Registration Dossier

Diss Factsheets

Toxicological information

Carcinogenicity

Currently viewing:

Administrative data

Description of key information

According to column 2 of REACH Annex X, information requirement 8.9.1, a carcinogenicity study does not need to be conducted if the substance is classified as a germ cell mutagen category 1A or 1B. Since the registereed substance is reasonably known to be mutagenic category 1B, it is considered justified to omit testing for this endpoint. Additionally, the substance is already classified as carcinogens category 1B. Further testing for carcinogenicity is neither considered necessary nor considered to be an expedient or responsible use of animals.


Please refer to the document ‘Consideration of the long term toxicity of Distillates (shale oil)’, as included in section 13, for further information.


 


Supporting data (as considered in a weight of evidence approach):


 


RIVM (Rat Study on benzo[a]pyrene)


This study demonstrated the carcinogenic potential of the test material in a variety of organs when administered by gavage to rats. Major target organs were liver, forestomach and epidermal structures, of which the liver is considered the most relevant for human risk assessment in terms of pathogenesis and sensitivity.


 


RIVM (Mouse study on benzo[a]pyrene)


Target tissues for carcinogenicity in mice exposed to benzo[a]pyrene via the diet were the forestomach, oesophagus and tongue. No clear treatment-related induction of liver tumours was found.


 


IARC Monograph of high-temperature crude shale oils, low-temperature crude shale-oils, fractions of high-temperature shale-oil, crude shale-oil distillation fractions, shale-oil bitumens and commercial blends of shale-oils


There is sufficient evidence for the carcinogenicity in experimental animals of high-temperature crude shale oils, low-temperature crude shale-oils, fractions of high-temperature shale-oil, crude shale-oil distillation fractions, shale-oil bitumens and commercial blends of shale-oils.


There is limited evidence for the carcinogenicity in experimental animals of raw oil shale, spent oil shale and a residue of shale-oil distillation.


There is sufficient evidence that shale-oils are carcinogenic in humans.

Key value for chemical safety assessment

Carcinogenicity: via oral route

Link to relevant study records

Referenceopen allclose all

Endpoint:
carcinogenicity
Data waiving:
study scientifically not necessary / other information available
Justification for data waiving:
a carcinogenicity study does not need to be conducted because the substance is classified as germ cell mutagen category 1A or 1B and the default presumption is that a genotoxic mechanism for carcinogenicity is likely
Justification for type of information:
JUSTIFICATION FOR DATA WAIVING
According to column 2 of REACH Annex X, information requirement 8.9.1, a carcinogenicity study does not need to be conducted if the substance is classified as a germ cell mutagen category 1A or 1B. Since the registereed substance is reasonably known to be mutagenic category 1B, it is considered justified to omit testing for this endpoint. Additionally, the substance is already classified as carcinogens category 1B. Further testing for carcinogenicity is neither considered necessary nor considered to be an expedient or responsible use of animals.
Please refer to the attached document '1904651.UK0 - 7899 Consideration of the long term toxicity of Distillates (shale oil)' for further information.
Endpoint:
carcinogenicity
Type of information:
other: discussion on a variety of mamallian studies.
Adequacy of study:
weight of evidence
Study period:
Not reported.
Reliability:
4 (not assignable)
Rationale for reliability incl. deficiencies:
other: Secondary review paper with very little experimental details. No quantitative results provided.
Qualifier:
no guideline available
GLP compliance:
not specified

Exposures

Commercial oil-shale processing industries presently exist only in the USSR and in China (a commercial facility is expected to begin operation in the USA in 1984), producing crude shale-oils that are used as fuels or chemical-plant feedstocks. Crude shale-oils were produced in the past in several countries; an industry existed in the UK for more than 100 years.

Workers involved in oil-shale mining and processing may be exposed to complex mixtures of dusts, gases and vapours. The dusts may contain significant levels of crystalline silica. Inorganic gases and vapours to which workers may be exposed include carbon monoxide and hydrogen sulphide. Workers may also be exposed to gases and vapours containing organic compounds, including low levels of polynuclear aromatic compounds. The composition of spent oil shale can vary widely in respect to the remaining tar material, depending on the variation of retorting methods. Skin contact with crude shale-oil may occur, but is limited primarily to maintenance workers in modern oil-shale processing facilities.

Contact with shale-oil liquids occurred extensively in the past in the Scottish shale-oil industry, and in the British cotton-textile industry where lubricants derived from shale-oils were used.

Experimental data

Extracts of raw oil shale from Scotland (UK) and the Green River Formation (USA) were tested by skin application in mice, and no skin tumour was observed. A suspension of raw oil-shale dust from the Green River Formation was also tested by intratracheal instillation in rats and hamsters, and no local tumour was observed; exposure of rats by inhalation to raw oil-shale dust resulted in the induction of lung tumours. Crude shale-oils from low-temperature retorting from different sources were tested for carcinogenicity in

various experiments by skin application in different strains of mice. Samples from Jurassic Chuvash (USSR), Estonia (USSR), the Green River Formation (USA) and Fushun (China) all resulted in the induction of benign and malignant skin tumours. A sample from Estonia was also tested by skin application in rabbits and induced benign and malignant skin tumours. Lung tumours were produced in mice following intratracheal administration of a crude shale-oil from the Green River Formation.

Three samples of crude shale-oils from high-temperature retorting from Estonia were tested for carcinogenicity by skin application in mice, and one sample was also tested in rabbits. All the samples resulted in the induction of benign and malignant skin tumours; these samples were more carcinogenic than crude shale-oils from lowtemperature retorting from the same source.

An extract of spent oil shale from the Green River Formation resulted in the induction of skin tumours in mice after topical application. Dusts prepared from this sample induced lung tumours in rats after inhalation exposure. No lung tumour occurred in rats or hamsters exposed by intratracheal administration to a suspension of the spent oil-shale dusts.

Various fractions of low- and high-temperature shale-oils were tested by skin application in mice and rabbits and by intramuscular application in mice; their carcinogenic activities did not necessarily parallel the benzo[a]pyrene content of the fractions.

Various crude shale-oil distillation fractions from Scotland were tested by skin application in mice; the less refined shale-oils were more highly carcinogenic to the skin than the more refined products. Heavy fractions of shale-oil from Estonia were more carcinogenic than the light fractions or the total oil when tested in mice by skin application; in contrast, in a study from Scotland, light distillation fractions of a lubricating oil induced more tumours than heavier fractions.

Comparative carcinogenicity studies in mice by skin application indicate that residual shale-oil bitumen (Estonia) was more active in inducing skin tumours than blown (oxidized) bitumen.

Commercial samples representing blends of shale-oils from Estonia induced skin tumours in mice after topical application; the carcinogenic effect increased with increasing content of crude shale-oil from high-temperature retorting. In similar experiments, commercial products containing low-temperature retorting oils did not induce skin tumours.

A pot residue of shale-oil distillation ('shale-oil coke') from the Green River Formation was carcinogenic to mouse skin after topical application in benzene; however, the same sample did not induce respiratory tumours in hamsters after intratracheal instillation.

No relevant data were available to the Working Group on the carcinogenicity in experimental animals of oil shale retort process-waters.

All the shale-derived materials tested in short-term tests came from sources in the USA, and were therefore all from low-temperature processes.

Chromosomal aberrations were induced in bone-marrow cells of rats following administration by gavage of a suspension of raw oil shale. In-vitro tests of extracts of raw oil shale in bacteria, yeast and cultured mammalian cells gave negative results.

Preparations of spent oil shale yielded contradictory results in bacterial mutation assays and were negative in mutation assays with eukaryotic cells in vitro and in a chromosomal assay in vivo.

Preparations of shale-derived crude oils from various sources and retort processes were mutagenic in bacteria, yeast and cultured mammalian cells following metabolic or photo-induced activation. Three crude shale-oil preparations did not induce mitotic gene conversion in yeast; two others induced sister chromatid exchanges in cultured mammalian cells. Both positive and negative results were obtained in mammalian in vivo assays for chromosomal effects.

As compared with the corresponding crude shale-oils, preparations of hydrotreated oils showed decreased activity or were negative in various short-term tests.

A preparation of refined shale-oil was not mutagenic in bacteria.

Oil-shale retort process-waters elicited DNA damage and mutations in bacteria and in cultured mammalian cells following metabolic activation or photoactivation. They induced chromosomal aberrations in cultured mammalian cells; and induced chromosomal aberrations but not sister chromatid exchanges in mouse cells in vivo.

Extracts of oil-shale ash were mutagenic in bacteria both in reversion and forward mutation assays in the absence of a metabolic system. Tests with eukaryotic systems gave negative results.

Human data

The association between shale-oils and skin cancers, particularly of the scrotum, was demonstrated by analyses of 65 cases of skin cancer, including 31 of the scrotum, from the Scottish shale-oil industry. In the UK, over 2000 cases of skin cancer ('mule-spinners' cancer') were recorded among cotton-textile workers and others exposed to lubricating oils (many of which are believed to have been shale-derived). The occupational etiology of these cases is supported by occupational mortality statistics for the UK and by an occupational comparison with fatal cases of penile cancer. In contrast, one study reported very few scrotal cancers among US cotton-textile workers employed in mills where shale-derived lubricants were not used. A cohort study of shale-oil workers in western USA found statistically significant excesses of total cancer and of colon cancer, although data on duration and time since first exposure were not available. A cohort study of shale-oil workers in Estonia found significant excesses of skin cancer, but not of cancers at other sites.

Conclusions:
There is sufficient evidence for the carcinogenicity in experimental animals of high-temperature crude shale oils, low-temperature crude shale-oils, fractions of high-temperature shale-oil, crude shale-oil distillation fractions, shale-oil bitumens and commercial blends of shale-oils.

There is limited evidence for the carcinogenicity in experimental animals of raw oil shale, spent oil shale and a residue of shale-oil distillation.

There is sufficient evidence that shale-oils are carcinogenic in humans.
Executive summary:

There is sufficient evidence for the carcinogenicity in experimental animals of high-temperature crude shale oils, low-temperature crude shale-oils, fractions of high-temperature shale-oil, crude shale-oil distillation fractions, shale-oil bitumens and commercial blends of shale-oils.

There is limited evidence for the carcinogenicity in experimental animals of raw oil shale, spent oil shale and a residue of shale-oil distillation.

There is sufficient evidence that shale-oils are carcinogenic in humans.

Endpoint:
carcinogenicity: oral
Type of information:
experimental study
Adequacy of study:
weight of evidence
Study period:
not reported
Reliability:
4 (not assignable)
Rationale for reliability incl. deficiencies:
secondary literature
Qualifier:
no guideline followed
Principles of method if other than guideline:
The report includes (as an Annex) a summary of findings from mouse diet studies which were conducted at NCTR (Culp et al., 1994 and 1998). The NCTR project incorporates a chronic oral carcinogenicity study with B[a]P.
GLP compliance:
not specified
Species:
mouse
Strain:
B6C3F1
Sex:
female
Details on test animals or test system and environmental conditions:
TEST ANIMALS
- Age at study initiation: 5 weeks
Route of administration:
oral: feed
Analytical verification of doses or concentrations:
not specified
Duration of treatment / exposure:
24 months
Frequency of treatment:
continuously
Dose / conc.:
5 ppm
Dose / conc.:
25 ppm
Dose / conc.:
100 ppm
No. of animals per sex per dose:
48 females per dose
Control animals:
yes
Remarks on result:
not measured/tested
Critical effects observed:
yes
Lowest effective dose / conc.:
5 ppm
System:
gastrointestinal tract
Organ:
oesophagus
stomach
tongue
Treatment related:
yes

With respect to B[a]P, at the highest dose tested a decrease in body weight was observed from week 50 onwards, most probably due to effects secondary to tumour-development in the upper digestive tract. B[a]P induces in these female B6C3F1 mice tumours in the epithelial layer of the tongue, the esophagus, and the forestomach. Also, increased incidences of sarcomas and of tumours in the squamous cell of the larynx were observed. Clearly, B[a]P treatment did not induce tumours in the liver and lung, nor hemangiosarcomas or histiocytic sarcomas.

Table 1: Incidences of neoplasms in female B6C3F1 mice treated with B[a]P. Number of tumour-bearing animals (TBA) per dose-group are indicated as: TBA/ number examined

Agent

B[a]P1

Dose2

Tongue3

Oesophagus4

Forestomach5

Liver6

Lung7

Small intestine8

Hemang.sarc9

B[a]P

0

 

0/48

0/48

0/48

2/48

5/48

no

1/48

5

0.6

0/48

0/48

3/47

7/48

0/48

no

2/48

25

3

2/46

2/45

36/46

5/47

4/45

no

3/47

100

12

23/48

27/46

46/47

0/45

0/48

no

0/48

CT control

0

 

0/48

0/48

1/48

0/47

2/47

0/47

1/48

 

1) B[a]P concentrations in the administered diets (in ppm);

2) (rounded) calculated daily intake of B[a]P in mg/kg bw, corrected for reductions in food consumption;

3) Tongue (dorsal surface of the base) papillomas and/or squamous cell carcinomas;

4) Oesophagus squamous cell papillomas and/or carcinomas;

5) Forestomach papillomas and/or carcinomas of the squamous epithelium;

6) Hepatocellular adenoma and/or carcinoma;

7) Lung alveolar/bronchiolar adenomas and/or carcinomas;

8) Small intestine adenocarcinomas;

9) Hemangiosarcomas;

Conclusions:
Target tissues for carcinogenicity in mice exposed to benzo[a]pyrene via the diet were the forestomach, oesophagus and tongue. No clear treatment-related induction of liver tumours was found.
Executive summary:

The report includes (as an Annex) a summary of findings from mouse diet studies which were conducted at NCTR (Culp et al., 1994 and 1998). The NCTR project incorporates a chronic oral carcinogenicity study with B[a]P.

During the study, groups of 48 female 5-week old B6C3F1 mice were administered via the diet B[a]P (5, 25, and 100 ppm; calculated daily intake of 0.6, 3, 12 B[a]P in mg/kg bw, corrected for reductions in food consumption)

for 24 months. Two additional groups of 48 mice served as controls, i.e. one was fed the standard diet, the other received standard diet that had been treated with aceton in a manner identical to the B[a]P diets (Solvent control).

With respect to B[a]P, at the highest dose tested a decrease in body weight was observed from week 50 onwards, most probably due to effects secondary to tumour-development in the upper digestive tract. B[a]P induces in these female B6C3F1 mice tumours in the epithelial layer of the tongue, the esophagus, and the forestomach. Also, increased incidences of sarcomas and of tumours in the squamous cell of the larynx were observed. B[a]P treatment did not induce tumours in the liver and lung, nor hemangiosarcomas or histiocytic sarcomas.

In conclusion, the target tissues for carcinogenicity in mice exposed to benzo[a]pyrene via the diet were the forestomach, oesophagus and tongue. No clear treatment-related induction of liver tumours was found.

Endpoint:
carcinogenicity: oral
Type of information:
experimental study
Adequacy of study:
weight of evidence
Study period:
not reported
Reliability:
1 (reliable without restriction)
Rationale for reliability incl. deficiencies:
test procedure in accordance with generally accepted scientific standards and described in sufficient detail
Remarks:
Method and results were sufficiently described, similar to OECD-guideline 451. The stduy was conduced as a combined sub-chronic toxicity/carcinogenicity study. The study summary of 3-month repeated dose toxicity study is reported separately under section 7.7. The main deviations from the current test guidelines include a limited number of organs examined at the terminal investigations.
Reason / purpose for cross-reference:
reference to same study
Qualifier:
no guideline followed
Principles of method if other than guideline:
Groups of 52 animals per sex, per dose, received doses of test material, by oral gavage, for 5 days a week over a period of 104 weeks . The animals were observed daily and their behaviour and clinical signs were examined. Body weights, food and water consumption was measured at least weekly. Following the end of the exposure period the animals were sacrificed and were subjected to macroscopic examination, additionally tissue samples were processed for further histopathological investigation.
GLP compliance:
yes
Specific details on test material used for the study:
B[a]P was from Serva (Heidelberg, Germany) and had a purity of 98.6 ± 0.4%.
Species:
rat
Strain:
Wistar
Remarks:
SPF Riv:TOX
Sex:
male/female
Details on test animals or test system and environmental conditions:
TEST ANIMALS
- Source: the animals were bred at and derived from the Animal Facility at the Institute
- Age at study initiation: 4-5 weeks at acclimation initiation. Animals were 6 weeks of age at study initiation (following acclimation).
- Housing: animals were housed in macrolon cages with a wire floor, two per cage
- Diet: ad libitum (SSP-Tox, Hope Farms BV, Woerden). From the start of treatment the applied food contained a reduced amount of soy oil to compensate for the soy oil used to administer the test material (standard SSPTox food contains 4.55% soy oil on a weight basis).
- Water: ad libitum (tap water - public drinkingwater, WMN, Utrecht)
- Acclimation period: 10 days

ENVIRONMENTAL CONDITIONS
- Temperature (°C): 19 - 25 °C
- Humidity (%): 40 - 72 %
Route of administration:
oral: gavage
Vehicle:
soya oil
Details on exposure:
PREPARATION OF DOSING SOLUTIONS:
The test material was dissolved in soy oil under heating (up to 60°C) and ultrasonic vibration.
Fresh solutions were prepared every week (week 1 - 8), every 2 weeks (week 9-18), and every 3 weeks (19-104), respectively.
The test material appeared to be stable under conditions of dissolution and storage (once dissolved). The test material solutions were prepared as follows. First, mean group weights were determined (per sex, per dose). Then, the mean group weight was estimated at half the dosing-period by simple extrapolation on the group’s growth curve (this ‘extrapolated’ mean group weight was used to prepare the solutions; this resulted in some slight initial overdosing).

DETAILS ON ROUTE OF EXPOSURE:
The length of the gavage needle used assured exposure of the distal half of the oesophagus (a known target-site for carcinogenesis).
Application was always in the morning for all groups and took about 1.5 - 2.5 hours.
Analytical verification of doses or concentrations:
yes
Details on analytical verification of doses or concentrations:
Test material solutions in use were regularly verified for their benzo[a]pyrene content.
Duration of treatment / exposure:
104 weeks
Frequency of treatment:
5 days per week
(Rats were not dosed at some national or christian celebration days)
Dose / conc.:
3 mg/kg bw/day
Remarks:
Due to several corrections (e.g. purity) actually achieved dose levels were somewhat lower, i.e. 2.9 ± 0.3 mg/kg bw/day
Dose / conc.:
10 mg/kg bw/day
Remarks:
Due to several corrections (e.g. purity) actually achieved dose levels were somewhat lower, i.e. 9.6 ± 1.0 mg/kg bw/day
Dose / conc.:
30 mg/kg bw/day
Remarks:
Due to several corrections (e.g. purity) actually achieved dose levels were somewhat lower, i.e. 29 ± 3 mg/kg bw/day
No. of animals per sex per dose:
52 rats (per dose, per sex)
Control animals:
yes, concurrent vehicle
Details on study design:
- Dose selection rationale: doses were selected on the basis of findings from short-term, and sub-chronic repeated dose toxicity studies (please see relevant robust study summaries in section 7.5.1 for further information).
- Rationale for animal assignment: animals were ad random assigned to the various dose groups by standardised procedures
Observations and examinations performed and frequency:
CAGE SIDE OBSERVATIONS: Yes
- Time schedule: The animals were examined daily for behaviour and clinical symptoms and by palpation.

BODY WEIGHT: Yes
- Time schedule for examinations: every week (week 1 - 8), every 2 weeks (week 9-18), or every 3 weeks (week 19-104), respectively.

FOOD CONSUMPTION: Yes
- Time schedule for examinations: twice weekly during the first 18 weeks, and once every two weeks for the remaining period.

WATER CONSUMPTION: Yes
- Time schedule for examinations: twice weekly during the first 18 weeks, and once every two weeks for the remaining period.

OPHTHALMOSCOPIC EXAMINATION: No

HAEMATOLOGY: No

CLINICAL CHEMISTRY: No

URINALYSIS: No

NEUROBEHAVIOURAL EXAMINATION: No
Sacrifice and pathology:
TERMINAL PROCEDURES: All animals were subjected to complete necropsy after death, or intercurrent or terminal sacrifice. The rats were killed by exsanguination from the abdominal caval vein under ether narcosis. From week 32 onwards exsanguination was performed under inhalation anaesthesia with CO2 / O2.

GROSS PATHOLOGY: Yes

HISTOPATHOLOGY: Yes
After macroscopic inspection the following organs were collected for further microscopic analysis: brain, pituitary, heart, thyroid, submandibular salivary glands, lungs (including trachea and infused with fixative), stomach (mounted on paraffin wax disks after opening along curvature maior and flushing), oesophagus, duodenum, jejunum, ileum, caecum, colon, and rectum (swiss rolls), thymus, kidneys, urinary bladder (after instillation with fixative), spleen, mesenteric, axillary and mandibular lymph nodes, liver, pancreas, adrenals, sciatic nerve, biceps femoris muscle, femur (diaphysis), vertebral column and head (after flushing the nasal cavity), skin including mammary tissue (ventral abdominal strip adjacent to midline, swiss rolled), ovaries/uterus or testis/accessory sex glands.
In addition, all gross abnormalities, in particular masses and lesions suspected of tumourous nature were sampled. After fixation (and for vertebral column, femur and skull: after subsequent decalcification in formic acid 20% as adequate) samples were trimmed, processed, and paraffin wax embedded. From the head, transverse sections (5) of the nose were taken, as well as a transverse section at the level of the auditory canal. Sections (5 μm) were cut and stained routinely with haematoxylin and eosin (H&E). Fixation, paraffin wax embedding, sectioning, staining, and all other histotechnical procedures were performed according to the Standard Operating Procedures of the Laboratory of Pathology of the Institute.
Clinical signs:
not specified
Description (incidence and severity):
Condition and behaviour, with particular emphasis on animal welfare aspects, remained reasonably well.
Dermal irritation (if dermal study):
not examined
Mortality:
mortality observed, treatment-related
Description (incidence):
In animals exposed to the test material, treament of both males and females resulted in a dose-related decrease in survival. Most intercurrent mortalitites were from euthanasia due to poor condition, ususally related to tumours (mainly in liver and stomach), while significant non-neoplastic effects were not observed.
The mortality in the highest dosed groups was 100% after about 70 weeks. This mortality was mainly due to sacrifice for humane reasons when the rats became emaciated, with distended abdomen in which frequently one ore more palpable masses were present in the cranial area (liver). Despite the presence of palpable masses, the condition and behaviour, with particular emphasis on animal welfare aspects, remained reasonably well.
Liver tumours were the most predominant tumour type encountered in this study in terms of morbidity and mortality.
In the control animals, survival after 104 weeks was about 65% and 50% in males and females, respectively. In these animals by far the main cause of death was tumour development in the pituitary, which is consistent with earlier findings in our historical controls.
Body weight and weight changes:
effects observed, treatment-related
Description (incidence and severity):
Only at the highest dose tested (30 mg/kg bw), treatment with the test material appeared to affect body weight gain in males only, i.e. from week 10 onwards. At the same dose level females appear to be less sensitive for this effect. Accordingly, it was concluded that adequate top-dose levels that were originally chosen were adequate, i.e. around or just below the maximum tolerated dose (MTD).
Food consumption and compound intake (if feeding study):
effects observed, non-treatment-related
Description (incidence and severity):
Treatment with the test material apparently had no significant effect on food consumption. For some short periods within the 104 weeks of treatment small (i.e. < 10%) statistically significant differences between the various treatment groups and controls were observed, though without a clear relationship to dose. An exception appears to be the more sustained reduction of food consumption though again of less than 10% of high dosed males starting at about week 36.
Food efficiency:
not examined
Water consumption and compound intake (if drinking water study):
effects observed, treatment-related
Description (incidence and severity):
Treatment with the test material had no major effect on water consumption in females. Water consumption in males appeared to be dose-relatedly increased, starting from week 13 onwards. High-dose males showed an increased consumption of water (more than 50%) briefly before their withdrawal from the study. It was noted that high-dose males also showed a decreased food consumption, and a decreasing body weight during this period. There is not yet a satisfactory explanation for these observations.
Ophthalmological findings:
not examined
Haematological findings:
not examined
Clinical biochemistry findings:
not examined
Urinalysis findings:
not examined
Behaviour (functional findings):
not examined
Immunological findings:
not examined
Organ weight findings including organ / body weight ratios:
not examined
Gross pathological findings:
effects observed, treatment-related
Description (incidence and severity):
See histopathological findings: neoplastic.
Neuropathological findings:
not examined
Histopathological findings: non-neoplastic:
effects observed, treatment-related
Description (incidence and severity):
Histopathological examination showed only minor changes such as slight reactive (proliferative) changes in the mucosa of the forestomach.
Histopathological findings: neoplastic:
effects observed, treatment-related
Description (incidence and severity):
Treatment with the test material for 2 years resulted in a dose-dependent increase in tumour incidence in a variety of organs and/or tissues. Clearly the most prominent carcinogenic effects of the test material were observed in the liver and forestomach, both organs with a low spontaneous tumour incidence in this rat strain, as illustrated by the incidence found in the concurrent controls of the present study. At the two highest dose levels multiple tumours in the liver was a common finding, together with tumour development in the forestomach. Besides these two target sites, treatment with the test material also induced soft tissues sarcomas at various sites (oesophagus, skin, mammary), as well as tumours of the auditory canal, skin, and oral cavity. Tumours were also observed in the small intestine (at the top dose; especially males), and in the kidney (the two highest dose levels, males only).

IN-LIFE OBSERVATIONS:
Next to the palpable liver tumours, frequently tumours were seen on the base of the ear which eventually could result in ulceration or eruption through the auditory canal. Often these occurred bilateral. In the intermediate groups, subcutaneous tumours were often found in the neck area between pharynx and axillary region. When these tumours compromised feeding or mobility, or became ulcerated, the animal was necropsied and the tumour was qualified as fatal.

NECROPSY:
Gross changes in the highest dose group included masses in the liver (99/104), papilliferous thickening or masses in the forestomach (89/104) and nodules in the lungs suspect for metastatic spread. Also, frequently masses in the skin (back / head / neck / lips) or at the basis of the auditory canal were found. In a number of cases (predominantly in groups 3 and 4) typically masses were found in the neck / prescapular or mediastinal area. Ususally these tumours were closely associated with the esophagus and / or neck musculature. Occasionally masses were observed on other organs and tissues, such as kidney, small intestine, etc. Other gross findings occurred occasionally or in accordance with common age and strain related background pathology. Histopathological observations for some major tumours will be briefly presented below.

LIVER:
Masses in the liver were found in the intercurrently died rats, the first case observed in week 35 (male, highest dose). These were most often found in the middle and right lateral lobes, but in advanced cases multiple tumours were found in multiple lobes. Tumours were often pleomorphic: solid, cystic, necrotic areas, haemorrhagic areas. In advanced cases, adhesions were formed with surrounding organs and omentum, and intraabdominal metastatic spread and ascites could be present. Liver tumours were the most frequent cause of death (i.e. indication for euthanasia). In the lower dose group, a much lower incidence and manifestation at a later time point was observed.
Tumours of the liver were usually complex, multiple and malignant in nature. A large proportion had metastasised to the lungs (59 of 150) and some to the abdominal cavity. The growth pattern and cytology were highly variable, while strikingly metastases were cytomorphologically often well differentiated. Therefore, cytomorphology was not a powerful criterion in discriminating between benign and malignant tumours. Malignant liver tumours were often associated with cystic structures that could be either of hepatocellar or cholangiocellular origin. Since these cholangiocellular lesions usually did not occur separately, but within or associated with malignant liver tumours, they have not been monitored as a separate entity with a few exemptions where they were clearly separated.
Histologically only the most progressed lesion was scored, thereby encompassing overdiagnosing. Thus, since tumours were usually multiple, if carcinoma was scored, this implies the concomitant presence of adenomas and foci of cellular alteration. Scoring was quantified as one, few or multiple.
Foci of cellular alteration (mainly clear cell type) were scored in tumour free livers, and quantified as one, few or multiple.

FORESTOMACH:
Masses in the forestomach were seen in most animals (86%) from the high dose group, and in 65 and 14% of the lower dose groups. These appeared as isolated small protrusions with a papilliferous surface. The common site of origin was the apex of the forestomach, and in advanced cases multiple masses became confluent. In some cases (pre)perforative local peritonitis had occurred, occasionally with evidence of intra-abdominal spread. In a few cases, the forestomach tumours were considered to be fatal, as decided upon presence of peritonitis and / or metastatic spread.
Lesions in the forestomach consisted of (multi)focal hyperplasia of basal cells, followed in advanced cases by papillomas, which ultimately could have resulted in invasive squamous cell carcinomas. In a few instances metastatic spread into the abdominal cavity and organs or lungs had occurred.
Forestomach lesions could be monitored in a time-sequence fashion swing to the intercurrent sacrifices. Small and /or early lesions were characterised by focal or confluent multiple hyperplasia of basal cells. In more advanced lesions hyperplasia of the squamous component, including squamous cell papilloma, became apparent. Invasive growth usually occurred by the squamous component (squamous cell carcinoma), in some cases resulting in perforation, contact metastasis in the abdomen, and lung and liver metastases. Consequently, the diagnosis of squamous cell carcinoma implicitly signifies the presence of both papilloma and basal cell hyperplasia, and the diagnosis papilloma implicitly includes the presence of basal cell hyperplasia.
The glandular stomach or oesophageal mucosa were generally unaffected.

SOFT TISSUE:
Most sarcomas were found in the subcutis of neck, prescapular or axillary area, often involving the oesophagus, thyroid or neck muscles by invasion. For this reason most have been entered under “oesophagus” or “skin and mammary”. Occasionally this type of tumour was found at other locations (chest wall, groin, abdominal cavity (3x), stomach wall (2x), uterus / ovary (2x) and nasal cavity. These sarcomas typically revealed a variety of characteristics within the same mass, such as rhabdo-myosarcoma, fibrosarcoma, malignant fibrous histiocytoma, undifferentiated sarcoma, etc. When one feature predominated, this served as the basis of classification. In a proportion of these tumours transparant spaces were observed with material suggestive of oil. These spaces were covered by thin elongated fibroblast or macrophage like cells, but without florid inflammatory response.

AUDITORY CANAL:
The tumours were histologically composed of squamous or sebaceous cells with the range of intermediate types, although in the larger tumours the squamous component predominated. These lesions apparently arose from the pilo-sebaceous units (sebaceous glands, including the Zymbal gland) in the auditory canal. In some cases a normal Zymbal gland existed in association with a tumour. This lesion was usually not preceded by (diffuse) hyperplasia, but focal dysplasia, hyperplasia or cystic change occurred instead. Malignant tumours of the auditory canal tended to infiltrate the masticatory musculature, and occasionally the skull and brain. Metastasis to the lung was found once.

ORAL CAVITY:
Tumours at this site usually were derived from pilosebaceous tissue, normally present as clusters in the lip commissure. In many cases, including control animals, hyperplasia of the units was found, often with abscesses, dilation of hair shafts / glandular outlets. The incidence of these hyperplastic and inflammatory lesions appeared not to be treatment related. Sizeable tumours usually were of the squamous or mixed squamous / sebaceous cell type, and were often multiple / bilateral, and dose-related increased.

SKIN:
Skin tumours were basically characterised as arising from epidermis (squamous cell papilloma, carcinoma or keratoacanthoma) or appendages (basal, hair follicle or sebaceous cell types and mixtures thereof). For that reason the diagnostic differentiation was not always unequivocal and was based upon the most prominent or most malignant component. In fact it may be realistic to pool these varieties of skin / appendage. These tumours were often multiple and variable, therefore the total number of tumours may seem overestimated when related to number of tumour bearing animals.

KIDNEY:
Tumours of the kidney mainly included cortical adenoma and occurred in the two highest dose group males. These benign tumours were usually small and could only be diagnosed incidentally by careful histopathology. A clear distinction with hyperplasia is not always feasible, even with the commonly accepted criteria (IARC / SNTP), since there is a gradual continuum between hyperplasia and adenoma. Therefore the lesions might be merged. Since the observation of these lesions is particularly chance-dependent, this lesion is under this protocol (one section of either kidney) most probably underscored. Occasionally adenocarcinoma, and tumours of pelvis or ureter were found.

SMALL INTESTINE:
Tumours of the small intestine were observed mainly in the top dose males and were sometimes multiple. Malignant tumours occasionally metastasised to the liver and were often associated with obstruction and / or peritonitis.
Description (incidence and severity):
See the table.
Relevance of carcinogenic effects / potential:
The B[a]P treatment, lasting for two years, resulted in a dose-dependent increase in tumour incidence in a wide spectrum of organs and tissues, the most prominent clearly being the liver and rumen, both being organs with a low spontaneous tumour incidence in this strain. Liver tumours were also responsible for the high mortality rate at the highest dose level in both sexes.The results show that only the top dose in males may possibly be regarded as exceeding the MTD, and therefore, the carcinogenic effects observed in the other dose groups are considered to be relevant for extrapolation to humans. A number of oesophagus- associated tumours were observed in the present study, but these were sarcomas instead of carcinomas, and attributed to accidental local tissue deposition of gavaged material (injection site sarcomas). The incorporated DNA adduct test showed no evidence of the possibility that either the total number (i.e. intensity) of adducts or the presence of specific adducts are responsible for tumourigenesis.
Dose descriptor:
LOAEL
Effect level:
3 mg/kg bw/day
Based on:
test mat.
Sex:
male/female
Basis for effect level:
body weight and weight gain
histopathology: neoplastic
mortality
Critical effects observed:
yes
Lowest effective dose / conc.:
10 mg/kg bw/day (nominal)
System:
other: Gastrointestinal tract, hepatobiliary
Organ:
liver
stomach
Treatment related:
yes
Dose response relationship:
yes
Relevant for humans:
yes

Table. Incidences of some major treatment-related neoplasms in female rats treated with B[a]P. The most advanced stage of lesions is scored.

    Females Males
Site Dose (mg/kg bw/kg) 0 3 10 30#1 0 3 10 30#1
forestomach  Examined 52 51 51 52 52 52 52 52
  squamous cell papilloma 1 3 20*** 25*** 0 7* 18*** 17***
  squamous cell carcinoma 0 3 10** 25*** 0 1 25*** 35***
Liver Examined 52 52 52 52 52 52 52 52
  hepatocellular adenoma 0 2 7* 1 0 3 15*** 4
  hepatocellular carcinoma 0 0 32*** 50*** 0 1 23*** 45***
auditory canal Examined#2 0 1 0 20 1 0 7 33
  squamous cell papilloma 0 0 0 1 0 0 0 4
  carcinoma#3 0 0 0 13** 0 0 2 19***

#1 note that this group had a significantly shorter lifetime;

#2: these tissues were examined only when abnormalities were observed upon macroscopic examination;

#3: composite tumours of squamous and sebaceous cells apparently arisen from the pilosebaceous units / "Zymbal glands".

*p<0.01, **p<0.001 ***p<0.00001,

Conclusions:
This study demonstrated the carcinogenic potential of the test material in a variety of organs when administered by gavage to rats. Major target organs were liver, forestomach and epidermal structures, of which the liver is considered the most relevant for human risk assessment in terms of pathogenesis and sensitivity.
Executive summary:

A study was conducted to provide an estimation of the cancer risks associated with the actually experienced daily exposure to dietary PAH in the Netherlands. Benzo[a]pyrene was considered to be a representative indicator of PAH. This carcinogenicity study in rats orally exposed to benzo[a]pyrene, was considered to adequately represent the carcinogenic fraction of PAH.

During the study, groups of 52 animals per sex, per dose, received doses of test material, by oral gavage, for 5 days a week over a period of 104 weeks . The animals were observed daily and their behaviour and clinical signs were examined. Body weights, food and water consumption was measured at least weekly. Following the end of the exposure period the animals were sacrificed and were subjected to macroscopic examination, additionally tissue samples were processed for further histopathological investigation.

Findings from the study clearly showed benzo[a]pyrene to be a potent carcinogen upon chronic oral administration. Tumours were induced at multiple sites in both sexes of rats, i.e. liver, forestomach, auditory canal, oral cavity, skin, and intestines, and additionally the kidney in males, and the mammary and oesophagus in females. The most potent carcinogenic effects of benzo[a]pyrene under the present conditions were observed in the liver and forestomach, both organs with a low spontaneous incidence in this rat strain. The lowest dose level associated with a significantly inceased tumour response is 10 mg/kg bw/day for females and 3 mg/kg bw/day for males.

Justification for classification or non-classification

The substance is classified as a carcinogen (category 1B) in consideration of the presence of polyaromatic hydrocarbons (PAHs) analysed in the substance at a level greater than the generic cut-off limits for classification for carcinogenicity. PAHs are widely acknowledged to possess CMR properties. The harmonised classification of the PAH substance benzo[a]pyrene is applied to this substance as it has been established that this substance is considered to be representative of the PAHs contained within the registered substance. Furthermore, the hazard of benzo[a]pyrene is well established in light of the wealth of toxicological data that are available on the substance.


Please refer to the document ‘Consideration of the long term toxicity of Distillates (shale oil)’, as included in section 13, for further information.

Additional information

According to column 2 of REACH Annex X, information requirement 8.9.1, a carcinogenicity study does not need to be conducted if the substance is classified as a germ cell mutagen category 1A or 1B. Since the registereed substance is reasonably known to be mutagenic category 1B, it is considered justified to omit testing for this endpoint. Additionally, the substance is already classified as carcinogens category 1B. Further testing for carcinogenicity is neither considered necessary nor considered to be an expedient or responsible use of animals.


Please refer to the document ‘Consideration of the long term toxicity of Distillates (shale oil)’, as included in section 13, for further information.


 


Supporting data (as considered in a weight of evidence approach)


 


RIVM (Rat Study on benzo[a]pyrene)


A study was conducted to provide an estimation of the cancer risks associated with the actually experienced daily exposure to dietary PAH in the Netherlands. Benzo[a]pyrene was considered to be a representative indicator of PAH. This carcinogenicity study in rats orally exposed to benzo[a]pyrene, was considered to adequately represent the carcinogenic fraction of PAH.


During the study, groups of 52 animals per sex, per dose, received doses of test material, by oral gavage, for 5 days a week over a period of 104 weeks . The animals were observed daily and their behaviour and clinical signs were examined. Body weights, food and water consumption was measured at least weekly. Following the end of the exposure period the animals were sacrificed and were subjected to macroscopic examination, additionally tissue samples were processed for further histopathological investigation.


Findings from the study clearly showed benzo[a]pyrene to be a potent carcinogen upon chronic oral administration. Tumours were induced at multiple sites in both sexes of rats, i.e. liver, forestomach, auditory canal, oral cavity, skin, and intestines, and additionally the kidney in males, and the mammary and oesophagus in females. The most potent carcinogenic effects of benzo[a]pyrene under the present conditions were observed in the liver and forestomach, both organs with a low spontaneous incidence in this rat strain. The lowest dose level associated with a significantly inceased tumour response is 10 mg/kg bw/day for females and 3 mg/kg bw/day for males.


 


RIVM (Mouse study on benzo[a]pyrene)


The report includes (as an Annex) a summary of findings from mouse diet studies which were conducted at NCTR (Culp et al., 1994 and 1998). The NCTR project incorporates a chronic oral carcinogenicity study with B[a]P.


During the study, groups of 48 female 5-week old B6C3F1 mice were administered via the diet B[a]P (5, 25, and 100 ppm; calculated daily intake of 0.6, 3, 12 B[a]P in mg/kg bw, corrected for reductions in food consumption)


for 24 months. Two additional groups of 48 mice served as controls, i.e. one was fed the standard diet, the other received standard diet that had been treated with aceton in a manner identical to the B[a]P diets (Solvent control).


With respect to B[a]P, at the highest dose tested a decrease in body weight was observed from week 50 onwards, most probably due to effects secondary to tumour-development in the upper digestive tract. B[a]P induces in these female B6C3F1 mice tumours in the epithelial layer of the tongue, the esophagus, and the forestomach. Also, increased incidences of sarcomas and of tumours in the squamous cell of the larynx were observed. B[a]P treatment did not induce tumours in the liver and lung, nor hemangiosarcomas or histiocytic sarcomas.


In conclusion, the target tissues for carcinogenicity in mice exposed to benzo[a]pyrene via the diet were the forestomach, oesophagus and tongue. No clear treatment-related induction of liver tumours was found.


 


IARC Monograph of high-temperature crude shale oils, low-temperature crude shale-oils, fractions of high-temperature shale-oil, crude shale-oil distillation fractions, shale-oil bitumens and commercial blends of shale-oils


There is sufficient evidence for the carcinogenicity in experimental animals of high-temperature crude shale oils, low-temperature crude shale-oils, fractions of high-temperature shale-oil, crude shale-oil distillation fractions, shale-oil bitumens and commercial blends of shale-oils.


There is limited evidence for the carcinogenicity in experimental animals of raw oil shale, spent oil shale and a residue of shale-oil distillation.


There is sufficient evidence that shale-oils are carcinogenic in humans.