A multihit, multistage model of chemical carcinogenesis.

Carcinogenesis involves the accumulation of genetic changes within a single cell. Tumor promotion functions in the initial clonal expansion of an initiated cell but is generally not considered to influence later stages. To investigate whether tumor promotion can influence later stages of carcinogenesis we developed a two-hit 7, 12-dimethylbenz[a]anthracene (D) protocol designed to enrich for keratinocytes that contain at least two D-induced genetic alterations. FVB/N mice were initiated with D and promoted with 12-O-tetradecanoylphorbol-13-acetate (T) or treated with acetone (A) vehicle for 6 weeks. At 7 weeks after the start of promotion, but before visible papilloma development, groups of mice were treated with a second dose of D or A and 1 week later T promotion was resumed. D/T/A/T mice developed 2.8 papillomas/mouse and D/A/D/T mice demonstrated an additive tumor response and developed 5.8 papillomas/mouse. Importantly, D/T/D/T mice developed 12.4 papillomas/mouse, thereby demonstrating a synergistic tumor response compared with D/A/D/T and D/T/A/T mice. D/T/D/T papillomas exhibited increases in suprabasal S phase cells and keratin 13 expression when compared with D/T/A/T papillomas. D/T/D/T mice developed squamous cell carcinomas (SCCs) 10 weeks earlier than D/T/A/T mice and demonstrated a 96% malignancy incidence and 1.71 SCC/mouse compared with D/T/A/T mice, which demonstrated a 28% malignancy incidence and 0.32 SCC/mouse. Greater than 90% of D/T/A/T and D/T/D/T papillomas and SCCs contained mutant Ha-ras, while a normal Ha-ras allele persisted in all cases, indicating that a gene other than the remaining normal allele of Ha-ras was a target gene for the second D hit. These data demonstrate that: (i) promotion between the first and second hits has a profound outcome on carcinogenesis, presumably by increasing the probability that a second hit will occur in a previously initiated cell; (ii) continued promotion after the second hit is required for full expression of malignancy; (iii) the classic initiation-promotion protocol can be extended to a multihit, multistage model.