What FDA’s Newest Gene Therapy Approval Tells Us About Durability: How Long is Long Enough?

December 8, 2022By Mark A. Tobolowsky & James E. Valentine

On November 22, 2022, FDA approved CSL Behring’s BLA for Hemgenix (etranacogene dezaparvovec), an AAV-based gene therapy for the treatment of adults with Hemophilia B who currently use Factor IX prophylaxis therapy, have current or historical life-threatening hemorrhage, or have repeated, serious spontaneous bleeding episodes.  This product was developed by uniQure prior to being licensed to CSL Behring.  This approval marked the 5th  viral vector gene replacement therapy product for genetic disorders (referred to in this post as “gene therapies” for simplicity’s sake), and the 3rd this year, approved by FDA following, in order, Luxturna (voretigene neparvovec), Zolgensma (onasemnogene abeparvovec) (see previous coverage here), Zynteglo (betibeglogene autotemcel), and Skysona (elivaldogene autotemcel).

As demonstrated by the few gene therapies approved by FDA, we are still in the early days of the development and evaluation of gene therapy benefits and risks in the United States.  In simple terms from a non-scientist, these gene therapies add functional versions of genes to patients that do not have them, and thus the idea is to have these genes function in the same way as they do in individuals without these disorders.

At times, there has been speculation that these can be actual “cures,” and at some point, they might well be.  However, there are many treatment goals that patients and their caregivers have that fall short of a cure, not only improvements but, for progressive conditions, a slowing or halting of worsening.  It takes many years to determine if something provides durable treatment and to what degree it may fade over time.  Thus, the current reality of gene therapy products presents a conundrum to FDA: If you have an efficacious one-time treatment, how long do you wait to see if the treatment stops working or stops working as well, while patients receive very real benefits in the interim?

Of course, neither “cure” nor permanency are the bar for FDA to find that a new drug is effective; simply because there is the potential for something to be a “cure” or offer permanent/near-permanent effects does not mean it should be required to be such.  At the same time, it is understandably important to be able to measure and characterize a treatment’s effect over time.  We know, for example, the use of viral vectors results in antibody production that could inhibit future administration of gene therapies using the same or similar viruses, which will inform treatment decisions by patients with their doctors.  However, with this most recent approval, FDA did not leave the question of durability as something to be answered postapproval, which signals to us that this issue looms large in FDA’s preapproval regulation of gene therapies.

We can see this evolution for requiring longer-term follow-up play out with this most recent approval of Hemgenix.  We can look to what information has been shared in the public domain as we do not yet have FDA’s summary basis of approval.  The gene therapy demonstrated that it increased Factor IX (“FIX”) plasma levels at 6 months, the original primary endpoint of the Phase 3 trial.  It seems reasonable to assume that there was some hope this would be sufficient for accelerated approval based on this surrogate endpoint.  However, while announcing in November 2020 that this endpoint was achieved, the sponsor shared that “[b]ased on interactions with the FDA and EMA,” it changed this primary endpoint for the trial to the co-primary endpoints of FIX activity and annualized bleeding rate (“ABR”), now measured at 52 weeks.  This was despite the 6-month data also demonstrating clinical benefits in reduction of bleeding events and declines in the use of FIX replacement therapy.

In June 2021, uniQure provided another regulatory update, announcing that FDA was requiring a demonstration of “durability of effect” based on data from patients followed for at least 52 weeks after FIX levels achieve steady state, which occurs by 26 weeks.  Essentially, FDA was now asking for 18 months of data following treatment.  Thus, the primary endpoint changed again to a sole primary endpoint of ABR at 78 weeks.  Again, this was despite the fact that subjects in the trial had demonstrated durable FIX activity to this point, with a mean of 41.5% of normal as compared to 39% at 6 months, that was coupled with substantial reductions in ABR and usage of FIX replacement therapy apparent as early as at 6 months that had now demonstrated persistence at 1 year.

That brings us to the product’s approval.  The FDA-approved Package Insert describes the “main efficacy outcome” as being “a non-inferiority test of annualized bleeding rate (ABR) during Months 7 to 18” as compared to the ABR during the lead-in period.  Therefore, the durability of response in FIX activity appeared to support the FDA-requested evidence of durability.

This approval suggests that FDA may have shown a reticence to allow such FIX levels, even with corresponding clinical benefit at earlier timepoints, to serve as a surrogate endpoint to support approval, including as a surrogate that is reasonably likely to predict clinical benefit for an accelerated approval.  Instead, FDA is requiring follow-up using this surrogate biomarker coupled with durable clinical benefits and may be delaying approvals due to the assumption that the biomarker predicts durability of effectiveness.

The fact that FDA seemingly required Hemgenix to demonstrate a degree of durability out to 18 months before determining it met the approval standard begs the question: would FDA have approved this one-time gene therapy treatment if the effects were not permanent (i.e., they waned over time), even though patients would have benefited over a shorter duration?  The regulatory bar has never required curative level or permanency of effect, so what magnitude of effect and for how long is enough for the Agency? And on what basis is FDA making such judgments?  We’ll have to stay tuned for the summary basis of approval to be posted here in the coming weeks to see what FDA has to say.