EMS in Viracept--initial ('traditional') assessment of risk to patients based on linear dose response relations

Prior to having performed in depth toxicological, genotoxicological and DMPK studies on ethyl methanesulfonate (EMS) providing solid evidence for a thresholded dose response relationship, we had prepared and shared with regulatory authorities a preliminary risk estimate based on standard linear dose-effect projections. We estimated that maximal lifetime cancer risk was in the order of 10(-3) (for lifetime ingestion of the maximally contaminated tablets) or 10(-4) for the exposure lasting for 3 months. This estimate was based on a lifetime cancer study with methyl methanesulfonate (MMS; as insufficient data were available for EMS) in rodents and default linear back extrapolation. Analogous estimates were made specifically for breast cancer based on short term tumorigenicity studies with EMS in rats, for the induction of heritable mutations based on specific locus and dominant lethal tests in mice and for the induction of birth defects based on teratogenicity studies in mice. We concluded that even under worst case assumptions of linear dose relations the chance of experiencing these adverse effects would be very small, comprising at most a minute additional burden among the background incidence of the patients.