A mechanistic interpretation of the oncogenicity of chlorothalonil in rodents and an assessment of human relevance.

Chronic dietary treatment of rodents with the fungicide chlorothalonil causes an increased incidence of papillomas and carcinomas of the forestomach squamous epithelium (rats and mice, both sexes) and adenomas and carcinomas of the renal proximal tubule epithelium (rats, both sexes; mice, males only); the product elicits no tumorigenic response in dogs. As a result, chlorothalonil is classified by EPA as a Group B2 "probable human carcinogen." However, chlorothalonil is not genotoxic and there is strong evidence that both the forestomach and renal tumors observed in rodents result from cytotoxicity followed by compensatory cell proliferation and hyperplasia. In the case of the forestomach, cytotoxicity results from sustained irritation of the squamous epithelium by chlorothalonil leading to inflammation, ulceration, and restorative hyperplasia. Cytotoxicity in the renal tubular epithelium is associated with formation of di- and trithiols that arise through the action of renal beta-lyase on cysteine S-conjugates derived from the corresponding glutathione conjugates of chlorothalonil. Renal cytotoxicity and cell necrosis in rodents result from the ability of the di- and trithiols to inhibit kidney mitochondrial respiration and disrupt cellular integrity. There is strong evidence that this mechanism is not operative in other species such as dogs and monkeys. The progression from cytotoxicity to hyperplasia to neoplasia is becoming increasingly well-recognized as a threshold-based mechanism of carcinogenesis. Unless exposure is excessively prolonged or intense, the cytotoxic effects will be fully reversible. Furthermore, the effects observed in rodents are not appropriate for evaluating the potential human cancer risk from chlorothalonil. Humans do not possess an organ equivalent to the rodent forestomach and the rat is a poor model for evaluating potential human risk for the renal tumorigenicity of chlorothalonil. Humans are likely to be very much less sensitive than rats to the nephrotoxic effects of chlorothalonil. In view of the fact that the tumorigenic effects of chlorothalonil are mediated through a well-understood, nongenotoxic, threshold-based mechanism of little or no relevance to humans, chlorothalonil should be a prime candidate for re-review under EPA's new risk assessment guidelines. Expert committees in both Europe and Canada have concluded that human risks to chlorothalonil should be evaluated by means of the NOEL/safety factor approach usually employed for noncarcinogenic materials.