The Dog Days of Drug Reviews?

Over at Forbes, the always thoughtful Matt Herper asks, “Remember when the FDA rejected drugs?

Per Matt, “As recently as 2008, companies filing applications to sell never-before-marketed drugs, which are referred to by the FDA as “new molecular entities,” faced rejection 66% of the time. Yet so far this year the FDA has rejected only three uses for new chemical entities, and approved 25, an approval rate of 89percent.”


(These numbers come from a new analysis commissioned by Forbes from BioMedTracker. The way BioMedTracker follows new molecular entities is slightly different from the way the FDA does. BioMedTracker users want to know about every use of a new medicine. That means that the 2015 rejection count includes rejections of Avycaz, a new antibiotic from Allergan, for hospital-acquired pneumonia, and selling Jardiance, a diabetes drug from Eli Lilly and Boehringer Ingelheim , in combination of metformin. But Avycaz was approved for two other uses and Jardiance is on the market by itself.)

Herper, “… it’s worth sticking to BioMedTracker’s definitions, because it allows us to compare this incredibly high approval rate with the past. And that tells a story of an agency that has been giving the green light more and more often.”

Interesting stuff – but what’s missing is a discussion of how the evolution of regulatory science has impacted the dynamic relationship between the FDA and the innovative pharmaceutical industry and has changed over the course of time (by design, and largely through the mechanism of PDUFA negotiations) the quality of NDAs reaching agency review.

More agency/sponsor meetings earlier in the process not only result in better submissions (more likely to be approved because of higher quality science and more sophisticated protocols), but fewer applications of questionable value . As one senior FDA official told me yesterday, “We’re seeing fewer dogs.”

Another factor that’s important to consider is that failed NDAs are expensive. The following figures are illuminating.

  • A 10 percent improvement in predicting failure before clinical trials could save $100 million in development costs.
  • Shifting 5 percent of clinical failures from Phase III to Phase I reduces out-of-pocket costs by $15 to $20 million.
  • Shifting failures from Phase II to Phase I would reduce out-of-pocket costs by $12 to $21 million.

When he was asked why he was so successful, Thomas Edison replied, “Because I fail faster than everyone else.” One of the reasons the FDA is seeing fewer dogs is because they are helping innovators to recognize failure earlier in the process. And that also means more money that can be reinvested in new clinical programs. Furthermore, a high turndown rate is not a badge of honor — it is indicative of a dysfunctional system.

The good news is that more R&D time, talent, and treasure is being focused on personalized medicine using more sophisticated tools (i.e, biomarkers). Failure is being found sooner, targeted clinical success is easier to predict earlier – and can be expedited through the regulatory process through many new and exciting review pathways (i.e., Breakthrough Designation).

(PS/ Those who don’t think the FDA has “adaptive licensing” opportunities don’t understand what’s going on. And those who choose to blame the FDA for biotech investor anxiety had better find some new excuses.)

Those who wave their arms about the FDA “approving everything” don’t see (or choose not to see) the important success story behind the headline. That dog don’t hunt.


Peter J. Pitts, a former FDA Associate Commissioner, is President of the Center for Medicine in the Public Interest and an Executive Partner at YourEncore.

Morning Consult