Success Rates In Clinical Trials: Methodology

The Clinical Development Success Rates 2006–2015 study aimed to measure clinical development success rates to strengthen benchmarking metrics for drug development. Success rates for investigational drugs were measured by analyzing individual drug program phase transitions from January 1, 2006, to December 31, 2015. For the 10 years studied, 9,985 transitions in Informa's Pharma Intelligence's Biomedtracker database were analyzed. A phase transition is the movement out of a clinical phase – for example, advancing from Phase I to Phase II development, or being suspended after completion of Phase I development.

These transitions occurred in 7,455 clinical drug development programs, making this the largest study of its kind. With this broad set of data, the goal was to capture the diversity in drug development across levels of novelty, molecular modalities and disease indications. As the study cuts across 1,103 companies, differences in clinical team experience, from both large and small companies, is taken into consideration.

Only company-sponsored, FDA registration-enabling development programs were considered; investigator-sponsored studies and combinations with other investigational drugs were excluded from this analysis. Generic products were not included, but generic manufacturers developing novel investigational drugs were represented.

Individual phase success rates were determined by dividing the number that advanced to the next phase by the total number advanced and suspended.

One of the key measures of success used in this report is the likelihood of approval (LOA) from Phase I. This LOA success rate is simply a multiplication of all four phase success rates, a compounded probability calculation. For example, if each phase had a 50% chance of success, then the LOA from Phase I would be 0.5 x 0.5 x 0.5 x 0.5 = 6.25%.

The data for drug program transitions used for this study were extracted from Biomedtracker using a Probability of Technical Success (PTS) tool, which identified all "Advanced" and "Suspended" drugs by development phase from January 1, 2006, to December 31, 2015. Biomedtracker tracks the clinical development and regulatory history of investigational drugs to assess their LOA by the FDA. Biomedtracker is populated in near real-time with updated information from press releases, corporate earnings calls, investor and medical meetings and numerous other sources. These data are recorded in Biomedtracker and tagged with a date. Biomedtracker also uses other sources, including regular communication with companies conducting clinical trials, to ensure accuracy and timeliness of the data.

Selection biomarkers are gene products used as inclusion or exclusion criteria for enrolling patients into clinical studies. The biomarker subject selection data used for this study were extracted from BiomarkerBase, which identifies selection biomarkers described in clinical trials posted at ClinicalTrials.gov. BiomarkerBase is a subscription-based service from Amplion that tracks selection biomarker usage in clinical trials, drug labels and tests (including laboratory-developed, FDA-cleared and FDA-approved tests). BiomarkerBase is updated weekly with information from these sources and publications, using supervised machine-learning algorithms for natural language processing (Amplion BiomarkerEngine) to identify selection biomarkers. BiomarkerBase includes primarily human genes and proteins. Other diagnostic measurements, including clinical blood chemistry, liver enzymes, white blood cell count, heart rate, blood oxygenation, blood glucose and clotting times, etc., are not included in BiomarkerBase.

This is an edited version of the methodology for the full report, "Clinical Development Success Rates 2006-2015," published in May 2016 and authored by David W. Thomas, Justin Burns, John Audette, Adam Carroll, Corey Dow-Hygelund and Michael Hay.