A comparative toxicity and carcinogenicity study was carried out for 2 years with estazolam, a benzodiazepine, via diet in Sprague-Dawley rats (0.5, 2, and 10 mg/kg/day) and in B6C3F1 mice (0.8, 3, and 10 mg/kg/day). In rats, no biologically significant changes were seen with respect to mortality, clinical signs, food consumption, or occurrence of palpable masses. Body weight gain in females (10 mg/kg/day) was depressed 12.6% and reflected a maximum-tolerated dosage (females). Spontaneous and incidental nonneoplastic lesions were consistent with aging in this species and unrelated to drug treatment. No biologically significant differences in tumor incidences occurred. Mice were more responsive to estazolam as suggested by (1) increased mortality (males) at 10 mg/kg/day, (2) increased food consumption and body weight gains (females), (3) withdrawal signs characterized by hyperactivity/aggressiveness and convulsions, and (4) appearance of dose-related nodular hyperplasia of the liver due to the relatively high dosages used coupled with the propensity of benzodiazepines to enhance liver enzyme induction. Several spontaneous benign and malignant tumors observed in all groups were not considered to be drug related. Based on the findings in these studies, estazolam was not considered to be carcinogenic when administered via diet to either rats at 0.5, 2, and 10 mg/kg/day or to mice at 0.8, 3, and 10 mg/kg/day for 2 consecutive years.