Formal Dispute Resolution: A Different Perspective on Wins and LossesOctober 9, 2018
In determining whether to submit a Formal Dispute Resolution Request (FDRR) appealing a Complete Response Letter (CRL) or other adverse action from FDA’s Office of New Drugs (OND), applicants necessarily consider how often such appeals are successful. The most cited relevant statistics were published by FDA in 2016. The analysis reported that, during the time period from 2003 to 2014, “[o]f the 140 unique appeals accepted and reviewed [131 of which were in OND], CDER granted 23 (16%) appeals and denied 117 (84%).” Eighty-four percent denied! Why would anyone appeal? Because “Denial,” like beauty, is in the eye of the beholder. In our experience, “Denials” often conceal important “Wins” in the sense of establishing a more favorable path to approval than originally presented by the review division.
For example, consider three distinct cases with highly similar outcomes that we were involved in: The first was an sNDA for a drug intended to treat a life-threatening condition; the second an NDA for an NCE that would be the first drug approved for what turned out to be a controversial symptomatic indication; and the third a 505(b)(2) NDA for an improved version of an existing therapy. For the first case, the sNDA, the applicant appealed the review division’s issuance of a second CRL which maintained that the applicant needed to conduct one or more additional adequate and well-controlled studies to demonstrate efficacy and safety of the drug. On appeal, OND technically “Denied” the client’s request but rather than require additional Phase 3 work, OND stated that it believed an advisory committee should review the data to determine if the benefits outweighed the risks. The drug was eventually approved with no additional Phase 3 data after a positive vote at the committee meeting. FDA’s published analysis tallied this as two “Denied” appeals while we count it as a single “Win.” FDA counts it as two appeals because the CRL was issued by the review division rather than the immediate office, therefore requiring two rounds of review before arriving at OND. We had alerted the client to the likelihood that the issues under appeal would require OND review and budgeted our time and resources accordingly. As such, we and the client considered the process as a single appeal.
In the second case, the CRL (which required an additional Phase 3 study) was issued at the immediate office level because the drug was an NCE. The FDRR therefore went directly to OND. As in the example above, OND “Denied” the appeal, but recommended resubmission with no new Phase 3 data and the convening of an advisory committee. Again, the vote was in favor of approval and the drug was approved with no additional Phase 3 work. The FDA published analysis considers this case a single “Denial,” while we consider it a “Win.”
In the third case, the 505(b)(2) application, OND, during a second round appeal, disagreed with both the applicant and the review division on the method for handling missing data. OND “Denied” the appeal and suggested a third statistical method of analysis. Application of OND’s own preferred method resulted in a statistically significant result and approval of the drug with no new studies. As above, FDA would count this as two “Denials,” while we see it as a single “Win.”
In our practice, these experiences are not out of the ordinary. We seek to come to a mutual understanding with the client of what an appeal “Win” might look like, the likelihood of obtaining such a “Win,” and how many rounds of review might be needed. To the extent that the achievable “Wins” are either too modest or too remote, we would not advise submission of a FDRR. In our experience then, the FDRRs that we pursue are the subset most likely to result in a palatable path forward regardless of the “Granted” or “Denied” designation.
Our most recent history tracks this pattern. From 2015 through 2018, we assisted with at least nine products that completed the FDRR process. We believe FDA would count the FDRRs for these nine products as twelve appeals, with three appeals “Granted” and nine “Denied” (due to counting each appeal level as a unique FDRR). By our count, two products were not offered any improved path forward, which we would count as losses, while seven products had “Wins” in that the applicant was provided with an alternative, more favorable path forward to approval than what was originally presented by the review division. This difference is important when considering the FDA reported statistics. By FDA’s count, our three-year track record would be 25% Granted versus 75% Denied. By our clients’ count, there were 78% Wins versus 22% Losses. We like those odds much better.