Volume 29, Issue 51 pp. 9411-9572 (28 November 2011)
Failure-to-success ratios, transition probabilities and phase lengths for prophylactic vaccines versus other pharmaceuticals in the development pipeline
Matthew M. Davis, Amy T. Butchart, John R.C. Wheeler, Margaret S. Coleman, Dianne C. Singer, Gary L. Freed
Research and development of prophylactic vaccines carries a high risk of failure. In the past, industry experts have asserted that vaccines are riskier to produce than other pharmaceuticals. This assertion has not been critically examined. We assessed outcomes in pharmaceutical research and development from 1995 to 2011, using a global pharmaceutical database to identify prophylactic vaccines versus other pharmaceuticals in preclinical, Phase I, Phase II, or Phase III stages of development. Over 16 years of follow-up for 4367 products (132 prophylactic vaccines; 4235 other pharmaceuticals), we determined the failure-to-success ratios for prophylactic vaccines versus all other products. The overall ratio of failures to successes for prophylactic vaccines for the 1995 cohort over 16 years of follow-up was 8.3 (116/14) versus 7.7 (3650/475) for other pharmaceuticals. The probability of advancing through the development pipeline at each point was not significantly different for prophylactic vaccines than for other pharmaceuticals. Phase length was significantly longer for prophylactic vaccines than other pharmaceuticals for preclinical development (3.70 years vs 2.80 years; p < .0001), but was equivalent for all 3 human clinical trial phases between the two groups. We conclude that failure rates, phase transition probabilities, and most phase lengths for prophylactic vaccines are not significantly different from those of other pharmaceutical products, which may partially explain rapidly growing interest in prophylactic vaccines among major pharmaceutical manufacturers.