In Draft Guidance on Acute Migraine Treatments, FDA Proposes Novel Efficacy Endpoint

October 23, 2014

By Etan J. Yeshua

Drug developers targeting acute migraines may now have a new path to approval.  In a draft guidance document issued on Tuesday by the Division of Neurology Products ("DNP"), FDA offers a novel primary endpoint that has not been used before to support migraine drug approvals (primarily the triptan class of drugs).  The guidance document also addresses safety considerations, class labeling, and other study design issues.

Migraines are characterized by more than one symptom (i.e., headache, nausea, photophobia, and phonophobia) and approved treatments, at the insistence of DNP, have traditionally demonstrated efficacy on all four of these as co-primary endpoints.  In the past several years, however, DNP has taken the view that a demonstration of efficacy on headache and only one additional co-primary endpoint – nausea – could be sufficient.  Now, DNP has apparently recognized that this paradigm, while lessening the regulatory burden from four endpoints to two, suggests (perhaps inappropriately) that nausea is a more important symptom than photophobia and phonophobia.  DNP has now opened the door to an alternative approach which appears to recognize that not all migraine patients suffer equally from the same secondary symptoms.  As explained in the guidance document, FDA would still require two co-primary endpoints: (1) headache, and (2) an effect on “the [other] most bothersome migraine-associated symptom” as prespecified by each individual subject.  In other words, FDA is considering “having patients prospectively identify their most bothersome migraine-associated symptom in addition to pain,” and efficacy would have to be demonstrated on headache pain and on “the most bothersome associated symptom” – whatever that may be for a given patient.  This new approach could be an effort by FDA to recognize the varied presentations of migraine symptoms as part of the agency’s recent emphasis on “patient-focused drug development.”

The extent to which this alternative approach actually lowers the bar to approval, however, is unclear.  Although DNP is reducing the number of efficacy endpoints necessary for a new migraine therapy to obtain approval, Tuesday’s guidance document leaves out some important details, which counsels toward thorough communication with DNP early in the development process.  For example, the guidance states that, in addition to the two co-primary endpoints, “all . . . migraine-associated symptoms (i.e., nausea, photophobia, and phonophobia) should be assessed as secondary endpoints,” but it does not specify whether and/or to what extent a drug candidate must demonstrate efficacy on one or more of these secondary endpoints.  While pivotal studies are typically not required to demonstrate statistical separation from placebo on secondary endpoints, the history of FDA’s required showing for triptans – efficacy across four endpoints whether they were itemized as co-primary, or primary and key secondary – suggests that the prudent drug developer obtain program-specific clarity on this point from DNP at or before the end of phase 2. 

In addition, the guidance document does not address whether FDA will require studies to enroll some minimum number or proportion of subjects who identify each of the three associated symptoms as the “most bothersome.”  Given that DNP has previously suggested that nausea is the most significant non-headache endpoint, it remains unclear whether a study in which the majority of subjects list, say, phonophobia as their most bothersome symptom, with success on that endpoint but little or no effect on secondary endpoints of nausea or phonophobia, would be sufficient for approval in acute migraine.  Finally, while randomization should assure a balanced distribution of most bothersome symptoms across treatment groups, it is unclear whether DNP would approve a drug that showed a benefit on the co-primary endpoint when, in fact, the results were driven by different identified symptoms in the placebo and treatment groups.

FDA asked that comments about the draft guidance be submitted by December 22, 2014.  Other topics discussed in the document include trial population inclusion criteria, dose selection, concomitant medications, additional secondary endpoints, frequency of data collection, statistical issues, safety considerations, pediatric studies, and labeling.