Genetic mutations and alterations in cell proliferation and apoptosis in UV-induced skin cancer: Application of biomarkers to human chemoprevention studies

Abstract

Human skin is continually subjected to UV-irradiation with p53 playing a pivotal role in repair of DNA damage and apoptosis. Mutations in p53 and p53 overexpression occur early in UV-induced skin carcinogenesis. We studied squamous cell carcinomas (SCC), actinic keratoses (AK), samples adjacent to AK (sun-damaged), and normal skin for p53 mutation, p53 immunostaining, apoptosis, and proliferation. The frequency of p53 mutation, p53 overexpression, and the number of apoptotic cells increased from normal skin, to sun-damaged skin, to AK, while AK and SCC had similar frequencies. Cell proliferation was significantly increased in the progression from normal skin to SCC. These data imply that apoptosis in samples with a high frequency of p53 mutation may not be p53-dependent. There may be two mechanisms for apoptosis. One is in response to DNA damage and the other in response to increased proliferation. It would be of great benefit to reduce the incidence of SCC and its precursor lesion, AK, through cancer chemoprevention. We applied surrogate endpoint biomarkers (SEBs) to skin cancer chemoprevention trials in subjects with multiple AK lesions. The primary SEB in these studies was reduction in the number of premalignant AK lesions, while secondary SEBs included p53 mutation frequency and exon distribution, p53 protein expression, and PCNA. An RXR-specific retinoid analog, targretin, demonstrated no significant reductions in SEBs, although there was a trend toward reductions in the number of AKs, PCNA, and p53 expression. This was a small study lacking power to detect a significant change. The ODC inhibitor, DFMO on the other hand, demonstrated significant reductions in the number of AKs, p53 expression, and although there was no reduction in the frequency of p53 mutations, there was a shift in exon distribution. Inhibition of polyamine synthesis in skin was determined in the DFMO study as a measure of drug effect. We conclude that measurement of SEBs in skin cancer chemoprevention studies is feasible, but it is imperative to choose SEBs that are appropriate to the chemopreventive agent and that have the potential for modulation. Sample size must have sufficient statistical power to detect a response to a chemopreventive agent.