Blog | September 21, 2021

Is This One Solution To The AAV Gene Therapy Full/Empty Capsid Dilemma?

Anna Rose Welch Headshot

By Anna Rose Welch, Director, Cell & Gene Collaborative
Subscribe to my blog ARW on CGT here!


It was a tragic headline to read: “Fourth Boy Dies In Astellas-Audentes Gene Therapy Trial, Raising Fresh Fears For The Field.”

Audentes has not had an easy go of things with its gene therapy AT132 for myotubular myopathy — first encountering hurdles with high doses and, now, facing a patient death after administering a lower dose. The cause of this most recent death still has yet to be determined. However, as you can imagine, this latest tragedy raises even more questions for the industry rather than provides a clear answer about how dose size and the overall purity of that dose may impact the overall toxicity of AAV gene therapies.

This was very timely news, given the marathon FDA meeting on AAV toxicity at the beginning of September. For those of you who watched or read anything about it after the fact, you’ll know that (unfortunately) little was decided about how to approach companies’ full/empty capsid ratios. In fact, the only thing the agency seemed certain about after this meeting was how much uncertainty remains.

As the officials explained, they are not currently in the best position to cap or provide blanket guidance as to the total vector genome dose/kg or the number/percentage of empty capsids because:

A.) there is too much unknown scientifically about the safety/overall clinical impact of empty capsids (the research is kind of all across the board — and what do we even mean by “empty” anyways?)

B.) there is too much analytical/measurement variability across AAV gene therapy company development programs; and

C.) there are too few/no reference standards for measuring the quality of AAV vectors. (FYI, this is a work in progress, especially as it relates to full/empty capsid ratios!)

While the agency and companies are sensitive about and striving to lower gene therapy dosage overall, reducing the percentage of empty capsids is an exercise rife with conflict. Not only does the industry point to reasons A, B, and C above, but the time and cost of making AAV can make regulatory pushback on full/empty capsid ratios even harder to swallow. As such, the percentage of empty capsids can vary widely across the AAV gene therapy board. 

This is why, when I read through the EndPoints News article linked above, one sentence (particularly the bolded portion) called out to me like a foghorn through the night: “Sixteen months after the first death, [Astellas/Audentes] have yet to publish a peer-reviewed paper on the study or release key detailssuch as the impurity levels in their therapyciting trade secrets.

On the one hand, reading this statement was hardly a surprise. I don’t have to tell you that manufacturing today is, or can be, a competitive differentiator, and in this business IP is paramount. Establishing analytical platforms and purifying vector today is no mean feat.

On the other hand, however, I’d also argue that something is going to have to give for the industry to attain the level of scientific clarity and regulatory guidance it craves around vector quality. Given just how big a head-scratcher full/empty capsid ratios are today for industry and regulators alike, I’d welcome (and am hoping for) a future in which companies are more willing to publicize this information.  

How this could (and maybe should) be done was a topic of conversation during this lively podcast episode featuring two brilliant C&G dynamos, UCSF’s Nicole Paulk and Dark Horse Consulting’s Anthony Davies. As they discussed, this information could be shared privately within an industry consortium between companies and perhaps the FDA (i.e., ARM, ASGCT, BIO). Or, on the more public end of the spectrum, in lieu of (theoretically) designating hard and fast caps on acceptable empty percentages, regulators could instead require companies to report the full/empty ratio on their pages.

At the end of the day, the goal isn’t to share all the nitty gritty proprietary stuff, including the genomic sequences or the capsid or promotor identity. Rather, Davies likened this exercise to providing a partially redacted COA (or, as this poet would label it, a regulatory erasure poem) that blacks out all the top-secret info aside from that ratio.

Providing greater transparency around full/empty capsid ratios is most likely not something the industry is overcome with excitement to share today. (If I listen closely, I can actually hear the hair on some lawyers’ necks stand up at just the suggestion.) But scientifically speaking, it’s a no brainer that such transparency would benefit academics, physicians, industry, and regulators at large. Only by having this data amassed in one (or even several) places will the industry finally be able to start mapping out and making connections (or uncovering a definitive lack of connection) between the presence/percentage of empty capsids and SAEs. After the past couple of weeks in the news, it’s clearly an answer we desperately want and need — but to get it, we're going to have to get more comfortable with sharing.