
Specific details on test material used for the study:
The two high BPA dosing suspensions were mixed by directly adding BPA solid to the vehicle with sonication, and the suspensions were stirred constantly. The three low BPA and the two EE2 dosing solutions were mixed by serial dilutions of stock solutions.
Species:
rat
Strain:
Sprague-Dawley
Remarks:
CD23/NctrBR
Details on species / strain selection:
The animal model used in these studies was the Sprague-Dawley rat maintained at NCTR. This colony had its origins in the late 1970s from Charles River Sprague-Dawley founders and has been used in toxicology studies with hormonally active agents for over a decade at NCTR.
Sex:
male/female
Details on test animals or test system and environmental conditions:
for details see chapter 7.5.1
Route of administration:
oral: gavage
Vehicle:
CMC (carboxymethyl cellulose)
Details on exposure:
for details see chapter 7.5.1
Analytical verification of doses or concentrations:
yes
Details on analytical verification of doses or concentrations:
At the start of the study and approximately every 8–10 weeks over the course of the study, all dose level preparations were assayed by the Chemistry Support Group prior to delivery to the animal rooms and certified to be within 10% of the target concentration with a % CV of ≤10%. In addition, at intervals spaced 4-7 months apart, BPA and EE2 dosing preparations from the animal rooms were assayed at the end of their use to verify the dose concentrations.
Duration of treatment / exposure:
From PND1 to PND21 (stop-dose arm study)
From PND 1 to 1 year or 2 years (continuous-dose arm study)
Frequency of treatment:
once daily
Post exposure period:
stop-dose arm: 1 year or 2 year
Dose / conc.:
2.5 other: µg/kg bw/day
Dose / conc.:
25 other: µg/kg bw/day
Dose / conc.:
250 other: µg/kg bw/day
Dose / conc.:
2 500 other: µg/kg bw/day
Dose / conc.:
25 000 other: µg/kg bw/day
No. of animals per sex per dose:
At weaning, up to a maximum of three pups/sex/litter were assigned to the chronic study. Same-sex littermates were not assigned to the same combination of study dose arm and time of sacrifice so that litter of origin was not a factor to be considered in the statistical analysis of endpoints collected after weaning.
20 to 26 pups/sex/BPA dose group were assigned to the one-year interim continuous-dose assessment
19 to 22 pups/sex/BPA dose group were assigned to the one-year interim stop-dose assessment
26 to 50 pups/sex/BPA dose group/dose arm were assigned to the two year continuous-dose assess
ment
46 to 50 pups/sex/BPA dose group/dose arm were assigned to the two year study stop-dose assess
ment
26 pups/sex/EE2 dose group were assignedto the one-year interim continuous-dose assessment
26 pups/sex/EE2 dose group were assigned to the two-year study continuous-dose assessment
The reason that stop-dose EE2 groups were not included in the study was solely an animal facility space and resource consideration, given the number of animals that needed to be provided and housed for both this study and the NIEHS-funded academic CLARITY-BPA grantee studies
Control animals:
yes, concurrent vehicle
Details on study design:
for details see chapter 7.5.1
Positive control:
ethinyl estradiol (EE2) - 0.05 and 0.5 EE2 μg/kg bw/day
Observations and examinations performed and frequency:
for details see chapter 7.5.1
Sacrifice and pathology:
Procedures were similar to those used for the interim sacrifice, except that the animals were not fasted, blood was not collected, and there were no hematology, clinical chemistry, sperm, or organ weight data collected (for details see chapter 7.5.1).
Other examinations:
for details see chapter 7.5.1
Statistics:
The full statistical analysis reports for all protocol-specified endpoints, including detailed methods and results for each analysis, including the omnibus tests, are found in Supplemental Appendices of the study report. The pairwise comparisons to the vehicle control and trend tests are the comparisons of interest that are presented in the tables in study report. The statistical methodology for each endpoint is summarized in the study report. Statistical comparisons were conducted within sex and, for data collected after weaning, within dosing arm (continuous-dose or stop-dose). For pairwise comparisons, the five BPA dose groups were compared to the vehicle control group. Similarly, the two EE2 reference estrogen dose groups were compared to the vehicle control. Tests were conducted at the 0.05 significance level and, in most cases, were two-sided. Exceptions were one-sided tests for the pairwise comparisons of histopathology lesion incidence and severity to vehicle controls and trend tests for abnormal estrous cycles. Although a p-value of <0.05 was used to flag a result as significant, the actual p-values are included in some of the tables in the study report and in all the statistical report appendices (Supplemental Appendices of the study report) to aid in the further evaluation of the statistical and biological significance of each result. Trend tests for treatment effect (either increased or decreased relative to vehicle control) with increasing dose were conducted only for vehicle control and BPA treatment groups, except for non-neoplastic and neoplastic lesions, where trend tests were also conducted within the vehicle control and EE2 groups. Because pups within litter and sex were assigned at weaning to different dosing arms and sacrifice times, litter correlation was not a consideration for endpoints evaluated after weaning in this study.
Clinical signs:
not specified
Mortality:
mortality observed, non-treatment-related
Description (incidence):
for details see chapter 7.5.1
Body weight and weight changes:
effects observed, non-treatment-related
Description (incidence and severity):
for details see chapter 7.5.1
Food consumption and compound intake (if feeding study):
not specified
Water consumption and compound intake (if drinking water study):
not examined
Ophthalmological findings:
not examined
Haematological findings:
effects observed, non-treatment-related
Description (incidence and severity):
for details see chapter 7.5.1
Clinical biochemistry findings:
effects observed, non-treatment-related
Description (incidence and severity):
for details see chapter 7.5.1
Urinalysis findings:
not examined
Behaviour (functional findings):
not examined
Immunological findings:
not examined
Organ weight findings including organ / body weight ratios:
effects observed, non-treatment-related
Description (incidence and severity):
for details see chapter 7.5.1
Gross pathological findings:
not specified
Neuropathological findings:
not examined
Histopathological findings: non-neoplastic:
effects observed, treatment-related
Description (incidence and severity):
for details see chapter 7.5.1
Histopathological findings: neoplastic:
effects observed, non-treatment-related
Description (incidence and severity):
Females:
In the stop-dose BPA study arm at two years, there was a statistically significant increase in the incidence of female mammary gland adenocarcinoma (22% versus 6%; p = 0.016) and the combination of adenoma and adenocarcinoma (24% versus 8%; p = 0.018) in the 2.5 μg BPA/kg bw/day dose group. No increase in female mammary gland neoplasms was observed in the continuous BPA dose arm at two years. There were no significant treatment-related non-neoplastic lesions in the mammary gland of interim or terminal sacrifice stop-dose BPA groups. In the interim and terminal BPA continuous dosing arm, there was an increase, significant by the secondary RTE test only, in female mammary gland atypical foci at 2.5 μg BPA/kg bw/day (14% versus 0% and 15% versus 4% for the interim and terminal dose group animals, respectively). Increased adenoma/adenocarcinoma incidence observed only in the stop-dose animals, lack of a dose response, absence of non-neoplastic lesions in interim or terminal sacrifice stop-dose animals, and comparison to limited historical control data for this strain of rats in experiments conducted at NCTR bring into question the biological plausibility of this lesion as a BPA treatment-related effect. In addition to mammary gland neoplasms, a significant trend (p = 0.037) for uterine stromal polyps in the interim sacrifice animals in the continuous BPA dose arm was observed; this was not observed in the terminal sacrifice animals.

Males:
There were no significant differences between treatment groups and vehicle controls in the incidences of neoplastic lesions in stop-dose or continuous-dose interim or terminal sacrifice males.
Relevance of carcinogenic effects / potential:
At the terminal sacrifice of females in the stop-dose BPA study arm, there was a statistically significant increase in the incidence of mammary gland adenocarcinoma (22% versus 6%, p = 0.016) and the combination of adenoma/adenocarcinoma (24% versus 8%, p = 0.018) in the 2.5 μg BPA/kg bw/day dose group. This incidence is higher than the limited data available for historical controls of this rat strain at NCTR utilizing the same diet,14; 15 which indicate a background rate of 11–16% for mammary gland adenocarcinoma in two-year-old control females. In the continuous-dose BPA study arm, the incidence of female mammary gland adenocarcinoma and the combination of adenoma/adenocarcinoma was 8% and 12%, respectively, in the control group. The incidence in each of the BPA continuous-dose groups varied between 6 and 18% for adenocarcinoma and between 9 and 20% for combined adenoma/adenocarcinoma, none of which was significant.
Evaluation of the totality of the evidence regarding the elevated incidence of mammary adenocarcinomas or combined adenomas and adenocarcinomas in the stop-dose females exposed to 2.5 μg BPA/kg bw/day makes it unlikely that this is a plausible BPA treatment-related lesion.
The only other statistically significant neoplastic lesion in BPA-treated females was a significant trend (p = 0.037) for uterine stromal polyps in the interim continuous-dose arm animals. This trend was driven by incidences of 3/20 (15%) and 3/24 (12%) in the 2,500 and 25,000 μg BPA/kg bw/day dose groups, respectively, which were not statistically different in pairwise comparisons with the control incidence of 1/23 (4%). No trend toward higherincidences in the higher BPA dose groups was evident in the terminal sacrifice continuous-dose animals, nor were there any significant pairwise comparisons between any BPA dose group and controls in the terminal sacrifice animals. Likewise, stromal polyps were not induced in the stop-dose BPA or in the interim or terminal sacrifice EE2 animals. The relatively small increased incidences leading to the significant trend in the interim sacrifice animals together with the lack of any effects in the terminal sacrifice animals indicate that uterine stromal polyps are not likely to be a consequence of BPA treatment.
For males, there were no significant differences in the incidences of neoplasms in treatment groups versus vehicle controls in any organ in stop-dose or continuous-dose interim or terminal sacrifice males.
Dose descriptor:
NOAEL
Effect level:
25 000 other: mg/kg bw/day
Based on:
test mat.
Sex:
male/female
Basis for effect level:
histopathology: neoplastic
Critical effects observed:
no
Executive summary:

The core GLP study was designed to characterize and evaluate the toxicologic potential of BPA following perinatal only (stop-dose study arm) or chronic exposure (continuous-arm study) in

Sprague-Dawley rats from the NCTR breeding colony (Sprague-Dawley/CD23/NctrBR) under the conditions of a chronic, extended-dose response design.

For that purpose first of all dams (P0) were daily dosed by gavage with 2.5, 25, 250, 2,500, and 25,000 μg BPA/kg body weight (bw)/day. Also reference estrogen groups (0.05 and 0.5 µg ethinyl estradiol/kg bw/day) and vehicle control group ( 0.3% CMC) were included. In utero exposure began on GD 6 and continued until the initiation of parturition.

Then the pups (F1) were daily dosed by gavage from postnatal day (PND) 1 (day of birth = PND 0) until termination at one year or two years was examined . BPA doses were unchanged 2.5, 25, 250, 2,500, and 25,000 μg/kg body weight (bw)/day. A vehicle (0.3% carboxymethylcellulose (CMC)) control group was also included. In addition to animals that were dosed daily throughout the study, a stop-dose study arm was included with animals dosed daily until PND 21 and then held without further treatment until termination at one year or two years. Two dose groups which received orally the active estrogen ethinyl estradiol (EE2; (0.05 and 0.5 μg/kg bw/day) were also included in the continuous-dose arm but not in the stop dose study arm of the study.

The approach used by the NTP to determine if neoplasms are likely related to treatment has always focused on a weight-of-evidence approach, encompassing a range of factors, including statistical analysis, consideration of consistency of responses within the study, and historical controls. This approach was followed in the present study to evaluate the histopathology data for the neoplastic and non-neoplastic lesions (for results see chapter 7.5.1). There were relatively few neoplastic lesions that showed potential treatment effects. There were many non-neoplastic lesions in both males and females that were variable across control and BPA treatment levels. One issue that contributed to the challenge of clearly establishing effects as treatment related was that most diagnosed lesions, whether neoplastic or non-neoplastic, were not rare lesions, but rather common lesions associated with aging.

The elevated incidence of mammary adenocarcinomas or combined adenomas and adenocarcinomas in the stop-dose females exposed to 2.5 μg BPA/kg bw/day and a significant trend (p = 0.037) for uterine stromal polyps in the interim continuous-dose arm of BPA-treated female animals were analyzed in depth. The results of the in depth analysis make it unlikely that these are plausible BPA treatment-related lesions.

For males, there were no significant differences in the incidences of neoplasms in treatment groups versus vehicle controls in any organ in stop-dose or continuous-dose interim or terminal sacrifice males.

That means that under the conditions of this bioassay, there was no convincing evidence that bisphenol A was carcinogenic for Spargue-Dawley rats.

Endpoint conclusion
Endpoint conclusion:
no adverse effect observed
Dose descriptor:
NOAEL
25 mg/kg bw/day
Study duration:
chronic
Species:
rat
Quality of whole database:
Life-long exposure studies are available in mice and rats

Carcinogenicity: via inhalation route

Endpoint conclusion
Endpoint conclusion:
no study available

Carcinogenicity: via dermal route

Endpoint conclusion
Endpoint conclusion:
no study available

Justification for classification or non-classification

Bisphenol A is included in Annex VI of Regulation (EC) No 1272/2008. No classification regarding carcinogenicity is required. 

Additional information

Recent data taken into account for the dossier update 

CLARITY Core study: 

The CLARITY-BPA core GLP study was designed to characterize and evaluate the toxicologic potential of BPA following perinatal only (stop-dose study arm) or chronic exposure (continuous-arm study) in Sprague-Dawley rats from the NCTR breeding colony (Sprague-Dawley/CD23/NctrBR) under the conditions of a chronic, extended-dose response design. To this, dams (P0) were daily dosed by gavage with 2.5, 25, 250, 2500, and 25000 μg BPA/kg body weight (bw)/day. Also reference estrogen groups (0.05 and 0.5 µg ethinyl estradiol/kg bw/day) and vehicle control group (0.3% CMC) were included. In utero exposure began on GD 6 and continued until the initiation of parturition. Then the pups (F1) were daily dosed by gavage from postnatal day (PND) 1 (day of birth = PND 0) until termination at one year or two years was examined . BPA doses were unchanged 2.5, 25, 250, 2500, and 25000 μg/kg body weight (bw)/day. A vehicle (0.3% carboxymethylcellulose (CMC)) control group was also included. In addition to animals that were dosed daily throughout the study, a stop-dose study arm was included with animals dosed daily until PND 21 and then held without further treatment until termination at one year or two years. Two dose groups which received orally the active estrogen ethinyl estradiol (EE2; (0.05 and 0.5 μg/kg bw/day) were also included in the continuous-dose arm but not in the stop dose study arm of the study. 

The approach used by the NTP to determine if neoplasms are likely related to treatment has always focused on a weight-of-evidence approach, encompassing a range of factors, including statistical analysis, consideration of consistency of responses within the study, and historical controls. This approach was followed in the present study to evaluate the histopathology data for the neoplastic and non-neoplastic lesions (for results see chapter 7.5.1). There were relatively few neoplastic lesions that showed potential treatment effects. There were many non-neoplastic lesions in both males and females that were variable across control and BPA treatment levels. The elevated incidence of mammary adenocarcinomas or combined adenomas and adenocarcinomas in the stop-dose females exposed to 2.5 μg BPA/kg bw/day and a significant trend (p = 0.037) for uterine stromal polyps in the interim continuous-dose arm of BPA-treated female animals were analyzed in depth. The results of the in depth analysis make it unlikely that these are plausible BPA treatment-related lesions. For males, there were no significant differences in the incidences of neoplasms in treatment groups versus vehicle controls in any organ in stop-dose or continuous-dose interim or terminal sacrifice males. That means that under the conditions of this bioassay, there was no convincing evidence that bisphenol A was carcinogenic for Sprague-Dawley rats. 

Delclos et al., 2014: 

A recent study by US NCTR (Delclos et al 2014) investigated mammary gland proliferation in rats. This study was evaluated by SCOEL and EFSA. Ethinyl oestradiol was used as a positive control of the estrogenic effects of Bisphenol A. The dose-matched vehicle control was carboxymethylcellulose. The doses were: (i) Bisphenol A 2.5, 8, 25, 80, 260, 840, 2700, 100 000, 300 000 μg/kg bw per day, (ii) Vehicle, (iii) EE2 0.5, 5 μg/kg bw per day. The study included a naïve control group and doses were administered by oral gavage. The protocol and methods, including statistical analysis were of the high quality and robust with treatment, body weight and litter randomisation and appropriate inclusion and exclusion criteria established prior to the start of the study. The target unit for analysis was 20 litters and 18-23 were achieved. F0 females were dosed from GD 6 up to labour onset and pups from PND 1 until tissue harvesting, up to PND 90. Additional groups were exposed from GD 6 to PND 21 for histopathological examination of the mammary glands.  

Overall conclusion concerning mammary gland: "Taking the incidences, the statistical testing results, and all pathologists and study authors opinions together, the authors of the NTP report (Gu and Mitkus, 2013), concluded that the evidence for duct hyperplasia in the mammary gland of females on either PND 21 or PND 90 was weak. They considered it an equivocal finding that may be the reflection of normal variability and/or a reflection of limits in tissue processing. Bisphenol A did not cause duct hyperplasia in the mammary glands of male rats, while conversely the reference estrogen EE2 induced hyperplasia in the male but not the female mammary gland". 

 

EFSA Opinion 2015 concluded:  

Proliferative and morphological changes potentially related to carcinogenesis 

"Earlier evidence for Bisphenol A effects on cell proliferation and differentiation in the mammary gland and other tissues (e.g. prostate or testis) has been supported by recent studies. The proliferative changes in the mammary gland reported in these new studies, including a non-human primate study, are insufficient to conclude that there is a link to cancer development in later life. However, there might be a possible role of Bisphenol A in increasing the susceptibility to mammary gland carcinogenesis later in life.  

 

The proliferative responses and possibly enhanced sensitivity to mammary gland carcinogens seen in animal studies might be of relevance for human health and are therefore included in the risk assessment.  

 

Using a WoE approach, the CEF Panel assigned a likelihood level of “likely” to Bisphenol A induced proliferative changes in the mammary gland. Therefore, this endpoint was brought forward for hazard characterisation and for uncertainty analysis. 

The CEF Panel considered that the evidence for proliferative changes induced by Bisphenol A in other organs (e.g. prostate or testis) is currently too limited to reach any conclusion." 

 

Carcinogenicity: 

“Very few epidemiological studies published to date have investigated a possible association between exposure to Bisphenol A and incidence of certain cancers, specifically breast cancer and meningioma. These studies do not allow any conclusion to be drawn regarding the carcinogenicity of Bisphenol A in humans.  

 

Bisphenol A was not carcinogenic in two standard oral carcinogenicity studies in rats and mice. In a more recent study, female but not male mice, exposed to approximately 10 mg/kg bw per day Bisphenol A from in utero up to postnatal day (PND) 21, developed significantly more hepatocellular tumours (adenomas and carcinomas together) with or without preneoplastic lesions after a stop-exposure period of 10 months. Additional rodent studies on perinatal exposure to Bisphenol A investigated the potential carcinogenic effect in mammary gland. Due to weaknesses in these studies the results do not provide convincing evidence that Bisphenol A is carcinogenic to the liver during adult life or in mammary gland following perinatal exposure.  

 

Using a WoE approach, the CEF Panel assigned a likelihood level of “unlikely to - as likely as not -” to carcinogenic effects of Bisphenol A. Since the likelihood level for this endpoint is less than "as likely as not” (see Appendix A), this endpoint was not taken into account in the evaluation of uncertainty for hazard characterisation and risk characterisation.“ 

 

SCOEL Recommendation 2014 concluded on carcinogenicity: 

"Thus, there is currently no convincing evidence of carcinogenicity of Bisphenol A when administered either during the adulthood or perinatally. However, as concluded by EFSA (2014), there are some data (including the data by Delclos et al 2014) that raise some concern for Bisphenol A effects on mammary gland cell proliferation after pre- and perinatal exposure. Whether this is linked to increased cancer incidence in later life or not remains to be shown."  

 

The 2008 updated EU RAR concluded:  

"The new information on the potential carcinogenic and/or promoting effects of Bisphenol A in prenatal and neonatal rat models supports the original conclusion that Bisphenol A does not possess any significant carcinogenic potential. This is based on one new study in which the full carcinogenic potential of Bisphenol A on the mammary gland was examined in a prenatal model. This study claimed that Bisphenol A induced preneoplastic and neoplastic lesions of the mammary gland, but its validity was hampered by serious methodological limitations and its findings are inconsistent with the absence of preneoplastic lesions of the mammary gland in the offspring from several standard multi-generation studies in rats and mice. Other new studies suggest that prenatal or neonatal exposure to Bisphenol A does not exert promoting activity on the carcinogenesis induced by established carcinogens/initiators in specific organs."  

 

The 2003 EU RAR concluded:  

"There are no human data contributing to the assessment of whether or not Bisphenol A is carcinogenic, but a dietary carcinogenicity study in rats and mice concluded that Bisphenol A was not carcinogenic in either species because the tumour findings were not considered toxicologically significant. No inhalation or dermal carcinogenicity studies were available, although in repeat exposure inhalation toxicity studies, Bisphenol A did not exhibit properties that raised concern for potential carcinogenicity. Taking into account all the animal data available, it was concluded that the animal evidence suggests that Bisphenol A does not have carcinogenic potential."