Lifting the Veil on FDA Subpart H Surrogate Endpoint Approvals

August 27, 2013

By Kurt R. Karst –      

In June 2013, FDA announced the issuance of a draft guidance document, titled “Expedited Programs for Serious Conditions – Drugs and Biologics” (Docket No. FDA-2013-D-0575).  As we previously reported, the draft guidance, as well as a related Manual of Policies and Procedures, provides important insight into FDA’s breakthrough therapy designation program, which was created by the 2012 FDA Safety and Innovation Act (“FDASIA”) (see our summary here).  But the draft guidance does much more than that.  It serves as a de facto desktop reference for FDA’s four expedited programs: fast track designation, breakthrough therapy designation, accelerated approval, and priority review designation. 

Section VII.C. of the draft guidance, titled “Evidentiary Criteria for Accelerated Approval,” describes several factors FDA weighs in assessing whether the available evidence is sufficient to allow the Agency to conclude that a proposed “surrogate endpoint” – i.e., an alternative measurement of the symptoms of a disease or condition that are substituted for measurements of observable clinical symptoms – is “reasonably likely to predict clinical benefit” and thereby constitute the basis for accelerated marketing approval, which is often referred to as a “Subpart H” approval.  This section of the draft guidance has not been the focus of much attention since the draft guidance was published.  But that could change with a recent analysis submitted as a comment to FDA on the draft guidance.

The analysis, conducted by Hyman, Phelps & McNamara, P.C.’s Frank J. Sasinowski and Alexander J. Varond, notes that the importance of Subpart H as a regulatory innovation and vehicle for providing patients suffering with serious and often rare diseases where there is inadequate available therapy has taken on significant added importance with two recent milestones.  First, FDASIA revised the statutory provisions of Subpart H to “facilitate somewhat broader use of accelerated approval to expedite patient access to important treatments for serious conditions[,] . . . provide additional flexibility[,] . . . provide clarification concerning the use of clinical endpoints[,] . . . [and] make clear that FDA has the authority to consider pharmacologic or other evidence . . . in determining whether an endpoint is reasonably likely to predict clinical benefit,” according to FDA.  Second, on September 25, 2012, the President’s Council of Advisors on Science and Technology released a report, titled “Report to the President on Propelling Innovation in Drug Discovery, Development, and Evaluation” (see our previous post here), that addresses the complexities of developing new medicines for Americans and that instructs FDA to expand the use of its Subpart H authority.  

According to the comment, “[t]he linchpin of the FDA Subpart H system was, and is, the surrogate endpoint that is ‘reasonably likely to predict clinical benefit’ (or intermediate clinical endpoint that is ‘reasonably likely to predict ultimate clinical benefit.’).”  But “[t]here have been many misunderstandings of this Subpart H system,” say the comments. 

Some have thought that it meant that the quantum or quality of evidence was somehow reduced, and the statutory requirement of “standard evidence of effectiveness” was in some way, in whole or in part, skirted or deferred.  While this seems not to be the case in statute, regulation or policy, the other extreme is just as likely not to “serve the public well” (quoting the Presidential Report at p. 59).  The other extreme is the view that unless the surrogate is validated, it cannot be relied upon in a Subpart H approval decision.  This is sometimes found in reviews that conclude that the Sponsor’s evidence failed to satisfy the standard of approval because the trial(s) attempted prove both the drug’s effect on the surrogate as well as on the clinical benefit and the clinical benefit showing was not robust enough to validate the drug’s effect on the surrogate.

Between these two extremes, there has existed a gaping hole that has begged to be addressed for nearly 3 decades and that is – what is the regulatory and evidentiary foundation for FDA’s determination that an unvalidated surrogate is capable of supporting a Subpart H approval.  Now the FDA in its June 2013 Draft Guidance has tackled this and laid out clearly discrete principles and factors.

With that backdrop, the analysis looks at each of the 19 Subpart H approvals (that are not for AIDS or cancer) in order to discern the types and patterns of evidence that FDA has found adequate to be the foundation for those Subpart H approval precedents from 1992, when the accelerated approval regulations were promulgated, to the present.  Each of the 19 Subpart H approval precedents is “scored” based on the factors laid out in Section VII.C. of FDA’s draft guidance according to the weights and scoring of the comment authors.  (The authors’ methods for conducting the analysis are laid out in the comment.) 

In the end, the comment authors conclude that FDA has shown great flexibility in applying its Subpart H standards to therapies under its review.  For example, “although most of the time a clear understanding of the pathophysiology of the disease process will facilitate access to reliance upon a surrogate, the absence of a complete understanding of the disease process or even the existence of a relatively weak understanding of the disease process is not, in and of itself, incompatible with Subpart H,” write the comment authors. 

FDA’s penchant for flexible application of evidentiary standards for drug approval not only exists with respect to Subpart H approvals, but in other contexts as well.  As we previously reported, a 2011 report authored by Mr. Sasinowski concluded that “two of every three orphan drugs approved show FDA’s historic flexibility in its review of effectiveness data on orphan drug therapies.”