How Platform Technologies Are Shaping The Gene Therapy Landscape
By Tyler Menichiello, contributing editor
This is a continuation of my coverage on gene therapy discussions at the 16th annual Bioprocessing Summit. There were two panel discussions that stuck with me: The first featured CMC experts talking about the challenges commercializing gene therapies, with a particular focus on late-stage process changes. You can read about that session here.
The second panel, a fireside chat moderated by Prime Medicine’s CTO, Ann Lee, Ph.D., featured leading experts in CRISPR, prime editing, base editing, and epigenetics: E. Morrey Atkinson, Ph.D., EVP and CTO at Vertex Pharmaceuticals, Manmohan Singh, Ph.D., CTO at Beam Therapeutics, and Heidi Zhang, Ph.D., EVP and head of technical operations at Tune Therapeutics.
Throughout the hour-long discussion, the panelists shared insights and lessons learned from their experience developing and commercializing gene therapies. They covered a wide range of topics —more than I can reasonably include in this review — but to me, the most valuable parts of the conversation centered around CMC strategy for platform technologies, product lifecycle management, and the role CDMOs play in moving the space forward (and how to effectively work with them to do so).
CDMOs Playing Catch Up
One of the biggest challenges in the space right now, at least in the short term, is the CDMO landscape, according to Singh. “Many of us still leverage a broad network of CDMOs who are building capabilities, but the expectation of the quality of some of the raw materials that need to go into genetic medicines is also evolving,” he said. “The CDMOs are just catching up to those quality attributes.”
To be fair, CDMOs are still in relatively uncharted territory and doing their best to supply the burgeoning market. “For advanced therapies, the innovator companies are ahead,” Atkinson said. “You’re still teaching CDMOs. They might be trying to come up with platforms, but we’re still teaching them, and that wasn’t the case in small molecules or biologics.”
That’s not to say that CDMOs are incapable of supplying the market — they just couldn’t have predicted how the needs and requirements around gene therapy manufacturing would evolve. “They were always planning on making reagents,” Atkinson said, not GMP materials or PPQ batches, and certainly not supplying raw materials that are regulated as drugs. Ultimately, the responsibility lies with sponsor companies to clearly define expectations for their product(s).
“We literally had to sit down with the CEOs of a couple of these companies and say, I know this isn’t what you were planning, but if we’re going to supply the market, this is what we need to do,” said Atkinson. “We had to teach them how to do PPQ batches, and how to validate their methods, how to control their incoming materials,” he continued.
As with any worthy endeavor, problems are part and parcel with ATMP manufacturing, so don’t trust a CDMO who says they “never had a problem,” said Lee, because they’re not telling the truth. “Even at Genentech, where we had manufacturing plants and we were leaders in antibodies, there were always problems,” she said. “The question is, how quickly are you able to pivot? How quickly are you able to really show that kind of resiliency?” Communication — honest communication — is paramount in the sponsor-CDMO partnership, she said.
Zhang acknowledges the challenges CDMOs face as well, saying, “They easily have the hardest jobs dealing with so many different sponsors with different interests.” The way CDMOs in the space have navigated the inherent ambiguity with relatively limited time and resources has been impressive, she said.
Leveraging Platform Technologies
At one point, the conversation turned to platform technologies and their potential to shape regulatory guidance around gene therapies. The hope is that, as more novel therapies are approved, their respective technologies (e.g., CRISPR, prime editing, and base editing) can be applied at a platform level, Singh said.
“If you’re going after a certain sequence or certain base in the DNA genome, one could argue it’s the same nuclease enzyme, all you’re changing is the guide RNA for specificity and editing,” said Singh. “The assays to really characterize and manufacture them are going to be the same.”
The panelists agreed that the bulk of CMC data generated in initial development for a platform should be applicable to subsequent products developed with the same platform technology, especially if all that’s changing is the guide RNA. There’s no sense in generating redundant data, said Lee.
Singh said he believes the onus is on sponsor companies to convince regulators that these processes do work at a platform level. This means having a solid control strategy for processes and generating enough data in the next several years (including with commercial products) to convince regulators that something like changing an RNA guide, for example, doesn’t impact product quality or safety.
More broadly, the concept of platforms has shifted from being something within a single company to more of a manufacturing process that’s established at CDMOs who serve multiple sponsors, Zhang said. Variations to platform processes are introduced with each sponsors’ product.
“Multiple sponsors could potentially come up with a more refined platform process at a given CDMO, and it just opens up partnership and collaboration to a whole new level, which I think has been very stimulating and also positive for the industry as a whole,” she said.
This adds texture to the panelists’ earlier point about CDMOs adapting to the needs and demands of ATMP manufacturing. It will take a lot of education and work up front, but as CDMOs expand their capabilities and work with more sponsor companies, best practices will become more established. This, in turn, should streamline subsequent product development.
Get Ahead Of Your CMC Strategy
Early in the discussion, Lee asked the panelists to share some regulatory “war stories,” which led to a conversation about CMC strategy and regulatory considerations. Atkinson shared a lesson he and his team learned from commercializing Casgevy. “If you’re doing ex-vivo cell therapies, you need to have a really good particle control strategy,” he told the audience.
“I think one of the things that kind of took us a little bit by surprise was the focus on particles,” he said. There are strict guidelines on particle control and the presence of particles, particularly in the U.S. “We had to come up with a very comprehensive particle control strategy at a very, very late date to make sure that we met the regulatory expectations,” he said. “It was tough, and they [FDA] were very serious.”
When it comes to process development, the panelists agree that ironing out wrinkles and having a solid process-control strategy in place early on is advantageous.
“I think the conventional wisdom is that you just want to get your product into the clinic and then you will figure out something else later, but there’s really no time later,” Zhang said. This is especially true for advanced therapies, which are often granted rapid regulatory pathways to BLA submission. “You don’t have the five, six, seven years that you can take to optimize your manufacturing platform process,” she said. Her advice is to solve problems now — don’t wait until tomorrow, because tomorrow, there will be other problems you need to solve.
Lifecycle Management Doesn’t End With Approval
Towards the end of the hour, as the panelists shifted their conversation to lifecycle management, all agreed: A product’s lifecycle doesn’t end with approval. Approval is great, but it’s just the beginning of a very long journey, Atkinson said. As soon as the Casgevy BLA was approved, Vertex had a whole list of prior approval supplements and other variations around the world “ready to go.”
“We knew that what we went to the market with was just 1.0, and we knew we had to get to 2.0, 3.0, et cetera, to really serve patients, to make it a more effective medicine for us to do globally, make it more cost-effective, more reproducible,” he said. After the approval, the company began filing plans for process changes, method changes, and sites to be added.
“The lifecycle aspect of the process is also dependent on what device or instrumentation, or automation is available when you build that process,” Singh said. Technology is evolving at break-neck speed, and by the time a product is approved, companies need to already be thinking about how to incorporate new technologies to improve process efficiency.
“Thinking about, when do you plug in that technology, how do you bridge your assays and your data to support a commercialization application later on, it’s super important,” Zhang said.
Overall, I think the panelists would agree that the industry is heading in the right direction. Problems and challenges are inherent in science and drug development, but thankfully, great scientists around the world are united in a shared goal: to overcome these challenges and bring life-changing therapies to patients. As companies continue forging ahead to bring advanced therapies to market, mistakes will be made, lessons will be learned, and patients around the world will benefit.