PROTOCOL DESIGN CONSIDERATIONS THAT RELATE TO DEMONSTRATING THE SAFETY AND EFFECTIVENESS OF CHEMOPREVENTIVE AGENTS

As with other drugs, applications for marketing approval of new chemop reventive agents in the United States must include data from adequate and well-controlled clinical trials that demonstrate effectiveness and safety for the intended use. Knowledge of a drug's pharmacologic acti ons and metabolism may benefit protocol design, by identifying the pat ient populations and dosing schedules associated with a Favorable risk /benefit profile. With availability of appropriate preclinical data, i ncluding standard assessments of an agent's toxicology, effects on rep roductive performance, and genotoxicity, initial Phase studies of 1-3 months may be performed in normal volunteers or an appropriate higher- risk population. For chronic dosing studies of longer duration, precli nical toxicology studies of longer duration are relevant. Enrollment i n chemoprevention studies should be directed toward individuals at suf ficient risk of developing cancer so that potential benefit may counte rbalance the unpredictable and possibly serious adverse effects that m ay be observed with prolonged administration of a study drug. Phase I and II studies with clinical dosing lasting up to 12 months often affo rd opportunities to assess drug effect on surrogate endpoint biomarker s that may correlate with endpoints of clinical effectiveness. Phase I II and late phase II chemopreventive investigations should routinely u tilize a prospective, randomized study design (double-masked and place bo-controlled, when possible). To support marketing approval, there mu st be evidence that a chemopreventive agent significantly delays or pr events the occurrence of malignancy, with acceptable safety. In some c ircumstances, modulation of a surrogate marker may provide a basis for marketing approval, before more definitive endpoint data become avail able. However, the acceptability of a surrogate depends on the nature and quality of the data supporting its predictive value. Given the con siderations of large study size, long duration, and high cost that may hamper development of potential agents, studies designed to examine t he predictive value of surrogate endpoint biomarkers are of great impo rtance to the future development of chemoprevention research. (C) 1998 Wiley-Liss, Inc.dagger