Registration Dossier

Administrative data

Description of key information

A number of studies have investigated the repeated dose oral toxicity of C14-17 chlorinated paraffins (40% or 52% chlorination) in rodents. These allow identification of an overall NOAEL of 23 mg/kg bw/day, from a 90-day dietary study in rats. Increased kidney and liver weights, liver enzyme induction, liver hypertrophy and slight decreases in plasma triglycerides and cholesterol levels were seen at the next higher dose of 222 mg/kg bw/day. At even higher doses in other 90-day rat dietary studies, additional kidney effects (‘chronic nephritis’ and tubular pigmentation) were reported as well as effects on the liver and thyroid. The liver effects included enzyme induction, increased weight and hypertrophy (all considered to be adaptive responses rather than toxicological in nature) and signs of peroxisome proliferation (of no relevance to man). The thyroid effects were concluded to be secondary to changes in liver enzyme activity and of no real significance for humans. In one-generation reproduction studies, NOAELs of 47 mg/kg bw/day and 100 mg/kg bw/day (as maternal doses) have been identified for the adverse effects on blood clotting seen in the offspring mediated via lactation and for the haemorrhaging potential in dams at parturition, respectively.

Key value for chemical safety assessment

Repeated dose toxicity: via oral route - systemic effects

Endpoint conclusion
Dose descriptor:
NOAEL
23 mg/kg bw/day
Study duration:
subchronic
Species:
rat

Additional information

In a good quality study (Elcombe, 2005b), Cereclor S52 (a C14-17 chlorinated paraffin; 52% chlorinated) was administered in the diet to Fischer 344 rats (10/sex/dose) for 90 days at concentrations of 0, 30, 100, 300 or 3000 ppm (about 0, 2.4, 9.3, 23 or 222 mg/kg bw/day). There were no treatment-related effects on growth, terminal body weight, food consumption, or histopathology in the kidneys or thyroid at any dose. However, at 222 mg/kg bw/day, minimal centrilobular hepatocyte hypertrophy was noted in the liver of male rats, liver and kidney weights were increased, and effects on clinical chemistry (slight decreases in plasma triglycerides and cholesterol levels) were seen. Slight effects on the thyroid hormones at 23 mg/kg bw/day and above were seen in males and females. The liver effects are considered to be an adaptive response of no toxicological significance, and the slight, inconsistent thyroid hormone changes as secondary effects related to increased liver metabolic activity and – in the absence of any histopathological changes – not adverse. Therefore, based on the increased kidney weights at the top dose (and slight decreases in plasma triglycerides and cholesterol) a NOAEL of 23 mg/kg bw/day was established.

 

Groups of Fischer 344 rats (15/sex/dose) were administered Cereclor S52 in the diet for 90 days at levels corresponding to doses of 0, 10, 100 or 625 mg/kg bw/day (Spicer, 1984). Survival, overt toxicity, food consumption, growth, haematology, blood biochemistry, urine composition, organ weights, ophthalmoscopy and the microscopic appearance of the tissues were monitored. There were no effects at 10 mg/kg bw/day. At 100 mg/kg bw/day, liver weights were increased. At 625 mg/kg bw/day, growth and food consumption were reduced and there were effects on the liver (increased weights and "trace" hepatocellular hypertrophy in most of the animals) and kidney (increases in blood urea nitrogen and urinary protein, as well as renal tubule pigmentation and a ‘mild chronic nephritis’ in the top-dose females and males respectively). In addition, the thyroid was affected in males (increased absolute weight and a greater severity of hypertrophy and hyperplasia). The effects seen in the liver and thyroid are considered to be of no relevance to human health. The study NOAEL is 100 mg/kg bw/day, with toxicologically significant kidney effects observed at the highest dose level studied (625 mg/kg bw/day).

 

NOAELs of 32 mg/kg bw/day in rats and 30 mg/kg bw/day in dogs have been observed following 90-day oral exposure (Conning et al. 1971a,b; reviewed by Birtley et al. 1980). [The draft RAR on MCCPs summarized another 90-day rat dietary study (Poon et al. 1995), but noted doubts over the validity of the histopathology scoring system. The reports of kidney and thyroid effects at the low dose of 4 mg/kg bw/day suggested that the study was unrepresentative of other MCCP studies and therefore should not be used for risk characterisation purposes (EU, 2008).]

 

In two reliable one-generation studies (Stamp [CXR], 2006; IRDC, 1985; see IUCLID Chapter 7.8.1 for more details) male and female rats were administered Cereclor S52 in the diet at up to approximately 100 and 400 mg/kg bw/day (the highest tested doses, respectively) for 4 weeks prior to mating, and throughout mating and (in females only) during gestation. In the earlier study, pup survival in all dose groups at birth was equivalent to that of the control group pups. However, reduced pup survival (by 11%) and related haemorrhaging was evident during the lactational period at the lowest-observed-adverse-effect level (LOAEL) of 74 mg/kg bw/day (maternal dose) and, although not statistically significant, is considered to be of toxicological importance (IRDC, 1985). No such effects were seen in the more recent reproduction study with exposure for 11-12 weeks at maternal dose levels of up to about 100 mg/kg bw/day (CXR, 2006). Overall, a NOAEL of 47 mg/kg bw/day (as a maternal dose) can be identified for the adverse effects seen in the neonatal offspring mediated via lactation. In another one-generation study in rats (CXR, 2004), haemorrhaging was seen at parturition in 16% of dams given 538 mg/kg bw/day, but not at up to 100 mg/kg bw/day in other studies. An overall NOAEL of 100 mg/kg bw/day (as a maternal dose) can be identified for the haemorrhaging potential in pregnant women at parturition. The reported internal haemorrhaging would appear to be a repeated dose effect to which newborn rats during lactation, and possibly pregnant female rats at the time of parturition, are particularly susceptible.

 

Overall, a NOAEL of 23 mg/kg bw/day was identified for repeated dose toxicity.

 

No information is available on the effects of repeated exposure to MCCPs in humans. There are no data in laboratory animals relating to repeated inhalation or dermal exposure. In accordance with column 2 of REACH Annex VIII and IX, repeated dose toxicity testing by the inhalation and dermal routes is not considered appropriate due to the low vapour pressure and lack of significant absorption through the skin (Johnson, 2005).

 

Mechanisms of toxicity in the liver, thyroid and kidneys

A number of studies have been conducted to investigate the possible underlying mechanistic events leading to the observed spectrum of toxicity in animals, with the aim of assessing their relevance to humans. In general, studies in rats have looked at the underlying mechanisms of effects in the liver, thyroid and kidneys (Elcombe et al. 1997; Poon et al. 1995; Wyatt et al. 1993, 1997), whilst studies in mice and guinea pigs have focused mainly on the liver (Elcombe et al. 1997; Meijer et al. 1981; Meijer and DePierre, 1987; Wyatt et al. 1993).

 

The findings of these mechanistic studies demonstrate that MCCPs are capable of eliciting hepatic enzyme induction and proliferation of smooth endoplasmic reticulum in response to an increased metabolic demand arising from MCCP exposure. These effects are considered to reflect physiological adaptation rather than a toxicological response (EU, 2008). At higher dose levels, there is evidence (based on microscopy, morphometric analysis and enzyme marker activity) that MCCPs can induce hepatic peroxisome proliferation (PP) in rats and mice, but not in guinea pigs, a species that is relatively insensitive to this effect (they did show a small elevation in beta-oxidation activity, but much less than in rats or mice). Humans are considered to be insensitive to PP induction (Ashby et al. 1994; Bentley et al. 1993). Thus, the liver changes seen in rats and mice are considered to be of limited relevance to human health (EU, 2008). Similar conclusions in relation to hepatic effects (i.e. the significance of changes related to xenobiotic metabolism and PP) were agreed in the final RAR for SCCPs (EU, 2000).

 

Exposure of rats to MCCPs (40% chlorination) by oral gavage for 90 days at 312 and 625 mg/kg bw/day (Wyatt et al. 1997) and 1000 mg/kg bw/day (the highest tested dose) for 14 days (Wyatt et al. 1993) resulted in thyroid effects (follicular cell hypertrophy and hyperplasia). No toxicologically significant effects on thyroid hormones and TSH levels were observed up to the top dose of 222 mg/kg bw/day in male rats in a recent, well-conducted 90-day study (Elcombe, 2005b). There is evidence (Wyatt et al. 1993) that the MCCP-induced increase in liver enzyme UDGP-transferase results in increased glucuronidation and consequent excretion of T4, leading to reduced T4plasma levels, followed by a pituitary response releasing more TSH, leading to increased thyroid production of T4, and the continuous thyroid stimulation ultimately results in hypertrophy and hyperplasia. In humans, the decreases in T4 (or T3) levels produced by altered hepatic clearance are typically insufficient to increase TSH levels. The decreased sensitivity of the human thyroid-pituitary axis to increased hepatic clearance of thyroxin appears to be influenced by several important quantitative differences between rats and humans (e.g. in humans, the half-life of T4 is much longer – probably due to higher-affinity binding to globulin in man -, the thyroid is less active, and TSH levels are much lower). This is discussed in detail in the draft RAR where it is concluded that humans are much less susceptible than rodents to fluctuations in levels of free plasma T4 and hence to any thyroid stimulation arising from a reduction in free T4 levels (EU, 2008). Similar effects on thyroid activity were observed for SCCPs (EU, 2000). Overall, considering the probable mechanisms outlined above, and the apparent association with the observed liver effects, it is concluded that the thyroid effects produced by MCCPs in rats would be of little relevance to human health at relevant levels of exposure.

 

From the data that are available, no adverse kidney effects were seen in male and female rats fed diets containing about 23 mg/kg bw/day for 90 days (Elcombe, 2005b). Changes seen in the kidneys, including increased kidney weight at 222 mg/kg bw/day (Elcombe, 2005b), and ‘chronic nephritis’ and tubular pigmentation at 625 mg/kg bw/day (Spicer, 1984) are considered as being potentially relevant to human health. Mechanistic studies indicated some deposition of alpha2u-globulin in proximal convoluted tubules of male rats only at higher dose levels. However, as this was unrelated to the pathological findings described above, this was not considered to be a classic male rat‑specific phenomenon.

 

Mechanisms of the internal haemorrhages

A study was conducted to investigate the possible mechanism of internal haemorrhages developing post-natally in pups (Hart et al. 1985; see IUCLID Chapter 7.8.3 for details). Groups of male and female Wistar rats were treated with 0 or 6250 ppm Cereclor S52 in the diet for 4 weeks before mating. After confirmation of mating, females were placed into one of five treatment groups (Group 1 - 16 females fed control diet rearing their own pups; Group 2 - 26 females fed MCCP diet rearing pups fostered from group 3 control females; Group 3 - 26 females fed control diet rearing pups fostered from group 2 treated females; Group 4 - 16 females fed MCCP diet rearing their own pups; and Group 5 - 16 females fed MCCP diet up to day 10 -17 of pregnancy, rearing their own pups while fed control diet. Blood samples were obtained from one pup/litter on days 3, 4, 5, 8, and 11, and 2 pups/litter on day 22 (the day of sacrifice) post-partum and analysed for clotting factors VIII and X. Prothrombin times were also measured at these timepoints and platelet counts on days 11 and 22. Samples of breast milk were taken from lactating dams of groups 1, 2 and 4 on day 14 post‑partum only, and analysed for MCCP. Sampling proved difficult and only one sample was obtained for each of groups 2 and 4; these were found to contain 570 and 1280 mg/l MCCP respectively. No overt toxicity was seen in the parental animals during the pre-mating period and pregnancy. During days 12-22 post-partum, a statistically significant increase in pup mortality was seen in control pups fostered to treated mothers (group 2) and treated pups reared by their own treated mothers (group 4); (77% and 67% deaths respectively compared to 4% for group 1). Increased mortality was not observed in the other groups (3 or 5). Haemorrhages were seen in 17 and 8%, respectively, of group 2 and group 4 offspring, with no sign of haemorrhaging in pups born to the other dams. Several findings indicative of internal haemorrhages were found in pups raised by dams treated with Cereclor S52 during lactation (i.e. groups 2 and 4), comprising dark red bulging eyes, blood clots within the membranes lining the cranium, and pale liver. In group 2, a significant reduction in pup body weight was observed from day 5 post-partum onwards (up to 11% on day 22). Throughout lactation, haematological analysis revealed a marked and statistically significant reduction in the concentration of clotting factor X in control pups fostered to treated mothers (i.e. group 2 - reduced by up to 45%) and pups reared by their own treated mothers (i.e. group 4 - reduced by up to 63%) relative to group 1 pups. The concentration of factor X in groups 3 and 5 was essentially similar to control group 1. Prothrombin times were increased (but not statistically significantly) in groups 2 and 4. No toxicologically significant changes were seen in factor VIII. Significant increases in liver weight were seen in group 2 pups (both sexes, up to a 5% increase) and in group 4 female pups (up to an 18% increase). The low (similar to control) incidences of pup death in groups 3 and 5, and the high incidences in groups 2 and 4, clearly indicate that the effect on the pups is focused on mother-to-pup transfer during weaning. The absence of effects in group 5 pups (where dams received MCCP until day about 10-17 of gestation) compared with the effects in group 2 and 4 pups (where there was continued feeding with MCCP throughout the study) suggests that continuous MCCP availability is needed for the effect to be mediated. The lack of effect in group 5 suggests that uptake of MCCP into fatty tissue (with subsequent mobilisation into breast milk) may not be a significant factor in this case. However, in the absence of measurement of MCCP levels in breast milk from group 5 dams, it is not possible to determine whether the lack of toxicity in the group 5 pups was due to no MCCP being present in the breast milk or whether the levels of MCCP (including any derived from fatty tissue) were below a threshold for this effect, however mediated (Hart et al. 1985). On the basis of the observed decreases in clotting factor X, the study authors proposed that the chlorinated paraffin tested may have been transferred in the breast milk (disrupting the pups’ clotting system) and/or that the pups received less vitamin K in the breast milk due to treatment-related effects upon their mothers (as a consequence the vitamin K-dependent clotting pathway was impaired). Both mechanisms could have led to an MCCP-induced reduction in haemostatic capacity in the pups, resulting in pup deaths. This conclusion would appear to be reasonable. Overall, therefore, MCCPs are considered to present a hazard to the neonatal offspring via the lactating mother. A NOAEL of 47 mg/kg bw/day as a maternal dose can be identified for this effect, from the previously described one-generation study (CXR, 2006).

 

An expert evaluation (EU, 2008) of a good-quality study (Powrie et al. 2003; summarised in IUCLID Chapter 7.9.3) concluded that although the administration of a C14-17 chlorinated paraffin (52% chlorination) to adult female rats at dose levels up to 1000 mg/kg bw/day caused a significant reduction in clotting factors, this was not of a sufficient magnitude to cause a biologically significant increase in prothrombin clotting times (EU, 2008). Overall, the draft RAR concluded that MCCPs are without effect on the blood clotting system in adult female rats treated for 3 weeks up to a dose level of 1000 mg/kg bw/day, suggesting that the haemorrhaging effects on the offspring are unlikely to be mediated by reduced vitamin K levels in breast milk under the conditions of this study (EU, 2008).

 

The other hypothesis (that MCCPs transferred to the pups through breast milk can disrupt the pups’ clotting system) was explored (CXR, 2004 [Barton and Daley, 2004]; see IUCLID Chapter 7.8.3 for details) in a one-generation assay modified to provide milk, blood and liver samples from lactating dams, and blood and liver samples from suckling pups. The blood and milk samples were analysed for levels of MCCPs, clotting factors and vitamin K, while the liver samples were examined for induction of the liver isozymes CYP2B1 and CYP2B2. Groups of 16 male and 32 female Sprague-Dawley rats representing the parental generation (F0) were treated in the diet with 0 or 6250 ppm Cereclor S52 (equivalent to a dose averaged over the first 4 weeks of treatment of 0 or 513 and 538 mg/kg bw/day for males and females, respectively) for 4 weeks prior to mating, then throughout mating, gestation and lactation. However, because of very high pup mortality in the treated group, the study was terminated prematurely approximately 2 weeks after the first litters were born (PND 12). The F0 animals were monitored for clinical signs of toxicity, body weight, food consumption and mating performance. The F1 offspring were monitored for survival and growth. Half of the dams from each group were assigned for pup sampling and half for milk sampling. Blood and liver samples (pup sampling) were obtained from one male and one female pup from half of the litters on days 1 and 4 of lactation (day 0 was the day of parturition) and at study termination (day 12 of lactation). From the remaining females not used for pup sampling, a milk sample was obtained on PND 1 and 4 and at study termination when samples of maternal liver were also collected. Five test dams (16%) died or were killed around the time of parturition. All five deaths were associated with difficulties in littering (although there was no obstruction or other hindrance to the delivery process) and were considered treatment-related. Of the five dams, four either gave birth to normal litters or were found to have a normal complement of live foetuses in their uterus. One exposed male was also found dead during the experiment. The clinical/necropsy findings in three out of these five dams and in the male rat found dead showed signs (abnormal red coloured urine, cage stained red, blood around vagina, placenta dark red, skin stained, eyes pale, skin pale) suggestive of haemorrhaging. Growth prior to mating was similar in both groups, but treated females showed slightly reduced weight gain during gestation and lactation. There were slight (about 10%) reductions in food consumption prior to mating in both treated sexes, and in treated females (17% reduction) during lactation. Cereclor S52 did not affect mating performance, duration of gestation, litter size at birth or pup mortality from birth to PND 4. However, after PND 4, pup mortality increased dramatically in the treated group such that few pups survived until the study was terminated around PND 12. At necropsy, the majority of these pups showed internal haemorrhages. In the treated group, mean pup weight on PND 1 was marginally lower (by 7-18%) but by PND 4 was noticeably lower (by 12-27%), and the difference was even more apparent (by 44-48%) by PND 7. At birth, liver weights were only marginally greater (by 7%) in pups from treated females compared to controls but the difference (up to 29%) was statistically significant by PND 4. MCCP analysis of dam milk on PND 1 found 1057 mg/L (no MCCPs were detected in control milk). Plasma vitamin K levels measured in samples from 10 animals from each group on PND 12 were significantly decreased in the treated dams (0.03 ng/ml) compared to the control dams (0.41 ng/ml). This finding seems to contradict the result obtained in the previous investigation (CXR, 2003) in which the plasma levels of vitamin K were unaffected by treatment. However, this apparent inconsistency could be explained by the fact that in the Powrie et al. (2003) [CXR, 2003] study, treatment with MCCPs lasted only three weeks whereas in this investigation (CXR, 2004) it lasted 7-8 weeks for the females. Furthermore, the 2003 study involved only treatment of adult nulliparous females whereas the 2004 study involved treatment of the females throughout the critical stages of pregnancy and lactation. Decreased maternal plasma vitamin K in the treated dams was reflected by decreased vitamin K levels in treated dam milk. Vitamin K was not found in pooled days 1 and 4 samples from four treated dams compared to a mean level of 0.28 ng/ml in samples from five control dams. This finding was confirmed on pooled days 9 and 12 samples from treated dams (0.36 ng/ml) and control dams (0.61 ng/ml) which showed an approximate 40% lower level of vitamin K in breast milk in the treated dams. As described above, the concentration of vitamin K in the plasma of adult females having gone through pregnancy and lactation was markedly decreased by Cereclor S52 treatment. This in turn produced a decrease in activity of the plasma clotting factors VII (24.2) and X (87.0) in the treated dams compared to controls (58.6 and 119 for factor VII and factor X respectively) on PND 12. However, this did not affect the prothrombin times in the dams, suggesting that the functional reserve in these adult animals was sufficient. Pup plasma volumes were insufficient to measure vitamin K directly but analysis of clotting factor activities (as surrogates) showed that Cereclor S52 treatment led to decreased clotting factor VII and X activities after PND 4.

 

The CXR (2004) data suggest that Cereclor S52 induced a perturbation of the clotting system in neonates of mothers treated at a dose level of about 540 mg/kg bw/day during the lactational period. In adult animals, decreased levels of vitamin K and of the clotting factors VII and X were found but prothrombin time was unaffected, indicating that the functional reserve in adults is sufficient. The foetus in utero apparently receives sufficient vitamin K via the placenta, but after birth becomes severely deficient in vitamin K and related clotting factors when reliant on the mother’s milk for these factors. Pups may also receive through the milk considerable levels of MCCPs which may also further reduce their vitamin K levels. This in turn will lead to a severe vitamin K deficiency in these neonates (who are already compromised in their vitamin K status) and consequently to haemorrhaging.

More recently, two studies have investigated the clearance and pharmacokinetics of a single dose of radio-labelled vitamin K, administered either by intraperitoneal injection (Powrie, 2010a) or gavage (Powrie, 2010b), to groups of 4 female rats pretreated with 14 daily gavage doses of Cereclor S52 at 1 g/kg bw (in corn oil). These studies show that pre-treatment with MCCPs leads to reduced blood and tissue reservoir levels of orally administered vitamin K (compared to vehicle controls), but not following injection, suggesting that MCCPs inhibit the intestinal absorption of vitamin K (Powrie, 2010a,b; see IUCLID Chapter 7.1.1 for further details).

From the studies available, an overall NOAEL of 47 mg/kg bw/day as a maternal dose can be identified for those effects mediated via lactation, and is selected for the risk characterisation of an adult population of breastfeeding mothers. However, it should be noted that the effects (an 11% reduction in pup survival and related haemorrhaging) observed at the LOAEL (74 mg/kg bw/day) were not statistically significant. Haemorrhaging was also seen in one study at the time of parturition in 16% of dams given Cereclor S52 at about 540 mg/kg bw/day, but not up to 100 mg/kg bw/day in other studies. The NOAEL of 100 mg/kg bw/day is appropriate for the risk characterisation of haemorrhaging effects potentially occurring in pregnant women at the time of parturition.

 

 

 

References (for studies for which no ESR has been created - move to reference list of CSR):

Ashby J et al. (1994). Mechanistically-based human hazard assessment of peroxisome proliferator-induced hepatocarcinogenesis. Human Experimental Toxicol. 13 Suppl 2 (cited in EU, 2008).

 

Bentley P et al. (1993). Hepatic peroxisome proliferation in rodents and its significance for humans. Food Chem. Toxicol. 31, 857-907 (cited in EU, 2008).

Elcombe C et al. (1997). Hepatic effects of chlorinated paraffins in mice, rats and guinea pigs: Species differences and implications for hepatocarcinogenicity. Unpublished report (cited in EU, 2008).

 

Meijer J et al. (1981). Effects of chlorinated paraffins on some drug-metabolising enzymes in rat liver and in the Ames test. Adv. Exp. Med. Biol. A136: 821-828 (cited in EU, 2008).

 

Meijer J and DePierre J (1987). Hepatic levels of cytosolic, microsomal and ‘mitochondrial’ expoxide hydrolases and other drug-metabolising enzymes after treatment of mice with various xenobiotics and endogenous compounds. Chem. Biol. Interactions. 62: 249-269 (cited in EU, 2008).

 

Wyatt I et al. (1993). The effect of chlorinated paraffins on hepatic enzymes and thyroid hormones. Toxicology. 77, 81-90 (cited in EU, 2008).

 

Wyatt I et al. (1997). Chlorinated Paraffins: Mechanisms of non-genotoxic carcinogenesis. Draft submitted to Arch. Toxicol. (cited in EU, 2008).


Repeated dose toxicity: via oral route - systemic effects (target organ) urogenital: kidneys

Justification for classification or non-classification

Under the EU CLP and DSD regulations, C14 -17 chlorinated paraffins would not be classified as a specific target organ toxicant as the observed kidney affects (considered of relevance to humans) in the 90-day oral rat studies were seen at doses significantly higher than 100 mg/kg bw/day.