Two mouse models for carcinogenicity testing proposed by the ILSI/HESI program were exposed to cigarette mainstream smoke (MS): the transgenic mouse expressing the human c-Ha-ras proto-oncogene (rasH2) and the heterozygous tumor suppressor p53 knock-out mouse (p53). Female mice (16/group) were exposed to air (SHAM) or MS from the reference cigarette 2R4F for 6 months (3 h/d [low] or 6 h/d [high], 5 d/wk) at a concentration of 240 µg total particulate matter/l. Satellite groups (8-16/group) were kept for 3 months post-exposure. Additional groups of 6 mice were initiated by i.p. administration of urethane (250 mg/kg) before exposure. Body weight gain and lung nodule incidence and multiplicity were assessed as endpoints. Lungs were made transparent (Tellyesniczky fixative) for macroscopic evaluation and a limited number of sections were taken for histopathological analysis of the nodules. Body weight gain of MS-exposed mice was reduced in rasH2 mice only (10%; end of exposure period). There were no statistically significant MS-mediated increases in the incidence and multiplicity of lung nodules in either strain. Incidence is given for sham, low dose and high dose groups, respectively. At the end of the exposure period incidence was 3/16, 3/16 and 5/16 in rasH2 mice and 2/16, 0/16, and 5/16 in p53 mice. At the end of the post-exposure period incidence was 2/15, ─ and 3/16 in rasH2 mice and 5/16, 1/8 and 2/16 in p53 mice. Treatment with urethane resulted in a high nodule incidence with no statistically significant differences vs sham for both mouse strains. In rasH2 mice incidence was 5/6, 6/6 and 3/6 at the end of the exposure period and 6/6, ─ and 4/6 at the end of the post-exposure period. In p53 mice incidence was 0/6, 3/6 and 0/6 at the end of the exposure period and 2/6, 3/8 and 0/6 at the end of the post-exposure period. Histopathological evaluation showed the presence of bronchiolar-alveolar adenomas and carcinomas in rasH2 mice only, irrespective of urethane injection. Under the conditions of this study, the mouse models for carcinogenicity testing, rasH2 and p53, were not suitable for detecting MS-induced lung tumors.