An Ounce of Prevention: Risk Mitigation in Pharma Development

Posted on Sep 23, 2009

The difficulty of getting oncology products to market was emphasized in a recent New York Times article, along with a good lesson about needlessly risky study execution.

The article emphasized the plight of a small company, Pharmacyclics, whose product was turned down by the FDA after research showed that it failed to work. The failure was attributed by the company’s CEO to a single site in France: “They got a very negative result there, and it really skewed the data.”
 
This example shows a key limitation of the traditional linear (non-adaptive) research model: there are little or no intermediary indications during the study. This black box approach keeps researchers in the dark until the end of the study, then greets them with an all-or-nothing result.  
 
In contrast, adaptive methodology applied to both design and operational aspects of a trial enables sponsors to get an idea of how their product is doing throughout the study, and pinpoint any difficulties with specific sites. Agile research tracks performance continuously during the study, with any suspicious results an indication for a closer look and immediate action if warranted.

In the end, Pharmacyclics didn’t know whether the difficulties they faced truly stemmed from a poor site, or from the drug itself. Variability is inherent in multicenter studies, which is the reason we do them to begin with ¾ but the FDA didn’t buy the bad site argument, and Pharmacyclics was left having to repeat the study.
 
Programs and studies should never be subject to this degree of risk. A powerful benefit of Agile Clinical Development is that it helps avoid this unnecessary risk entirely.