Non-melanoma skin cancer event rates in a formalized clinical trial setting: considerations for clinical trial design

Abstract

Background Here we report clinical risk factors and event rates for the development of new non-melanoma skin cancer (NMSC) in a randomized, double-blind, placebo-controlled trial of the irreversible ornithine decarboxylase (ODC) inhibitor, difluromethylornithine (DFMO), over a 3–5-year follow-up. Methods 147 placebo patients (white; mean age 60.2 years; 60% male) were evaluated for event rates and association of initial skin biomarkers and baseline patient characteristics with the development of squamous cell (SCC) and basal cell (BCC) carcinomas. Results Post-study evaluation (median follow-up 4.4 years) indicates the measures of prior NMSCs (P ≤ 0.001), prior BCCs (P ≤ 0.001), prior SCCs (P = 0.011), prior tumor rate (P = 0.002), hemoglobin (P = 0.022), and gender (P = 0.045) as significant predictors for new NMSC development. Similarly, all measures of prior BCCs and NMSCs (P < 0.001), prior tumor rate (P = 0.014), and SCCs in the prior 2 years (P = 0.047) were statistically significant predictors for new BCC development. Total prior NMSCs and those in the prior 5 years (P < 0.001), total prior SCCs and those in the prior 5 years (P < 0.001), total prior BCCs and those in the prior 5 years (P ≤ 0.001), prior tumor rate (P = 0.011) as well as age (P = 0.008), hemoglobin (P = 0.002), and gender (P = 0.003) were statistically significant predictors of new SCC development. TPA-induced ODC activity at baseline showed no statistically significant association with the development of new NMSC (P = 0.35), new BCCs (P = 0.62), or new SCCs (P = 0.25). Conclusion In the studied population, the history of and rate at which prior NMSCs occur are predictive and should be controlled for in future NMSC prevention trials.