To take a look at current trends, developments, and modalities we’re seeing in CGT today, DHC COO Katy Spink sat down for a discussion with DHC’s Heads of Cell and Gene CMC (Sara Mills and Scott Cross, respectively) and Nate Manley, Head of Nonclinical, with a guest appearance by Senior Consultant Christina Fuentes.
KATY: Let’s jump right in with something that came up back at our JPM event in January: a surge of interest in in vivo CAR T. Is this something we’ve seen reflected in recent client asks? And if so, what kinds of problems are we helping clients with for these approaches?
SCOTT: Yes, I was surprised to see an increase in asks for in vivo lenti in 2023. Back in the early 2000s a few companies were looking into an in vivo approach using lenti, but it didn’t gain much traction at the time and the area had been quiet until the last two years. We have continued to see more companies pursuing this in 2024. The asks have been varied, mostly around CMC considerations, but we’ve also had discussions involving safety testing and regulatory considerations.
SARA: Cell Tx CMC has supported more in vivo CAR recently as well, particularly in LNP-based delivery. A challenge we’re seeing reflected here is that the potency test methods are a bit different because the activity is different, which has made for a creative process with some of those teams. We’ve crossed over with our nonclinical team as well to work through potency preclinical assays.
NATE: From a nonclinical perspective, characterization of product safety can be a challenge for in vivo gene therapy approaches. For example, safety is impacted by a given product’s cell/tissue tropism. Whether we’re talking in vivo lenti or in vivo LNP, generally we’ve found that model systems for cell/tissue tropism aren’t great, especially in small animal model systems. In other words, small animal testing isn’t as predictive as we’d need it to be. This is particularly relevant for LVVs or LNPs engineered to have novel targeting capabilities. A second, related safety consideration is the potential for genetic modification in non-target cells or tissues, with the potential for germ-line modification viewed as one of the greatest theoretical safety concerns. Use of a non-integrating transgene approach (e.g., LNP delivery of mRNA) avoids concerns of genotoxicity, but instead brings its own challenges of therapeutic durability and the potential need for repeat dosing. Design of informative dose finding studies is another challenge worth noting here. Species-specific differences in cell targeting or efficiency of transgene uptake/expression can make it difficult to extrapolate from animal models to a human dose that is expected to be safe and potentially effective.
SCOTT: Right, and LNPs target the liver quickly, so identifying a dose with maximum overlap of safety and efficacy is challenging.
KATY: From a product development standpoint, what does this mean for our clients? Is that a slower start in the clinic with more monitoring and staggering, for example?
NATE: Yes, although the lack of predictive small animal models can be partially addressed through heavy reliance on in vitro models, inclusion of primate studies, or a combination of the two. However, in many cases, limited availability of relevant nonclinical models may warrant a more slowly structured clinical protocol with lower doses and more staggering between patients.
KATY: Makes sense. That’s certainly consistent with what we saw at Geron in the early days – since it was difficult to know how predictive xenograft models would be of safety issues for our human embryonic stem cell-derived products, FDA required a slower safety progression, beginning with lower doses with long staggers between patients and frequent monitoring for potential safety concerns. It follows that a similar approach might be required by regulators in this case.
SARA: Exactly…think about how far we’ve come in the sixteen years since that first IND clearance for pluripotent stem cells. We’ve now reached the point where well over a decade of clinical data from multiple products have proven to regulatory bodies that if we have strong CMC packages and analytical methods, we may not need long-term spiking studies to address tumorigenicity. Of course, it’s a different use case: for the early pluripotent stem cell programs the indications (for example, spinal cord injury and diabetes) were typically very serious but not immediately life threatening. However, with CAR Ts these therapies are designed to treat quick-acting life-threatening illnesses. That alters the risk/benefit profile from the perspectives of the drug developers and Health Authorities; sometimes these clinical trials may be the best or last hope for a patient and their family.
SCOTT: The pluripotent stem cell historical example does provide a view into the future on the reasonable expectation that in vivo will eventually gather a large enough body of data to allow for a scaling back of some safety concerns. The question becomes how to get from here to there.
SARA: Yes, that’s a common challenge in cell and gene therapy. For a new product modality, there’s often no ‘yellow brick road’ in existence that you can just follow. Each product has its unique challenges that need to be overcome. And that becomes a creative process, between all of our cross-functional SMEs and our clients. For example, we tap into Nate’s group to work through the strengths and limitations of different available in vitro and in vivo model systems to demonstrate efficacy and de-risk potential safety concerns. Then my group ensures that the manufacturing process is robust and that product Critical Quality Attributes (CQAs) are well defined to further lend assurance of product potency and safety. And we can call on our Regulatory group of ex-FDA reviewers to get their perspectives on whether the combined approach is likely to be sufficient for the regulators. That’s an example of why working with DHC is such a completely different experience than working with anyone else in the space. Our hands-on “lived experience” between our prior roles in industry and as regulators, and the breadth and depth of what we see in working with about 200 clients a year is incomparable.
SCOTT: And “lived experience” also encompasses mistakes, right? I think each of us can point to examples just off the tops of our heads when we’ve been able to say to clients, ‘hey, I’ve made that mistake before when I was in your shoes, so I can actively help you avoid it now.’
NATE: Absolutely. Knowing what didn’t work previously can be one of the most valuable tools available when we’re helping clients build that yellow brick road: the safest, wisest, most efficacious path. Areas lacking an established yellow brick road are where DHC really excels. Anyone can follow a formula for success, but when a formula doesn’t yet exist, we know how to break ground…how to review the interplay of elements under consideration…and we get the specifics behind possible adjacent paths.
SCOTT: I agree with Nate. DHC is particularly strong when a path isn’t clearly defined, but also, so much of our day-to-day work is in the weeds solving common everyday issues that arise during the lifespan of a product. For instance, viral vectors are manufactured in the same basic way now as they were 20 years ago. Our staff has years of experience in these areas and know what works best in different scenarios. We are also very good at describing the risks associated with different paths, whether those be speed to clinic, scale up, tech transfer, etc. Every organization has a different tolerance for risk, and we work closely with a company to find the best path forward that meets their needs and their risk tolerances.
KATY: Very true Scott, so, to summarize…we excel at building a yellow brick road when one doesn’t exist, but also at selecting the right yellow brick road for clients from the existing available options. Changing gear to consider other trends in the space, I would say that another area generating a lot of recent interest is nonviral delivery, whether through methods like LNPs and exosomes in vivo, or through electroporation and related methods ex vivo. Can each of you speak about some of the challenges and opportunities you are seeing in this space?
SARA: My group has worked with several clients developing EVs and exosomes. Some advantages of EVs are that they can target specific tissues and delivery sites more readily than the challenges we have with LNPs currently. And another advantage to EVs is that they have little to no immunogenicity challenges of the sort that LNPs are still grappling with. But, of course, a disadvantage is that their production throughput is incredibly low compared to what you can get out of an LNP system. So, there's pros and cons to both, right? LNPs, of course, also have a relatively huge track record for consistent large-scale manufacturing and clinical safety due to the Covid vaccines whereas EVs have limited clinical data and are still challenging to control from the CMC perspective.
The use of non-viral delivery techniques is certainly increasing across the space, and we’ve recently seen LNPs gain commercial approval to deliver a gene therapy. Developing and manufacturing of these products certainly has its own set of challenges (e.g., specialized equipment, formulation challenges, and scalability), but DHC accurately anticipated the growth in this area and has an experienced set of SMEs that we deploy specifically for these projects.
KATY: And of course, another aspect of nonviral delivery of in vivo gene therapy is that the nucleic acid being delivered (whether as a plasmid, guide RNA, or mRNA) may be drug substance, rather than a starting material or critical raw material of viral or gene modified cell manufacturing.
SCOTT: True. Of course, our team already does a lot of work on plasmids for the purpose of viral vector production, so in working with non-viral gene therapy approaches we often leverage that great depth of experience with plasmids.
But regarding DNA as a drug substance, it is typically manufactured in the same way you’d manufacture plasmid…just to a higher level than you would otherwise for a starting material plasmid. We started with a spec of 95% supercoiled, which is incredibly difficult to manufacture. The reason for maintaining a spec greater than 90% supercoiling is that it helps to protect the DNA once it enters the body, so it doesn’t get degraded right away.
Part of that challenge, then, is identifying who can manufacture to those specs, but even more than that, the challenge is the assay to validate it to that degree. Manufacturing it is less difficult than finding an assay capable of detecting that level.
CHRISTINA: The transient nature of mRNA is particularly advantageous for gene editing applications where endonucleases delivered as mRNA can be transiently expressed to edit the genome and effect a lasting change. In other words, the gene editing component is expressed long enough to make a permanent change to the genome, without having the editing tool/delivery mechanism stick around forever. Long term and persistent expression of genome editing tools provides more opportunities for safety risks associated with off-target edits or chromosomal rearrangements.
KATY: Thanks, Christina. Are there any other particular trends the rest of you are seeing that we haven’t yet addressed?
SARA: From a cell therapy CMC perspective, I’ve been interested in the number of clients who used to (or had at least intended to) manufacture internally shifting over to outsourcing that manufacturing need. So many companies a few years ago shared the intent to build their own manufacturing space with a goal to manufacture their own drug product for clinical trials all the way up through and including eventual commercialization. Recently, though, more clients are looking to outsource their manufacturing. We have supported quite a few clients in finding the right manufacturing partner and in leading the tech transfer of their process to that partner. We have really refined the budgetary analysis portion of this process over time, finding more accurate ways to evaluate capabilities and costs across organizations for our clients, especially in a time where every penny counts.
KATY: That dovetails with the broader financial picture in the space, doesn’t it? I suspect that a reason for this trend is because money is tighter now than it was a few years ago; the equation on building and validating one’s own facility can switch dramatically when funding is tight. Are we seeing similar trends with our gene therapy clients as well, Scott?
SCOTT: Definitely. We are seeing fewer and fewer companies deciding to build or buy, due in large part to sticker shock when actually mapping out how much it costs to build or acquire. In one case, after helping a client assemble a cost and timeline proposal to build their own facility, the board was hit with the reality that the true cost was going to be four to five times their initial estimate.
SARA: And there’s room available at the CDMOs that are already built and validated, so tapping into that space can really help accelerate your timelines versus a facility build. The leverage exists on the sponsor side in the current climate. In the recent past those who were outsourcing manufacturing may have had to deal with long waitlists and lots of non-negotiables from CDMOs. But right now, there’s a lot more leeway in negotiation between the developers and the manufacturing facilities. CDMO selection, tech transfer, and relationship management are key areas where DHC excels. In fact, they fit that “lived experience” point you made earlier, Scott; our tech transfer packages and CDMO project management support are so valuable to clients because our PMs know all the things that can go wrong and how to save money by proactively planning to prevent those issues from arising.
SCOTT: Another way in which we are seeing resource constraints impact how clients engage us is in the increased demand for our interim technical leadership support. Whether the path was via layoffs or simply restricted hiring…it’s relatively common for clients to engage us to fill a role on a part time or interim basis rather than hiring a full-time employee to carry out that function. In addition to getting immediate leadership support from someone with years of experience in the field, many clients take advantage of DHC’s ability to build a team of SMEs to cover multiple vacancies or simply provide support across a range of areas like CMC, regulatory, quality, and analytical.
SARA: And I would point out that in this field there’s a ton of product diversity and things change, fast. Some things that used to be risky are no longer considered to be, whereas new risks in different areas have been uncovered and need to be mitigated. DHC has not only the strength of understanding and thought-leadership, but recent and plentiful hands-on experience for nearly anything a CGT client could encounter. This is particularly true when it comes to interacting with regulators. There was a funny internal story going around recently about a consultant who was asked what their most recent hands-on experience was in consulting with FDA on a particular topic. They checked their calendar so as to be precise and, well, their most recent interaction on that topic was only a few weeks prior. That constant finger on the pulse is invaluable in a field that’s changing and growing at the rate we are in CGT.
KATY: Before we run out of time, are there other trends any of you would like to touch on?
SCOTT: I’d like to make mention two of our new offerings, which are next-generation sequence (NGS) analysis and off-target analysis. There’s an important distinction to draw here. We’re not in the business of performing the actual sequencing; rather, we support clients in interpretation of the data. We support a lot of clients in the AAV space, and the need for NGS analysis was becoming a more and more frequent ask, so we decided to offer it as a formal service.
NATE: How to design a robust off-target analysis plan and how to de-risk findings coming out of that are two increasingly common asks at DHC. Because we routinely sit in on clients’ regulatory engagements, we maintain an up-to-date read on regulatory expectations regarding off-target analysis. This is definitely an area of increased focus for the FDA and other health authorities. Some aspects of off-target analysis are becoming more standardized, but at the same time, emergence of newer analytical tools, alternative gene editing modalities, and more clinical experience keeps us on our toes.
SARA: There’s actually so much else I’d love to have this group discuss—raw materials assessment being only one of a handful of hot topics in the space we haven’t yet touched on yet, especially in light of the recent FDA draft guidance.
KATY: Let’s consider this discussion to be Part I of a two-part series. I’d love to reconvene in a few months to take a look at the next batch of topics that come to mind. See you all again here soon.
