Weight-of-the-evidence review of acrylonitrile reproductive and developmental toxicity studies

Abstract

Risk assessment of acrylonitrile (AN) toxicity to humans has focused on potential carcinogenicity and acute toxicity. Epidemiological studies from China reported reproductive and developmental effects in AN workers, including infertility, birth defects, and spontaneous abortions. A weight-of-the-evidence (WoE) evaluation of the AN database assessed study strength, characterized toxicity, and identified no-observed-adverse-effect levels (NOAELs). The epidemiological studies do not demonstrate causality and are not sufficiently robust to be used for risk assessment. Rodent developmental studies showed fetotoxicity and malformations at maternally toxic levels; there was no unique developmental susceptibility. NOAELs for oral and inhalation exposures were 10 mg/kg/day and 12 ppm (6 h/day), respectively. Drinking-water and inhalation reproductive toxicity studies showed no clear effects on reproductive performance or fertility. Maternally toxic concentrations caused decreased pup growth. The drinking-water reproductive NOAEL was 100 ppm (moderate confidence due to study limitations). The inhalation exposure reproductive and neonatal toxicity high confidence NOAEL was 45 ppm (first generation 90 ppm) (6 h/day). The inhalation reproductive toxicity study provides the most robust data for risk assessment. Based on the WoE evaluation, AN is not expected to be a developmental or reproductive toxicant in the absence of significant maternal toxicity.

Keywords::

• Acrylonitrile
• reproduction
• derelopment
• toxicity

Acknowledgments

Funding for this review was provided by The Acrylonitrile Group, Inc., the European Acrylonitrile Producers Association, and the Japan Acrylonitrile Producers Association.

B.H. Neal and J.C. Lamb were retained by the US acrylonitrile industry association to develop the weight-of-the-evidence review of the reproductive and developmental toxicity of acrylonitrile, and have provided consulting services to the association. D.E. Strother was previously employed by a manufacturer of acrylonitrile, chaired the Health Sciences Committee of the US acrylonitrile industiy association, and currently serves in a consulting capacity to the association. J.J. Collins reports no conflict of interest. The authors alone are responsible for the content and writing of the paper.

Notes

¹An additional drinking-water reproductive toxicity assessment was conducted by Svirbely and Floyd in 1964. The original report is unavailable but the study was summarized in an unpublished review by Kapp et al. (1998). Per Kapp et al., the Svirbely and Floyd study was terminated early and did not proceed to the second generation. It reportedly showed decreased fertility, gestation, and viability indices at 500 ppm, and growth retardation of the F1 pups at 500 ppm. The no-observed-adverse-effect level (NOAEL) reported was 100 ppm. Further details are not available; however, the Litton Bionetics three-generation reproductive toxicity study (Friedman and Beliles, 2002) was designed to attempt to replicate the study design and clarify these findings, as the Svirbely and Floyd study was considered inadequate for evaluation. The Svirbely and Floyd study is not discussed further in this review.

²This study was conducted by WIL Research Laboratories and comprehensively reported by Nemec et al. (2008). This WoE review relied on data from both the unpublished study report and the published paper.

³This study was performed by Litton Bionetics and results published by Friedman and Beliles. This WoE review relied on both the unpublished study report and the published paper.

⁴This study was conducted by the Toxicological Research Laboratory, and the results subsequently reported by Murray et al. (1978). This WoE review relied upon data both from the unpublished study report and the published paper.

⁵This study was conducted by the Toxicological Research Laboratory, and the results subsequently reported by Murray et al. (1978). This WoE review relied upon data both from the unpublished study report and the published paper.