Styrene

Endpoint summary

Key value for chemical safety assessment

Effects on fertility

Additional information

A well-conducted two-generation inhalation study (Cruzan et al., 2005a) found no effects on fertility and reproductive performance in rats exposed to up to 500 ppm (2165 mg/m3, ca.300 mg/kg/day) styrene, a concentration causing parental toxicity (degeneration of the olfactory epithelium and reductions in body weights). In more detail, reproductive performance and fertility were unaffected by styrene exposure in the parental generations (F₀& F₁). Degeneration of the olfactory epithelium of the nasal cavity was observed in the high-exposure group of the F₀and F₁parental animals. Also, body weights were statistically significantly reduced during the pre-mating interval in the mid-exposure males of the F₁generation (by 6-7%) and in the high-exposure males and females of both the F₀and F₁generations (by 6-13%). Body weights of females during gestation were statistically significantly reduced (by 6-7%) only at 500 ppm in the F₁generation. There were no significant maternal effects on body weight at 150 ppm, although body weights of the 150 ppm F₁females were reduced during gestation by 5-6%. In conclusion, from the results of this study, 50 ppm can be identified as the NOAEC for parental toxicity based on body weight reductions at 150 and 500 ppm and on the degeneration of the nasal olfactory epithelium at 500 ppm. Based on body weight reductions and on the degeneration of the nasal olfactory epithelium at 500 ppm, the NOAEC for maternal toxicity is 150 ppm. No effects on reproductive performance and fertility were seen up to the highest tested concentration of 500 ppm.

From the other relevant studies available, there is no convincing evidence that styrene can impair reproductive performance, produce testicular toxicity, sperm abnormalities or adversely affect the reproductive organs (Cruzan et al., 1997; 1998; 2001; NCI 1979; Quast 1979). Thus, taken together, the data available indicate that styrene does not have the potential to impair fertility and reproductive performance in animals. 

 

A literature search from 1998 to January 2010 overlapping the date of the UK RAR (June 2008) was conducted. Overall, the studies identified did not lead to a modification of the conclusions reached by the UK RAR and reinforced some of the aspects elaborated above.

In the study of Kuwada et al. (2002) described below testicular weight was decreased on day 21 but reached control levels on day 35 and 50 post partum. In addition histopathological alterations of testes were reported at d 21 but without giving any details for the 8 different test chemicals studied. The relevance of these findings is difficult to interpret due to the unphysiological application route, insufficient reporting (i.e. number of animals/group is unclear), no other reproductive organs or hormones were studied, and since only one dose level was used not allowing a dose response interpretation.

Chamkhia et al., (2006) treated male rats (2-3 months of age) over ten days with 600 mg/kg bw/d by the intraperitoneal route. A significant increase of the relative testicular weight was found as well as histopathological alterations of testes including loss of gametes in the seminiferous tubules. Motility and number of epididymal spermatozoa was decreased. These finding are clearly in contrast to those of the guideline 2-generation study and other long-term studies by the oral and inhalation route. Furthermore, the relevance of this publication is questionable due to the unphysiological application route and since only one dose level was used not allowing a dose response interpretation.

In a series of investigations male rats (8 months of age; number of animals not given) were orally exposed to styrene (0, 50, 100, and 500 mg/kg bw/d) over 5 days. There was a dose related decrease of sperm counts and motility and an increase of the percentage of abnormal sperm. These effects were statistically significant at the highest dose level. Reactive oxygen species were increased in sperm at all dose levels. Furthermore, the mRNA expression of catalase and superoxide dismutase-2 was suppressed by styrene treatment. The authors conclude that styrene may act on sperm cells by enhancing oxidative stress (Chun et al., 2005). The same group further investigated the differential gene expression in testes of rats subjected to the same exposure schedule (Han et al., 2007). The expression of clusterin mRNA was decreased in a dose related manner in testes (but not in other organs) while some other mRNAs were upregulated. As clusterin is involved in apoptotic processes in several hormone dependent tissues and may be involved in sperm maturation according to the authors this finding indicates a role of clusterrin repression in styrene mediated toxicity.

The studies mentioned above must be critically assessed against recent subchronic and chronic guideline studies by inhalation exposure showing no treatment related effects on the reproductive organs (including testes)(Cruzan et al., 1997; 1998; 2001) and against the 2-generation study (Cruzan et al., 2005a) where no effects on spermatogenic endpoints were noted. Therefore, in summary the weight of evidence does not give an indication for adverse effects of styrene on male reproductive parameters.

 

 

Luderer et al. (2005) published a comprehensive review based on the literature available by March, 2005, on the reproductive and developmental toxicity of styrene. The report was prepared for thefor the Evaluation of Risks to Human Reproduction (CERHR). The panel comprised 10 government and non-government scientists and the report was reviewed by CERHR staff scientists and by members of the Styrene Expert Panel. The NTP-CERHR is headquartered at NIEHS. The CERHR basically came to the same conclusions as the UK Risk Assessment Report with only minor deviations:

The human data are insufficient to conclude that styrene is a reproductive toxicant (emphasis added). There is suggestive evidence that exposure to styrene in occupational settings is associated with increased serum prolactin and depletion of peripheral blood dopamine metabolizing enzyme activities relative to unexposed individuals. Although the clinical relevance of these findings is unclear, the findings warrant further investigation.

In the update of literature until Oct. 01, 2015, the following new information was found:

Eldesouki et al. (2013) studied 60 painters (mean age 39.2 years; mean exposure duration 13.7 years) out of 1000 male workers in a furniture manufacturing plant exposed to a mixture of volatile organic chemicals in comparison to 27 non-exposed controls. Only 5 % of the workers used regularly protective equipment. About half of the exposed subjects complained about musculoskeletal pain and/or half about mucous membrane irritation. Testosterone levels were significantly decreased while LH and FSH were significantly increased in the exposed workers with a significant negative correlation between exposure duration and testosterone levels and a positive correlation with FSH. No information is available for the selection of the 60 workers out of a total of 1000 and about participation rate. The highest exposure concentrations were noted for toluene (150 mg/m³), xylene (100 mg/m³), styrene (90 mg/m³), and bromobenzene (50 mg/m³), while for benzene the concentration was 1.3 mg/m³. No information is given when and how these measurements were made. It is not possible to assign the effects reported to any specific substance,

 

Available experimental data in rodents are sufficient to conclude that styrene is not a reproductive toxicant. These rodent data are assumed relevant for humans:

- A NOAEL of 500 ppm by inhalation for male reproductive toxicity was identified. This was the highest dose tested. A NOAEL of 250 ppm in drinking water (18 mg/kg bw/day) was identified. This dose was the highest level tested.

- A NOAEL of 500 by inhalation for female reproductive toxicity was identified. This was the highest dose tested. A NOAEL of 250 ppm in drinking water (23 mg/kg bw/day) was identified. This dose was the highest level tested.

Based on the styrene experimental animal data, the Expert Panel expressed negligible concern for reproductive toxicity in humans; however, there is insufficient epidemiologic evidence to support this conclusion. There is the outstanding question of the clinical relevance of the prolactin findings observed with occupational exposures.”

Short description of key information:
- rat (males and females), 2-gen study: NOAEC = 500 ppm for fertility (Cruzan et al. 2005a)

Effects on developmental toxicity

Description of key information

- rat (males and females), 2-gen study: NOAEC = 150 ppm for developmental toxicity (Cruzan et al. 2005b)

Additional information

Summary of developmental toxicity studies in animals

Data from inhalation and oral developmental toxicity studies in a number of species are available, but most are either poorly designed or reported. There are no studies using the dermal route of exposure.

In the rat, inhalation exposure produced no evidence of significant effects on conventional parameters assessed in the foetus at non-maternally toxic exposure concentrations of up to 600 ppm styrene (Murray et al., 1978). However, developmental delays have been reported postnatally in a number of non-GLP, non-OECD studies at 300 ppm styrene in the absence of overt maternal toxicity (Kishi et al., 1995; Katakura et al., 2001). This has been confirmed by a recent, well conducted two-generation study in which, a pattern of developmental delay (delays in attaining some pre-weaning developmental landmarks and in acquiring preputial separation, decreased swimming ability, slight shift in the normal pattern of motor activity and reductions in forelimb grip strength), including decreased body weights, was evident mainly in the F₂ pups of the high exposure group (500 ppm) (Cruzan et al., 2005b). It is noted that, in contrast to previous investigations, in this OECD- and GLP-compliant study the delay in pup development was seen at an exposure level causing some maternal toxicity (reductions in body weights of 6-7% and degeneration of the nasal olfactory epithelium). No specific developmental neurotoxicity was seen in this study up to the highest tested concentration of 500 ppm. In this investigation styrene exposure caused a statistically significant decrease in body weight gain of the high-exposure F₁pups (by 11%) and in body weight of the mid- (by 7-10%) and high- exposure F₂pups (by 10-13%). Generally, a pattern of developmental delay was evident mainly in the F₂high-exposure group. The delay in attaining some pre-weaning developmental landmarks (pinnal detachment, surface righting response, incisor eruption and hair growth) and in acquiring preputial separation, the decreased straight channel swimming ability, the slight shift in the normal pattern of motor activity and the reduction in forelimb grip strength observed in this study are parameters known/likely to be body weight-sensitive and hence are not considered to represent a direct and specific expression of styrene developmental neurotoxicity, but the consequence of generalised toxicity.

Other species investigated using the inhalation route of exposure included the mouse, rabbit and Chinese hamster. No styrene exposure-associated effects were observed in the rabbit study (Murray et al., 1978). The majority of the studies conducted in mice suffer from many limitations (Kankaanpää et al., 1980) and so, no conclusions can be drawn from them. In a recent study (Ninomiya et al., 2000), reduced placental and foetal weights were observed at an exposure level (100 ppm) associated with impairment of maternal growth. Significantly higher numbers of dead or resorbed implants were observed at 1000 ppm in the Chinese hamster study (Kankaanpää et al., 1980). However, in the absence of information on maternal toxicity, which is most likely to have occurred at such a high exposure concentration, no firm conclusions can be drawn from the results of this study.

Only studies in the rat are available using the oral route of exposure; generally no significant effects on any of the conventional parameters assessed in the foetus were seen at dose levels up to 250 mg/kg/day, at which maternal toxicity was observed (Murray et al., 1976; Srivastava et al., 1990). In one study, an increased incidence of a fetal anomaly (dilated renal pelvis) associated with a significant reduction in maternal body weight was observed following the administration of 1147 mg/kg/day styrene (Chernoff et al., 1990). Fetotoxicity (reduced foetal weight and increased embryonic/foetal death) was also reported at 400 mg/kg/day in another study, but this again was in the presence of maternal toxicity (Srivastava et al., 1990). An increase in the number of brain dopamine receptors in rat pups exposed to 200 mg/kg/day styrene during gestation and lactation or during lactation only was reported in one study (Zaidi et al., 1985), however, the toxicological significance, if any, of this finding is unknown. 

Overall, it can be concluded that styrene does not cause developmental toxicity in animals as evaluated by structural endpoints at inhalation exposures of up to 600 ppm and oral exposures of up to 250 mg/kg/day and by neurological endpoints at inhalation exposures of up to 500 ppm. However, reduced pup growth and pup developmental delays (delays in attaining some pre-weaning developmental landmarks and in acquiring preputial separation, decreased swimming ability, slight shift in the normal pattern of motor activity and reductions in forelimb grip strength) were seen postnatally in rats at exposure levels (300-500 ppm) causing, in some cases, maternal toxicity. With the exception of a small reduction (up to 10%) in pup body weight, no developmental effects were observed at 150 ppm (»120 mg/kg/day) in a well-conducted 2-gen study.

Taking into account all of the available information, it is suggested that 150 ppm is taken forward to the risk characterisation as the NOAEC for potential effects of styrene on development. Although at 150 ppm there was a decrease in pup body weight, since this was small (up to 10%), limited to the pre-weaning period of the F₂generation only and not accompanied by other related effects, it was not considered sufficient to set the NOAEC for effects on pups at the next lower exposure concentration. This information will be taken into account when judging the adequacy of the Margins of Safety in the risk characterisation.

 

There are four effects noted in F2 pups in the 2-generation study (Cruzan et al., 2005a) that warrant a more detailed discussion as they may be taken as an indication for a specific developmental toxicity of styrene. These effects are:

1. patterns of delay in attaining developmental landmarks at 500 ppm


2. reduced forelimb grip strength on pnd 60 at 500 ppm


3. increased straight channel swim time at 500 ppm


4. decreased body weight in F2 offspring of F1 rats at 150 ppm in the absence of maternal toxicity.

The basic question is whether these are direct, specific effects on development, or rather non specific delays associated with maternal toxicity in F0 and F1 generations combined with some “chance” variations inherent in the large number of observations and measurements in the course of this investigation. In this respect it should be taken into consideration that in total more than 140 different observations are statistically analyzed.

Ad 1: delays in certain developmental landmark at 500 ppm. The table shows the days of acquisition of the developmental landmarks and only for incisor eruption statistical significance (*p<0.05) is obtained

Landmark	Pina detachment	Surface rightning	Incisor eruption	Hair growth	Eye opening	Vaginal opening	Preputial separation
0 ppm	4.0 (0.07)	5.1 (0.10)	9.3 (0.30)	11.8 (0.82)	15.1 (0.95)	33.5 (1.26)	45.3 (1.48)
500 ppm	4.1 (0.22)	5.2 (0.18)	9.8 (0.57)*	12.5 (1.12)	15.5 (0.98)	34.1 (2.48)	47.2 (4.07)

 

On pnd 0-21 the mean F2 pup weights at 500 ppm were statistically lower (7-13%) than those of the 0 ppm control animals as shown in the following table (* p<0.05)

pnd	Female 0 ppm	Female 500 ppm	Male 0 ppm	Male 500 ppm
1	7.1	6.4*	7.4	6.9*
7	14.3	12.4*	15.2	13.5*
13	24.3	21.2*	25.4	22.7*
21	40.5	35.4*	42.6	38.0*

 

Therefore, the delay of acquisition of some developmental landmarks in F2 pups, especially incisor eruption, is clearly related to reduced body weights at 500 ppm and not an indication of selective toxicity. This corresponds to the conclusion of the study report linking these effects to reduced body weight. Furthermore, there were no effects found on developmental landmarks in the F1 pups. The EU Risk Assessment report concludes that reduced pup growth and pup development are of no lasting consequence with no indications of selective toxicity on neurodevelopment.

Ad 2: reduced forelimb grip strength on pnd 60 at 500 ppm: It is well known that body weight and grip strength are correlated (Maurissen et al., 2003). But it may be argued that the reduction in bodyweight in the F2 pups (pnd 63) is insufficient to explain the reduction in forelimb grip strength on pnd 60:

-        males: 23.7% reduction in strength; weight reduction 8.6%

-        females: 28.4% reduction in strength; weight reduction 6.3%.

There are good reasons that the reduction in grip strength is not an indication for a specific developmental effect but is rather attributable to a combination of reduced body weight and chance variations in data.

-        although reductions in body weights of F2 pups on pnd 63 might be considered moderate (6.3 and 8.6%), during pre-weaning the weight reductions were much more pronounced (10-13%) resulting in continued reductions throughout the study

-        there is no simple proportionality between grip strength and body weight, e.g. body weights in male rats increased by 500% from pnd 22 to pnd 45 while grip strength increased only be 258%

-        there was a large variability in grip strength measurements with few animals out of the control range, i.e. 5/20 males and 3/20 females

-        F2 pups at 500 ppm also showed slight delays in a variety of developmental landmarks that were related to body weight

-        There were no effects noted for grip strength on pnd 22 and 45

-        Forelimb grip strength in 500 ppm pups were not significantly different from historical control values from 8 studies of the laboratory

-        There was a significant increase in forelimb grip strength in females at pnd 45 in the 150 ppm group indicating to the variability of these measurements

-        Significant effects were noted only for forelimbs, but not for hindlimbs with longer (more sensitive) nerve fibres.

Ad 3: increased straight channel swim time at 500 ppm on pnd 24. It may be argued that the reduction in bodyweight (10-13%) in the F2 pups is insufficient to explain the increase in swim time on pnd 24 (30-40%)

-        males: 10.6 seconds (500 ppm) versus 7.5 seconds (0 ppm)

-        females: 11.4 seconds (500 ppm) versus 7.8 seconds (0 ppm)

Straight swim time provides the base line for swimming ability in thewater maze; it is indicator of locomotor function and not of memory and learning. There were no effects on learning and memory in 12 trials. There are good reasons that the increase in straight channel swim time is not an indication for a specific developmental effect but is rather attributable to slight developmental delays as a result of a combination of reduced body weight and chance variations in data.

-        An effect was only noted for straight swim time on pnd 24 but not in the trial on pnd 62.

-        The historical data of the laboratoryfrom 20 studies show the swin time for the F2 pups from the 500 ppm group are within the historical control values while the swim time of the concurrent controls is particularly fast:

 

	No. of studies	Historical control data		Styrene
				Controls		500 ppm
		male	female	male	female	Male	female
PND 22	17	11.2	11.7
PND 24				7.5	7.8	10.6	11.4
PND 25	2	9.2	8.5
PND 26	1	8.4	9.3

 

As swimming ability and grip strength are basically both indicators of locomotor activity these data are consistent with the slight delay of developmental landmarks related to reduced body weights. Therefore, this finding is not to be taken as a specific effect on the nervous system development.

Ad 4: decreased body weight in F2 offspring of F1 rats at 150 ppm in the absence of maternal toxicity. There were statistically significant reductions in bodyweights at pnd 13 and 21 in male F2 offspring of F1 rats exposed to 150 ppm. This may be taken as not being a general response to maternal toxicity as no statistically significant effects on bodyweights of F1 females at 150 ppm were observed during premating and gestation. Bodyweights of the male and female F2 offspring are given in the table (* p<0.05):

 

	PND1	PND 4	PND 7	PND 13	PND 21	PND35	PND49	PND70
Male 150 ppm	7.3	10.5	14.2	23.1*	38.2*	127	248	372
Male Control	7.4	10.7	15.2	25.4	42.6	134	260	388
Female 150 ppm	6.9	10.0	13.7	22.6	37.4	112	171	233
Female Control	7.1	10.1	14.3	24.3	40.5	120	179	248

 

For comparison, the body weight reductions of male and female parental F0 and F1 animals (week 0-7 and week 0-10) and of maternal animals at gd 14 and 21 are shown in the following table. Reductions (* p<0.05) were related in % to the body weight of control animals (data taken from Cruzan et al., 2005a):

 

FO	m 150	m 500	f 150	f 500		f 150	f 500
W 0 – 7	6*	8*	4	7*	GD14	1	4
W 0 – 10	5	7	3	8*	GD21	0	2
F1	m 150	m 500	f 150	f 500		f 150	f 500
W 0 – 7	7*	8*	5	7*	GD14	4.5	6*
W 0 – 10	7	8*	5	8*	GD21	4	6

 

There are good reasons for proposing that the significant decrease in body weights seen in male F2 offspring of F1 rats exposed to 150 ppm is consistent with systemic toxicity and cannot be regarded as a specific developmental effect:

-        The table shows that there are statistically significant decreases in male bodyweights also periodically in the F0 and F1 parental generations exposed to 150 ppm.

-        There are also clear reductions in bodyweights in both the F1 and F0 females exposed to 150 ppm but without reaching statistical significance, including the F1 maternal animals during gestation

-        While the bodyweights of F2 offspring males at 150 ppm were significantly different from F2 control body weights (i.e. 38.2 g versus 42.6 g) they are almost identical to the control male bodyweights of the F1 offspring generation (i.e. 38.4 g), again showing variability in data

-        Weight reductions seen in the F0 generation (exposed as adult animals) clearly indicate that the response is related to general toxicity and not a specific developmental effect.

Overall evaluation:

When evaluating data from DNT studies it is important to take into account patterns of effects rather than focus on individual measures as isolated endpoints. For example Moser (1991) introduced the concept of functional domains which involves grouping inter-correlated endpoints to examine patterns of response (Moser V. Applications of a Neurobehavioral Screening Battery. JTox 10:661-669). This approach of examining patterns of effects is shown in the following tables for the 500 ppm exposure group:

Results relevant to the neuromuscular domain:

	PND 20-28	PND 60-74
Grip Strength	None	Decrease (M, F)
Mobility	None	None
Gait	None	None
Motor activity	None	None
Swim Time –Straight Channel	Increase time	None
Neuropathology	Not evaluated	None

 

Results relevant to activity and excitability domains:

	PND 20-28	PND60-72
Ease Removal	No effect	No effect
Ease Handling	No effect	No effect
Arousal	No effect	No effect
Home cage-posture	No effect	No effect
Motor Activity	(Increased activity?) n.s.	No effect
Swim Time –	Increase time	No effect
Startle (Vmax)	No effect	No effect
CNS Neuropath	None	None

n.s.: not statistically significant, but noted as slight developmental shift

 

This detailed analysis of the data obtained in the course of the 2-generation study, especially for the F2 offspring, leads to the following conclusions:

• There is no consistent pattern of effects to indicate a direct impact on development of the nervous system


• Inconsistent effects on grip strength and swim time together with slight delays in various developmental landmarks are considered secondary to the general toxicity characterized by reduced body weights and nasal pathology of parental animals


• Reduced body weights in F2 rats are also seen in F1 and F0 parental animals as an indication of general toxicity


• Biological variability in the data may also account for some observational differences between control and treated animals

 

 

There are some non-guideline investigations on postnatal effects after styrene exposure that may be discussed in this context. Neurochemical effects reported by Zaidi et al. (1985), Kishi et al. (1992), and Katakura et al. (1999) will not be considered here because the toxicological relevance can hardly be assessed without a clear association to other guideline parameters.

In addition to effects on dopamine receptors Zaidi et al. (1985) described behavioural effects (amphetamine-induced locomotor activity and apomorphine induced stereotypy) after gestational and neonatal exposure of maternal rats to styrene (200 mg/kg bw/d, orally). The small number of only 3 or 4 mother animals per group precludes any firm interpretation. Similarly, Kishi et al. (1995) studied postnatal parameters in only 5, 2, and 5 litters exposed to 0, 50, and 300 ppm during pregnancy.

A follow up study by the same group exposed 9 and 14 pregnant rats to 50 and 300 ppm using an ad libitum fed control group (14 dams) and a pair fed control group (12 dams) (Katakura et al., 2001). The pair fed control group and the 50 ppm group were given the same quantity of feed as that consumed by the 300 ppm exposure group. Only this study will be discussed in more detail. Dams were exposed during gestation and pups were observed until weaning (pnd 21). Although maternal feed intake of the 300 ppm group was significantly reduced during gestation in comparison to the ad libitum controls (p<0.01), there was no statistically significant effect on body weight gain during gestation. Maternal clinical symptoms were not reported nor body weight or feed consumption after delivery. Therefore maternal toxicity cannot be reliably assessed. The 300 ppm male pups showed a statistically lower body weight at pnd 21 in comparison to both control groups. Incisor eruption, eye opening, and acquisition of air righting reflex were delayed. In contrast, surface righting reflex was not affected in the guideline study of Cruzan et al. (2005b) neither in F1 nor in F2 animals. Thus, at 300 ppm some effects observed in this investigation corresponded to those reported by Cruzan et al. (2005b). Nevertheless, the overall interpretation should be based on the guideline study (Cruzan et al., 2005b) as discussed above in detail.

 

 

For DNEL derivation (long-term general population and worker) the effects found in the combined 2-generation/developmental neurotoxicity study Cruzan et al., 2005a; b) need to be discussed in relation to the lifetime phases when they occurred.

Body weights (Cruzan et al., 2005a):

As exposure related reductions in body weights represents the toxicological endpoint used by the UK HSE to develop the DNEL a short description of these data is provided here below:

-         F1 pups: there was no treatment related effect on body weights during pre-weaning at all exposure levels, i.e. 50, 150 or 500 ppm styrene. During post-weaning, (when rats where exposed to styrene), significant reductions in body weights were noted in both male and female rats in the 500 ppm exposure group and in male animals in the 150 ppm exposure group.

-         F2 pups: body weights were significantly reduced during pre-weaning in males at 150 and 500 ppm and in females at 500 ppm. The reduction persisted throughout the post-weaning period in the 500 ppm treatment group although during this period the reduction in body weights failed to reach statistical significance (Cruzan et al., 2005b).

-         F0 and F1 parents before mating: significant reductions in body weights, (mostly by about 6-8%), were measured in both males and females rats exposed to 500 ppm styrene. Numerically and statistically significant reductions in body weight were also measured in F0 and F1 male rats respectively in the 150 ppm treatment group

-         F0 and F1 females during gestation: body weights were numerically reduced, but statistical significances were observed only at some time points for F1 mothers.

Based on these data the UK HSE Risk Assessment Report on Styrene suggested (quote) “that 150 ppm is taken forward to the risk characterization as the NOAEC for potential effects of styrene on development. Although at 150 ppm there was a decrease in pup body weight, since this was small (up to 10%), limited to the pre-weaning period of the F2generation only and not accompanied by other related effects, it was not considered sufficient to set the NOAEC for effects on pups at the next lower exposure concentration. This information will be taken into account when judging the adequacy of the Margins of Safety in the risk characterization”.

An important question relating to the above data is whether the small reductions in body weights reflect a specific developmental effect or simply a general low level toxicological or behavioral consequence of high exposures to styrene. As body weight reductions were seen:

• in the F0 generation, (i.e., fully developed adult animals), and


• there were no effects up to 500 ppm in F1 pups during pre-weaning, (a true developmental effect would become evident especially in this early phase),

it is proposed that the reduced weight gains are a general toxicological or behavioral consequence of high dose exposures to styrene and not a specific developmental effect. This is supported by the fact that reduced body weights in the F1 generation occurred only after direct exposures by inhalation to styrene began on PND 22. In addition a reduction in body weight is a common effect seen in adult rats exposed to styrene for 90-days (Cruzan et al., 1997). Similar effects were reported in a 2-year study (Cruzan et al., 1998) where decreased body weights were noted in males starting at 500 ppm and in females starting at 200 ppm. The overall weight of evidence thus suggests the reductions in body weight gains is not a direct developmental effect of styrene but more likely a secondary consequence of other effects such as the severe nasal irritation characterized by degeneration of the olfactory epithelium which occurs in rodents following inhalation exposure to styrene. The sensitivity of rats to nasal irritation by styrene is explained by the high metabolic capacity of rodent nasal epithelium which converts styrene to toxic intermediates. By contrast human nasal epithelium is almost devoid of the metabolizing enzymes rendering human nasal epithelium resistant to nasal irritation caused by toxic metabolic intermediates.

In summary the effects of styrene exposure on body weights is probably a secondary consequence of a general debility in the exposed rats related to the severe olfactory irritation, unpleasant odour with styrene residues remaining on skin and fur following whole body exposures causing palatability problems, and mild narcosis at high doses.

Although no effects on body weights were seen in the F1 pups during pre-weaning reduced body weights were measured in the F2 pups (both sexes at 500 ppm) during the pre-weaning period. Leaving aside the relevance of effects in F2 pups in general for setting a workplace exposure limit (this is discussed later), the effect in the F2 pups is explained by the fact that in contrast the F0 generation which were mature animals at the start of exposures to styrene the F1 generation were basically still growing and developing when they were first exposed to styrene on PND 22. Thus it is perhaps not surprising that a greater effect on body weights was seen in the F2 generation compared to the F1 animals. In addition, although styrene exposures did not produce severe overt toxicity in the F1 dams, the doses were high enough to produce weight loss and respiratory effects as well as subtle, reversible neurobehavioral alterations probably impacting the level of maternal care. This is supported by the findings of several other investigations:

-         Savolainen, Pfäffli (1977) described that rats at 300 ppm (6 h/d, 5 d/week over 2-11 weeks) tended to be somnolent after the end of daily exposures in the early phases of the experiment although this tendency levelled off towards the end of the experiment.

-         Styrene exposure impaired rat operant behaviour at 500 and 1000 ppm, but not at 50 or 150 ppm. Animals were consecutively exposed three times over three weeks (4 h/d, 5 d/week) followed by two weeks without exposure. In the exposure free intervals the operant behaviour immediately returned to normal. Thus, the effect was reversible and only short lasting (Fumiko, 1988).

-         Kulig (1989) described impairments for response speed and accuracy in a two-choice visual discrimination task in rats at 350 ppm (16 h/d, 5 d/week, over 18 weeks). The effects were most pronounced on the first day of exposure and thereafter rapidly levelled off.

-         Coccini et al. (1999) observed that rats exposed to 50 and 300 ppm (6 h/d, 5 d/week) appeared lethargic during exposure, but promptly recovered at the end of exposure.

-         Cruzan et al., (1998) determined body weights on two exposure days before and after the 6h exposure period. Weight loss occurred in all exposure groups in males (50 – 1000 ppm) and in females at 1000 ppm thereby demonstrating a short-lasting impairment of the animals during exposure.

 

These subtle and reversible changes are often associated with inhalation exposure to high concentrations of chemicals which exhibit general and non-specific depression of the central nervous system. Overall, these findings in combination with the severe nasal toxicity give good evidence that maternal care has been impaired at 500 ppm in the multigeneration study.

 

In conclusion, sufficient indication for developmental toxicity per se cannot be derived from body weight effects in offspring at 500 ppm.

 

 

Relevance of effects in F1 and F2 pups for a worker-DNEL:

As no evidence for either reproductive toxicity or developmental neurotoxicity was found in the studies reported by Cruzanet al(2005a, b) it is not possible to calculate a DNEL for either of these endpoints. It is however possible to calculate a DNEL based on the highest dose for which no effect on either reproduction or neurodevelopment could be found, i.e. 500 ppm.

In the absence of reproductive or neurodevelopmental toxicity it might be suggested that there was a subtle developmental effect, characterized by the reduced body weights in F2 pups in the 500 ppm treatment during weaning. Thus the NOAEC for a non-specific delay in development is, (in accordance with the assessment in the UK HSE RAR), 150 ppm. For the F1 generation there was no retardation of growth during pre-weaning, so the NOAEC is the highest dose of 500 ppm. Despite the questionable relevance of the reduction in weight gain for human health it can be argued it represents a treatment related effect and is suitable for DNEL calculations. Based on this argument it must be decided whether the F1 or the F2 generation should be the starting point to derive a worker-DNEL. To help reach a decision it is useful to compare the experimental exposure schemes with real-world workplace situations.

 

While it is difficult to reconcile the exposure conditions experienced by experimental animals in a 2-generation reproductive study with exposures of pregnant women in the workplace, the maternal and paternal exposure patterns for the F1 generation (i.e. offspring from the F0 generation) are closer to the human workplace exposure than F2 generation. The reason is linked to the fact that styrene exposures in the F0 generation began when the rats were fully mature, corresponding to the time when humans will enter the workforce. Also, while exposure during pregnancy and lactation (via the milk) might occur in rare circumstances (e.g. if maternity leave not taken), it is most unlikely that inhalation exposures to styrene will occur in humans directly after post-weaning through to beginning of maturity which is the exposure regimen experienced by the F1 generation producing the F2 pups. Based on these exposure considerations the developmental toxicity results for the offspring of the F0 generation, (i.e. the F1 generation), are more relevant to the human experience at the workplace than those for offspring of the F1 generation (i.e. the F2 generation). The effects observed in F2 pups are only relevant for an environmental exposure scenario, but here concentrations prevail orders of magnitude lower.

Based on the consideration described above it is proposed the workplace DNEL should be based on the developmental effects observed in the F1 generation, i.e. on a NOAEC of 500 ppm the highest dose tested in the studies of Cruzan et al. (2005a, b). On the other hand, the long-term DNEL for the general population should be based on the NOAEC of 150 ppm observed for the F2 generation.

 

 

Summary of studies in humans

A range of epidemiological studies, particularly focussing on developmental effects, have been conducted but most of these lacked adequate exposure information and were too small to be conclusive. Nevertheless, the studies have been generally negative and the available human data certainly provides no reliable evidence for styrene exposure-related increases in spontaneous abortions (Lindbohm et al., 1985; 1990; McDonald et al., 1988; Taskinen et al., 1989), congenital abnormalities (Kurppa et al., 1983;Härkönen et al., 1984;Holmberg et al., 1986; Ahlborg et al., 1987; Taskinen et al., 1989), birth weight (Ahlborg et al., 1987; Lemasters et al., 1989), menstrual disorders (Lemasters et al., 1985), fecundity (Kolstad et al., 1999a), male or female fertility or sperm quality (Kolstad et al., 1999b; c) within the exposure ranges investigated. Overall, there is no clear evidence of an effect of styrene on human reproduction, but data are too limited to exclude the possibility for effects.

There is suggestive evidence that exposure to styrene in occupational settings is associated with increased levels of serum prolactin relative to unexposed individuals (see repeated dose toxicity section, human data, biochemical studies related to nervous system functioning). The interpretation of the clinical relevance of these effects is uncertain because the average elevation was not outside the normal range and because menstrual function and other reproductive endpoints were not evaluated in these studies. Therefore, overall, there is no clear evidence of an adverse effect of styrene on human endocrine function.

 

 

Luderer et al. (2005) published a comprehensive review based on the literature available by March, 2005, on the reproductive and developmental toxicity of styrene. The report was prepared for thefor the Evaluation of Risks to Human Reproduction (CERHR). The panel comprised 10 government and non-government scientists and the report was reviewed by CERHR staff scientists and by members of the Styrene Expert Panel. The NTP-CERHR is headquartered at NIEHS. The CERHR basically came to the same conclusions as the UK Risk Assessment Report with only minor deviations (quote):

“The human data are insufficient to conclude that styrene is a developmental toxicant (emphasis added). Available experimental data indicate little or no potential to produce developmental toxicity in laboratory animals. These data are assumed to be relevant for humans. Accordingly, dose levels were identified from experimental animal studies for use in this evaluation:

- A NOAEL of 600 ppm by inhalation and 300 mg/kg/day by oral gavage for prenatal developmental toxicity were identified. These were the highest doses tested. A NOAEL of 150 ppm and LOAEL of 500 ppm (by inhalation) were identified for postnatal development. The 500 ppm level produced maternal toxicity.

- A NOAEL of 500 ppm by inhalation was identified for developmental neurotoxicity. This dose was the highest level tested.

The Expert Panel was not able to separate the developmental delays and growth effects of styrene from effects that may be due to maternal toxicity at the same exposure levels

Based on the styrene experimental animal data, the Expert Panel expressed negligible concern for developmental toxicity in humans; however, there is insufficient epidemiologic evidence to support this conclusion. There is the outstanding question of the clinical relevance of the prolactin findings observed with occupational exposures.”

 

 

References not listed in the basic data set

Bachmann S. et al 1998	Drug Chem Toxicol 21 (Suppl. 1): 1-30
Hirano S. et al. 2002	J. Agric. Food Chem. 2001 49 4127-4131
Fail P. A. et al 1998	Drug Chem Toxicol 21(Suppl 1): 101-121
Fail P. A. et al 1998	Drug Chem Toxicol 21(Suppl 1): 101-121
Ohyama K. and Nagai F. 2002	Environmental Health Perspectives 110: 7
Ohyama K. et al. 2007	Exp Biol Med (Maywood) 232(2): 301-8
Ohyama K. et al. 2007	Exp Biol Med (Maywood) 232(2): 301-8

Toxicity to reproduction: other studies

Additional information

Studies of endocrine disruption activity

Overall, there is no evidence that styrene possesses significant endocrine disruption activity. This evaluation was based on negative in vitro studies for estrogenic activity (Nobuhara et al., 1999; Ohno et al., 2001) and androgenic activity (Nobuhara et al., 1999) and on a negative uterotrophic assay in female rats (Nobuhara et al., 1999). In an inhalation study with male rats at exposure concentrations up to 1484 ppm no effects were reported for prolactin and dopamine in serum nor for catecholamines and their metabolites in the striatum and mediobasal hypothalamus (Jarry et al., 2002).

A literature search from 1998 to January 2010 overlapping the date of the UK RAR (June 2008) was conducted. Overall, the studies identified did not lead to a modification of the conclusions reached by the UK RAR and reinforced some of the aspects elaborated above.

When styrene was tested in a series of screening assays no indication for hormonal activity was obtained in any of these tests (Date et al., 2002; Azuma et al., 2000). The test battery included: Estrogen and androgen receptor binding assays, uterotrophic assay with prepubertal and ovariectomized adult rats, Hershberger assay for anti-androgenic activity measuring the weights of seminal vesicle, ventral prostate, and levator ani plus bulbocavernosus muscle, thyroid hormone receptor binding assay, and a rat serum prolactin assay by treatment of ovariectomized rats over 3 days. In addition, no effect indicative for an estrogenic activity was found in the MCF-7 cell proliferation assay (Yamada, 1999).

Kuwada et al. (2002) treated neonatal rats on the day of birth with a single subcutaneous injection of 2 mM styrene in sesame oil. Assuming a bodyweight of 5 g this dose would correspond to approximately 1 mg/kg bw. Steroidogenesis was investigated on day 21 and 50 after application by incubation of testicular microsomes with progesterone. The synthesis of testosterone, 17-hydroxyprogesterone, and androstene-3.20-dione was clearly reduced on day 21 but had returned to control levels on day 50. The relevance of these findings is difficult to interpret due to the unphysiological application route, insufficient reporting (i.e. number of animals/group is unclear), no other reproductive organs or hormones were studied, and since only one dose level was used not allowing a dose response interpretation.

 

In order to investigate a possible impact of styrene on male fertility, Jarry et al. (2004) exposed male rats to 150, 500, and 1500 ppm styrene over 5 days (6h/d). Directly after exposure the serum levels of LH (luteinizing hormone) and testosterone were increased. This was not accompanied by a corresponding change of glutamate and gamma-aminobutyric acid in the striatum or mediobasal hypothalamus. After a 24 h recovery period gamma-aminobutyric acid was increased in the striatum of the 500 and 1500 ppm groups without a dose relationship. The authors concluded that this study suggests, but does not unequivocally proves, that styrene does not have reproductive toxicity effects in men.

Completely different results were reported by Chamkhia et al., (2006) after intraperitoneal treatment of male rats (2-3 months of age) over ten days with 600 mg/kg bw/d. 24 h following the last application the serum concentrations of testosterone had dropped significantly by 83% and those of LH and FSH (follicle stimulating hormone) had increased by nearly 5- and 3-fold, respectively, in comparison to the control group. The relevance of these findings is difficult to interpret due to the unphysiological application route and since only one dose level was used not allowing a dose response interpretation.

Taken together, the more recent studies cannot be taken as evidence for an endocrine disrupting activity of styrene.

By the update of literature up to Oct. 01, 2015, the following new publications were obtained:

Gelbke et al. (2015) reviewed the literature on potential endocrine effects of styrene for (anti)estrogenicity, (anti)androgenicity, interruption with thyroidal regulation and effects on prolactin serum levels. Based on screening and animal data, epidemiological studies and mechanistic information it was concluded that styrene lacks primary endocrine disrupting properties.

Kuwada et al. (2012) using the same procedure as Kuwada et al. (2002) investigated the effects on the ovary of 13 substances known or suspected for interaction with the estrogenic or androgenic pathway. The concentrations administered varied between 2 mM (for styrene) and 40 mM. Some of the test substances had already been studies by Kuwada et al. (2002). Again no rationale for dose selection was given but interestingly the highest concentration of 40 mM was used for the most potent androgens (like dihydrotestosterone or testosterone) or estrogens (like estradiol-17ß or diethylstilbestrol). For styrene it is only noted that ovary weights “were restored to those of the untreated” controls on postnatal day 50 but the findings on day 21 or 35 are not described. No effects for histopathology of the ovaries were reported. Metabolism of testosterone to estradiol-17ß by ovarian homogenate was increased on day 21 and was “considerably restored … but still reduced” on day 50. But numerical data and a statistical analysis are missing. The same deficiencies are noted for this study as for Kuwada et al. 2002).

In conclusion, there are no indications that styrene may act as an enocrine disriuptor.

Other effects, other studies:

The non-specific delay in development characterized by reduced weight gain seen in the two-generation reproduction study was accompanied by a slight delay in acquisition of few individual parameters used to investigate motor activity which were measured as part of a developmental neurotoxicity investigation linked to the 2-generation reproductive study – see below.

In the neurobehavioural study a slight developmental delay in locomotor activity was observed in F2 pups at 500 ppm:

Reduction of forelimb grip strength, but the mean was within the historical control range of the laboratory and only a few animals showed a reduction outside the concurrent control range. This suggests that the effect may actually be an expression of normal variation without toxicological significance. Grip strength has been correlated with body weight (Maurissen et al., 2003). Thus, the reduction in forelimb grip strength observed on PND 60 only is considered to be the consequence of the reduced body weight seen in these pups.

-         The “normal” age-related pattern of motor activity appeared to be slightly shifted. However, this slight shift was considered to be related to the growth delay evident in this group of animals particularly in the pre-weaning stage.

-         At PND 24, the mean time to escape in the straight channel swimming trial was statistically significantly increased in males and slightly but not significantly in females. However, no difference was seen at PND 62. Swimming time in the straight channel is a measure of basic swimming performance and does not represent an effect on learning and memory. Historical control data from the contract laboratory showed that the swimming time values at PND 24 were within the historical control ranges and that the observed increase was due to an unusually low value in the concurrent controls. This suggests that this increase in swimming time may actually be an expression of normal variation and have no toxicological significance and that this effect was consistent with the reduction in body weights.

-         Incisor eruption was delayed in F2 but not in F1 offspring and this si probably secondary to the reduced pup weight.

-         In addition, a non significant delay in preputial separation being within the historical control range was observed in the high-exposure F1 males. The correlation between body weight and preputial separation in rats is clearly established. Therefore, this effect is likely to be a consequence of the decrease in body weight observed in this group following direct exposure after weaning. As reductions in body weight around exposure concentrations of 500 ppm have also been observed in other studies (see above) this effect cannot be regarded as a consequence of developmental toxicity.

The findings in the F2 pups of the high exposure group were related to the reduced pup body weight being a consequence of reduced maternal care and maternal toxicity (reductions in body weights of 6-7% and degeneration of the nasal olfactory epithelium) and are not an indication of developmental toxicity. Overall, the 2-generation reproductive toxicity study combined with developmental neurotoxicity study found no evidence for styrene being either a specific reproductive toxicant or a selective developmental neurotoxicant.

Justification for classification or non-classification

In November 2012 RAC (2012) assessed the evidence for developmental toxicity based primarily on the study of Cruzan et al. (2005b) on developmental neurotoxicity in rats. Four effects noted in the study of Cruzan et al. (2005b) were assessed in detail in the RAC opinion leading to the conclusion to classify styrene as a category 2 developmental toxicant:

- a decreased pup growth in F2 offspring at 150 and 500 ppm (7-10% and 10-13%, respectively),

- a decreased relative pituitary gland weight (22% in F2 males),

- a decreased forelimb grip strength (24-28%), and

- an increased time to escape in straight channel swimming trials (38% in males).

The arguments of RAC for a cat. 2 classification were as follows (taken from RAC (2012) verbally):

“The RAC notes the dose-dependent decrease in pup weights in the second generation offspring (F2), and although there were some effects on the F1 maternal body weight at the top dose (reduction by 7-8%), the reduced growth of the pups at the mid dose supports that this could be a direct effect on the offspring. There were no effects on the weights of the first generation offspring (F1), even though F0 maternal weights were clearly affected (7-8%).

The relative pituitary weight was clearly decreased in males at the top dose, and of such a magnitude to indicate this to be an adverse effect even in the absence of any pathological findings. It has been argued in the PC (argument given by Industry during Public Consultation - explanation added) comments that the large variability in the weight of the pituitary between animals at PND 21 (but not in adults) makes it difficult to draw firm conclusions from the mean values observed on PND 21. On the other hand, if it is the developmental rate of the pituitary that is affected, the lack of pathological findings may be consistent with the decreased weight. Although this finding may constitute some evidence of developmental effects, the robustness of this finding is decreased by the lack of pituitary effects in F2 females (PND 21) or the F1 generation (PND 21 offspring or adults).

Forelimb grip strength was reduced (24-28%) in both sexes at 500 ppm at day 60 (but not on days 22 and 45). The magnitude of the effect on day 60 was larger than the effect on body weight, potentially indicating a neuromuscular effect. However, it cannot be ruled out that the effect is caused by the decreased growth rate. Hind limb grip strength was decreased (18%) at day 45, but only in 500 ppm exposed males, and without any effects on days 22 and 60. Overall, the grip strength tests give some evidence of neuromuscular effects, but the findings are clearly weakened by only being observed on one of the three occasions when it was studied, and by the observed effects on forelimb and hind limb grip strength not occurring on the same occasions. Discrepancies between effects on fore and hindlimb strength have, however, been reported in other studies (Maurissen et al., 2003).

Time to escape in straight channel swimming trials is assumed to reflect swimming ability and motivation to escape. The time to escape was increased by 38% in males of the 500 ppm group at day 24. Effects of a similar magnitude were observed in the females, but were not statistically significant (information from the EU RAR). Findings of similar effects on day 62 when it was studied again would have strengthened this observation. However, the dossier notes that the positive controls PTU and methimazole also only affected this parameter on day 24 but not on day 62. It is not explained in the dossier why PTU and methimazole can be considered as “positive controls”.

Observations of delayed development at 300 ppm in the two rat inhalation developmental studies are referred to as supporting information. The RAC notes that in both studies there are observations of effects on time of eye opening, righting reflex, and incisor eruption, and that these effects fit the pattern of effects observed in the two-generation study. The use of pair-fed control dams in one of these studies indicates that the effects are not caused by a decreased pup growth rate. The two developmental studies have some methodological deficiencies and can only be used as supportive studies.”

On this basis the RAC arrived at its classification proposal as follows (taken from RAC (2012) verbally):

“For styrene, there are indications of effects on development but the rather inconsistent effects (e.g., decreased pup growth in F2 but not in F1, decreased grip strength only at some time points, and effects on swimming trials at day 24 but not on day 62) cannot qualify as the ‘clear evidence’ required by the CLP. Furthermore, some relationship between decreased pup growth and the other effects cannot be completely ruled out. Thus, in the opinion of the RAC, classification with Repr. 1B, H360 (CLP) is not appropriate. As the criteria for DSD are very similar to the CLP criteria, classification with Repr. Cat 2; R61 according to the DSD is likewise not warranted.

The generally well performed two-generation study provided some evidence of long-lasting, delayed pup development, as exemplified by dose-dependently decreased F2 pup body weights at 150 and 500 ppm (10-13% at 500 ppm), a decreased pituitary weight in male 500 ppm F2 pups (22%), and decreased grip strength (24-28% forelimb grip strength) and swimming abilities at 500 ppm. In the weight of evidence assessment made by the RAC, it has been taken into consideration that styrene did not affect other parameters studied in the two-generation study. Two developmental studies in rats, in the absence of maternal toxicity, also indicated a delayed development of newborn pups (delayed eye opening, righting reflex, and incisor eruption), and decreased pup weights (8-11% at day 1 and 15% at day 21 at 300 ppm in one study and 8% at day 21 at 300 ppm in the other study) although there are some deficiencies in these studies. There might be a relationship between decreased pup growth and the other findings, but it is noted that the effect in the two-generation study on the pituitary weight, and decreased grip strength cannot be fully explained by the decreased growth rate. The possibility of the effects being caused by general pup toxicity rather than by specific developmental toxicity is discussed in the comments, but the RAC finds it difficult to distinguish between the two based on the available data.

Maternal toxicity was also discussed in the comments received during the public consultation as potentially explaining the observed effects. Maternal effects were only noted at the top dose (500 ppm) in the two-generation study. They consisted of nasal toxicity and a reduced body weight gain, such that the final body weights of the females were 7-8 % lower than control weights in both F0 and F1. It is not likely that the maternal nasal toxicity can explain the effects noted on the pups. Likewise, the reduced maternal weight gain does not seem to be of a sufficient magnitude to constitute marked maternal toxicity or to explain the pup effects.

Whether the pup effects were caused by the pre- or postnatal exposure has also been raised, and it is acknowledged that it is always difficult to determine when such effects have been initiated. However, in the two-generation study, F2 pup body weights were reduced already on day 0 (“decreases in body weight…were observed…throughout the pre-weaning period (PND 0-21)”). In the two developmental toxicity studies (where treatment of the dams stopped prior to birth), pup body weights were reduced and developmental landmarks were delayed, occurring later during the pre-weaning phase, indicating that the effects were attributable to the gestational exposure. Placental transfer of styrene has also been shown in mice.

In adult rats, styrene causes ototoxicity (loss of hearing) and toxicity to the nasal epithelium. In humans, styrene causes hearing loss, affects colour vision and long-term exposure may also lead to brain damage (chronic encephalopathy). In rat pups, styrene consistently affects the growth of the pups, resulting in delayed development of the offspring. There are also indications of neurological/neuromuscular deficits in the offspring, and although there are inconsistencies in these data, these effects should be interpreted in the context of the neurotoxic effects of styrene on adult animals. There is evidence of developmental toxicity noted in three different studies.

Overall, the RAC is of the opinion that there is sufficient evidence of developmental effects to warrant classification as Repr. 2, H361d (CLP).

The criteria also state that if the effects are considered to be of low or minimal toxicological significance (e.g., small effects on foetal weights, or small differences in postnatal developmental assessments), classification may not necessarily be the outcome. The types of effects observed in the styrene studies might initially suggest that this is a borderline case for classification, but the RAC considers that the overall pattern of long-lasting developmental delays and neurological/neuromuscular deficits fulfill the requirements for classification with Repr. 2, H361d (CLP). As the criteria for reproductive toxicity classification under DSD are very similar to the CLP criteria, classification with Repr. Cat 3; R63, is warranted according to the DSD.“

We note that the arguments of industry submitted during Public Consultation to RAC concerning the weight of evidence that can be derived from these four endpoints were not adequately taken into consideration nor discussed by RAC. Therefore we want to reiterate these arguments although some of them were already mentioned in the CSR above (for body weight, forelimb grip strength, straight channel swimming trials). First of all, the following general considerations must be taken into consideration:

Point of life span and generation affected: All the effects taken by RAC were only observed in the second generation. In a detailed analysis Piersma et al. (2011) concluded that an evaluation of the F2 offspring will very rarely provide critical information as compared to findings in the F1 generation.

Severity of effects: All the effects taken by RAC are mild and can by no means regarded as severe.

Maternal toxicity with possible effects on maternal care: an assessment in this regard should not solely be based on the crude parameter of statistically significant maternal body weight. The evaluation of the large database for styrene shows that 1. repeated inhalation to 500 ppm styrene produces degeneration of the olfactory epithelium but more subtle histopathological alterations occur already at 50 ppm after 12 months showing that olfactory function is affected already at 50 ppm. 2. prolonged exposure often leads to reduction of body weight already at 200 or 500 ppm. 3. respiratory tract irritation occurs at 200 ppm. 4. mild narcotic effects that may impair maternal care are described at 50 to 300 ppm with a short lasting weight loss during exposure even at 50 ppm. The exposure concentrations in the Cruzan study (i.e. 50, 150 or 500 ppm) are thus within the range known to already lead to other relevant effects that may impair maternal care.

Statistical considerations and number of endpoints investigated: statistically significant effects (at p<0.05) can be expected as chance findings in DNT studies which include measurements of 143 endpoints in 2 genders resulting in a total of 286 datasets (without histopathological investigations, organ weight determinations and interval/trial data).

Functional domains: to overcome the problem of statistical assessment of such a large number of endpoints the concept of functional domains has been developed by Moser (1991). Evaluation of such domains in 500 ppm F2 offspring did not show a consistent pattern, especially not for the neuromuscular domain.

The four endpoints taken by RAC for their classification decision require that much more details must be taken into consideration to come to a well-founded decision:

Body weight effects:

There were numerical reductions in body weights in F1 and F0 dams already at 150 ppm during premating and gestation although not reaching statistical significance.

While body weights of F2 males at 150 ppm (pnd 21) were significantly different from F2 controls they were almost identical to F1 controls.

No statistically significant reductions in the bodyweights of F2 exposed to 150 ppm were observed during the post-weaning phase.

A pattern of body weights effects were seen only in F2 but not in F1 offspring.

The persistence of reduced bodyweight of F2 pups (500 ppm) throughout postweaning may be attributed to the unexceptionally high control F2 weights.

As the body weights of control F2 offspring were clearly higher than those of control F1 offspring the body weight effects of exposed F2 pups may be a chance finding.

Weight reductions observed in parent generations of the 2-generation studies and in various other toxicological investigations indicate that body weight effects noted in offspring may rather be a general toxicological consequence of styrene exposure of parents, not being a specific developmental effect.

Pituitary weight:

Determination of pituitary weights in pnd 21 F2 pups is not required by the test guidelines and historical control data are not available. The high variability of pituitary weights in conjunction with the very low absolute weights (between 0.6 and 10.9 mg for F2 pups) may lead to misinterpretations caused by chance variation.

Significantly reduced absolute or relative pituitary gland weights were only observed in pnd 21 F2 pups but not in pnd 21 or adult F1 pups.

No histopathological alterations were noted in the pituitary of 500 ppm exposed male or female F1 adult animals.

The absolute and relative weights of the exposed F2 groups with statistically significant reductions were comparable to those of F1 control pups of the same age.

Only a few exposed F2 pups fell below the range of the F2 controls.

The UK RAR (2008) concluded: “given the lack of any associated histopathology, it is reasonable to assume that these pup organ weight reductions (including pituitary weight - added) are unlikely to represent adverse developmental effects”

Forelimb grip strength:

Grip-strength shows considerable variation. Of the 500 ppm offspring only 6 of 20 males and 3 of 20 females fell outside the range of concurrent controls and the majority were within normal range.

Grip-strength is influenced by body weight and body weights of 500 ppm animals were lower than controls. But a simple proportional correlation is misleading. Although reductions in body weights on PND 63 were only moderate, during preweaning weight reductions were much more pronounced.

There were no statistically significant effects on fore-limb grip strength on either PND 22 and 45.

There was no statistically significant difference in hind-limb grip-strength although peripheral nerve damage typically leads to more pronounced effects on hind-limb grip-strength.

A significant increase in fore-limb grip strength in females at PND 45 in the 150 ppm group indicated to variability of the effect.

The group mean values for fore-limb grip-strength are within the historical control range supporting the conclusion that the findings reflect normal variability.

There was no underlying histopathology and therefore the difference in grip-strength is unlikely to represent a specific neurological effect.

The UK RAR (2008) concluded: “the reduction in forelimb grip strength observed on PND 60 only is considered to be the consequence of the reduced body weight seen in these pups. Furthermore, as there was no similar difference in hindlimb grip strength at the same time point and no underlying histopathology, it is unlikely that the reduced forelimb grip strength represents a specific neurological effect of styrene”.

Time to escape in straight channel swimming trials:

A “real” increase of straight channel swimming time would mean a generalised impairment of swimming performance that should also affect the total swimming times in the part for learning and memory in the maze test. But this was not the case.

The mean swimming times of the contemporary control animals were low compared to historical control data while at 500 ppm the values were within the historical control range.

The mean swimming times are derived from 4 consecutive trials. The most prominent difference was obtained in trial 1, while swimming times in trials 2-4 were similar over all treatment groups. Therefore, the difference in group mean swimming time reflects the unusual contemporary control value and a chance finding.

The UK RAR (2008) concluded: “Historical control data … show that the swimming time values … were within the historical control ranges and that the observed increase was due to an unusually low value in the concurrent controls. This suggests that this increase in swimming time may actually be an expression of normal variation and have no toxicological significance.”

In summary, if such a detailed analysis is carried out for the four endpoints that were the basis for the RAC (2012) decision and taking into account the general considerations mentioned above, a category 2 classification can hardly be justified.

References

RAC (2012). Opinion proposing harmonised classification and labelling at EU level of styrene; adopted 28 November 2012.

Response of the Styrene Producers Association to the CLH proposal (Sept. 2011) for the classification of styrene as a Cat. 1B reproductive toxicant (developmental effects) according to Regulation (EC) No 1272/2008 (CLP); submitted to ECHA.

Piersma, A. H. et al., Combined retrospective analysis of 498 rat multi-generation reproductive toxicity studies: On the impact of parameters related to F1 mating and F2 offspring. Reprod. Toxicol. (2011), doi: 10.1016/j.reprotox.2010.11.013.

Moser V (1991). Applications of a neurobehavioral screening battery. J. Am. College Toxicol. 10: 661-669.