The objective of this study was to examine the effect of retinol and isotretinoin on the incidence of nonmelanoma skin cancer in high-risk subjects. A total of 525 participants with a history of at least four basal cell carcinomas (BCCs) and/or cutaneous squamous cell carcinomas (SCCs) were entered into a randomized, double-blind, placebo-controlled trial, performed in free-standing study clinics. Participants were randomly assigned to receive oral retinol (25,000 units), isotretinoin (5-10 mg), or placebo supplementation daily for 3 years. The time to first new occurrence of BCC or cutaneous SCC was used as the outcome measure. During the study period, 319 BCCs and 125 cutaneous SCCs were diagnosed clinically and pathologically. There were no differences between those who received retinol, isotretinoin, or the placebo, with regard to the time to first occurrence or to the total number of tumors noted. No beneficial effects were noted with regard to the prevention of nonmelanoma skin cancer with either retinol or isotretinoin.