The design and analysis of cancer vaccine studies represent unique challenges from an industry as well as regulatory perspective. Many studies have been initiated but few have succeeded in confirming clinical benefit. Cancer vaccines have not generally been able to induce clinical tumor shrinkage and therefore traditional endpoints such as tumor response may not be helpful in demonstrating antitumor activity. Overall survival is the gold standard for demonstration of clinical benefit in oncology but studies designed to demonstrate an effect on survival may take several years to complete. Some authors have suggested that a redefinition of traditional oncology response criteria may provide better support for initiation of confirmatory studies or even support licensure. FDA is not convinced that this approach will be helpful in improving the process of identifying better products suitable for licensure. FDA convened a workshop in 2009 to discuss lessons learned from completed cancer vaccine studies. The focus was on optimal design of early phase studies on which to base confirmatory studies. Many of the unsuccessful completed Phase 3 studies were initiated based on results from smaller single arm studies with comparisons to historical controls which likely confounded interpretation. FDA encourages sponsors to consider initiation of randomized phase 2 studies with time to event endpoints in order to provide at least some preliminary evidence of clinical activity prior to initiation of Phase 3 studies. In addition it may be feasible to develop innovative adaptive study designs in order to identify which cancer vaccines are worthy of further clinical development and to facilitate rapid initiation of confirmatory studies in promising products. FDA has published a draft Guidance for Industry, Clinical Considerations for Therapeutic Cancer Vaccines which attempted to address some of these challenges.