GAO Analysis Says FDA is Meeting PDUFA Commitments

It’s no secret that FDA’s review and approval metrics are closely watched.  In fact, these metrics are integral to evaluating whether FDA has met the commitment to industry it made during Prescription Drug User Fee Act (“PDUFA”) negotiations.  As explained in our summary of the 2017 reauthorization of PDUFA, for fiscal years 2018-2022, FDA has committed to reviewing 90 percent of new molecular entity (“NME”) NDAs in 10 months and 90 percent of new priority NDAs in 6 months of the filing date (60 days after actual submission); for non-NMEs, the review goals are 10 months and 6 months after submission, respectively.  At the direction of Congress, the GAO just released a new report evaluating exactly how FDA is doing in meeting these goals.

The bottom line: Application review times largely reflect agency goals.  The GAO reached this conclusion based on its analysis of 637 NDAs and BLAs submitted to the Center for Drug Evaluation and Research throughout fiscal years 2014 and 2018.  Oddly, the GAO Report does not explain much about its conclusion that “review times largely reflect agency goals,” mentioning only in a footnote that “FDA completed its initial review of 97 percent of NDAs within the time frames established under its PDUFA goals—a greater percentage than the 90 percent goal stated in its PDUFA V and VI commitment letters.”  There are, however, some exceptions – though fewer than one might think.  The GAO identified only 5 NDAs for which the review time was exceptionally long in comparison to the PDUFA goal date.  GAO excluded these NDAs from its some of its analyses based on FDA’s explanation that “these reviews were substantially delayed because of complicated manufacturing site issues, complicated legal and regulatory issues, or emerging public health issues requiring last minute advisory committee meetings” – conditions that GAO “deemed sufficiently unusual to exclude these five NDAs from further statistical analyses of review times.”

Much of the GAO Report’s focus is on the differences between applications that may impact goal dates.  The Report explains that goal dates are directly related to four key features of NDAs: whether the NDA receives priority review designation; whether the NDA involves an NME; whether the applicant submitted a major amendment; and whether the NDA qualifies for one of FDA’s expedited programs (such as Accelerated approval, Breakthrough designation, or Fast track designation, all of which are available for NDAs for drugs intended to treat serious or life-threatening conditions).  Thirty-two percent of the 637 NDAs included in the GAO review had a priority review designation; 36 percent involved a new molecular entity; 12 percent involved a major amendment; and 18 percent qualified for one expedited program while 9 percent qualified for two or three.   These key features, the Report explains, directly affect the timing of review because they dictate the PDUFA goal date.

The GAO Report also explains that the features that affect goal dates impact some review divisions more than others.  As such, the Report examines the distribution of NDAs with these key features across review divisions and analyzes the differences in time taken to complete initial reviews between divisions.  The analysis found that each division differed in the average number of days to complete an initial review of an NDA, but these differences reflected the differences in the key features of the NDAs (i.e. priority review, expedited programs, major amendments, and NME status) submitted to each division.  Unsurprisingly given the goal date commitments, NDAs with key features that resulted in shorter goal dates had shorter review times.  Controlling for these differences in goal dates, the GAO Report found that most of the divisions’ average review times were similar to (within 2 weeks of) each other.  However, the Hematology and Oncology divisions reviewed about 2 or 3 weeks faster than other divisions.

The Report includes a breakdown of the applications submitted to each division with details about the number of applications with each key feature.  FDA’s Oncology and Hematology divisions reviewed the greatest number of NDAs—while still reviewing faster than other divisions—closely followed by the Antiviral Division, the Anti-infective Division, the Metabolism and Endocrinology Division, and the Anesthesia, Analgesia, and Addiction Division.  The distribution of NDAs with key features varied significantly across divisions, which led to differences in the review time for each division as well.  For example, fifty-six percent of the Anti-Infective Division’s NDAs had priority designation, mandating a faster review time, while thirty-seven percent of the Gastroenterology and Inborn Errors Division’s NDA were priority.  With this difference in priority workload, the median time taken to complete an initial review of an NDA by the Anti-infective division was about two months faster than the Gastroenterology and Inborn Errors division.

Finally, in response to a request from Congress, the Report also details actions FDA has recently taken to evaluate and facilitate the use of different sources of evidence to support NDAs.  As a result of the 2016 Cures Act, FDA started implementing initiatives, including the Real-World Evidence Program, Patient-Focused Drug Development, Complex Innovative Trial Designs, Drug Development Tool Qualification Programs, and Model-Informed Drug Development.  Because the amount and nature of the evidence needed can be an important determinant of when and whether new therapies become available to the public, these programs can impact review times.  While implementation is still in progress for all of the initiatives—meaning only minimal data was available for GAO review—the GAO did note that there has been an increase in discussions with the Agency relating to these new initiatives.

The GAO Report included four appendices of interesting data.  For those of you that have the time while social distancing, it could make for some noteworthy quarantine reading.