Who Invited The CAPAs?! Bracing For CGT Manufacturing "Party Fouls"

By Anna Rose Welch, Editorial & Community Director, Advancing RNA

When I was in college, the “Kappa’s” were known as the “fun” sorority. This also explains why they were ultimately banned from campus for doing things the school thought were a bit too fun. (And before you ask, no, I have no first-hand knowledge on how or why that happened; I was a hop-skip-and-three-grand-canyons away from the sorority realm.)

In the world of manufacturing in general — and the cell and gene therapy (CGT) space in particular — corrective and preventative actions (CAPAs) can in no way be likened to rambunctious sorority sisters. In fact, they’re the exact opposite. They’re how you figure out where the most “fun” was had; why that “fun” was so appealing to pursue in the first place; and how you can make sure no one tries it again. In other words, they’re the person at the party that will try to “logic” you away from setting off fireworks in the bathroom. (“No, the colors and loud noises won’t fill the void that has been growing in you since the tender age of 11 when you lost your favorite pet rabbit.”)

I know — beer tells us we should all hate that guy. But as we approach the end of the year in the CGT space, we’ve been reminded that not everything goes as well as we want it to, especially when it comes to CGT CMC. Frankly, if you’ve been reading the news, it’s been a doozy of a few weeks. For starters: 

• We’ve heard that one of Pfizer’s U.S. gene therapy trials remains hung-up because it had to go back to its potency assay matrix to appease the FDA (despite the fact quite a few other regulators signed off on it as-is).  
• Janssen and Legend, whose BCMA-directed CAR-T was due for an FDA decision at the end of this month, have found their PDUFA date extended to late February. The FDA needs more time to review new information submitted about an “updated analytical method following an FDA information request.” Frustration, thy name is method validation.  
• Gamida Cell, which is on the cusp of a BLA for its off-the-shelf HSC therapy (Omidubicel), must now demonstrate comparability between its Phase 3 clinical manufacturing site and its commercial manufacturing facility. (What I found particularly interesting was that the company specified the FDA did not ask for additional clinical data; analytical comparability will be just fine — and really, thank heavens for that.)

But if we step beyond the tricky analytical and comparability realm, we also had this pause-worthy manufacturing news out of GenSight.

Earlier this year, the eye disease gene therapy company caught my eye — pun absolutely intended — when it’s Phase 3 trial failed because it improved vision in the treated eye… and in the eye that received the placebo. (I actually laughed when I read this news.) Well, now the company has announced that its manufacturing is in the spotlight. Though the company produced three “consistent” batches of the therapy to submit its marketing authorization application in the EU, a downstream “hiccup” will require the company to redo its validation batches.

I was drawn to this information in the company's press release:  Root cause investigations have assessed the technical issue as related to operational conditions and not at all to the intrinsic design of the manufacturing process.

In particular, this called to mind a snippet of conversation during a gene therapy webinar which identified automation and the fragility of single-use technologies as two common reasons CGT companies find themselves up a creek without a paddle holding a leaking bag.

Project Farma EVP Tony Khoury (formerly of AveXis) put it best: When we were getting the first facility up and running, we had a lot more nonconformances and CAPAs than we initially planned, especially due to single-use technology [SUT] breaking and the prominence of manual operations. The sheer number of these incidents was a major curveball for us. We basically had a whole team handling nonconformances and CAPAs.

Though this example wasn’t presented as advice, Khory’s (and GenSight’s) experience lays out an important best practice for companies. The root cause of an issue doesn’t have to be paradigm-changing to be a royal pain and a time-suck. CGT companies shouldn’t underestimate the amount of operational issues that can and will arise from what remains a highly manual, SUT-laden process. A team dedicated to addressing these manufacturing “party fouls” might not be the coolest "clique" to invite to the party (no offense, guys...); but it’s the one that will keep you alive to party another day.