Cell and Gene Therapy Modalities: Business and Manufacturing Strategies Influencing the Decision to Develop One Therapy Type Instead of Another

Moderator Patricia Seymour (BPTG, BDO USA) with Justin Skoble (Caribou Biosciences, Inc.), Adam Fowler (Locanabio), and Eytan Abrahan (National Resilience)

Moderator Patricia Seymour opened the panel by asking each presenter to introduce himself and his company’s technology.

Caribou Biosciences: Cell Therapies
Panelist Justin Skoble (vice president of technical operations with Caribou Biosciences) is responsible for developing the company’s next-generation CRISPR genome-edited cell therapies. Before joining Caribou in 2019, he was chief development officer at Acton Therapeutics and served in positions of increasing responsibility in immuno-oncology and vaccine development at Aduro Biotech, Anza Therapeutics, and Syros. His experience encompasses development of small molecules, antibodies, microbial platforms, and cell therapies.

Skoble: Caribou Biosciences is a small start-up biotechnology company based in Berkeley, CA, and spun out of Jennifer Doudna’s laboratory. We focus on precision genome engineering with a second-generation CRISPR technology, chRDNA (prounced “chardonnay”) technology, for CRISPR hybrid RNA/DNA. It enables engineering of mammalian cells with a high degree of specificity to conduct multiplex genome editing at high rates with low off-target consequences. The technology is broadly applicable to a number of cell types, with allogeneic CAR T-cell therapies in the initial pipeline. Caribou’s next generation of cell therapies will include induced pluripotent stem cells.

Our main goal is to improve patient accessibility to such therapies. Issues involve timely manufacturing of autologous cell therapies so that those cells can be brought back to patients without the need for bridging therapies. Patients need to have healthy enough immune systems to provide their own apheresis material. But with an allogeneic cell therapy, healthy donor T cells are good starting material for an off-the-shelf product that can be made available immediately. Using healthy donors allows for a more robust memory phenotype: The cells are less likely to be exhausted by all the previous cancer chemotherapy regimens that patients have gone through.

Manufacturing an off-the-shelf product and reducing the cost of goods for patients will broaden access to such potent CAR T-cell therapies. However, chief among the drawbacks to using allogeneic cells is that they are not recognized as self. Caribou’s approach is to improve the potency, durability, and persistence of those cells once delived to patients.

One strategy incorporated into the CB-010 program is to knock out the PDCD1 gene to abrogate PD-1 signaling access in the CAR T cells themselves. That results in a robust, immediate antitumor response. A second strategy to improve persistence is incorporated into the CB11 program (currently preclinical) and referred to immune cloaking. It simultaneously eliminates endogenous HLA class 1 presentation on the surface of the CAR T cells and expresses HLA-E, a minor HLA class 1 antigen, to blunt both T- and NK-mediated rejection of a CAR T-cell therapy.

Locanabio: RNA Modifications
Adam Fowler is the senior vice president of technical operations at Locanabio, a San Diego–based private biotechnology company that he joined in 2020. He has led a number of early to late-stage CMC drug development activities and has participated in more than 20 INDs and two licensed biologic projects. His experience spans a broad range of biologic process development, manufacturing, and supply-chain management, including nine years in gene therapy. He has driven development and transfer of scalable processes using AAV, lentivirus, and adenovirus and has led internal manufacturing operations.

Fowler: Locanabio is pioneering a class of targeted therapies that correct or modify dysfunctional RNA, the root cause of many genetic diseases. The CORRECTX platform is a modular system that can modify RNA by deleting or degrading a defective or mutant sequence; by partnering the binding system with a nuclease; or depending on the mode of action, sometimes just by blocking or splicing a sequence. We can target multiple sequences with the same single vector and have plans to offer RNA editing. We use an AAV serotype as the delivery mechanism but also are exploring other opportunities.

Locanabio is focused on two specific neuromuscular disease areas: myotonic dystrophy 1 and Duchenne’s muscular dystrophy. Myotonic dystrophy 1 is target of our lead program, for which the company hopes to file an IND in the second half of 2023. Locana also is exploring approaches to treating central nervous system (CNS) disorders such as Huntington’s disease and ALS. Following our start as an early stage research company, we now are building out development functions.

Resilience Cell Therapy: Advanced Therapies Manufacturing
Eytan Abraham heads the Resilience Cell Therapy franchise, which focuses on partnering with pharmaceutical and biotechnology clinical centers to deliver best-in-class process development and clinical and commercial biomanufacturing. Previously, he headed the personalized medicine business unit at Lonza and its cell and gene therapy business, based in the Maryland area. Abraham has over 15 years of experience in the cell and gene therapy space focusing on R&D, CMC, and commercial development.

Abraham: Resilience is a fairly new company that was formed during the pandemic to address a severe lack of capacity and of technologies to scale and manufacture advanced therapies as needed — focusing on some of the key technological gaps in those modalities. Another differentiator that we feel strongly about is forming long-term partnerships with our collaborators: We believe in partnering for the long term with therapeutic companies in a way that’s sustainable and allows them to commercialize their products. Our novel business structures ensure that we align incentives and work together to commercialize therapies.

Focusing on analytical aspects, we believe in driving a deep understanding of the mode of action and critical quality attributes of both product and process (MoAs, CQAs, and CPPs). From a digital standpoint, we strive for a manufacturing 4.0 approach using digital batch records and digital infrastructure for information coming out of a process to help us understand failures or successes and ensure data security and cybersecurity.

We are focused on five modalities: biologics, vaccines, nucleic acids, and cell and gene therapies. We have 11 sites in North America, with 1,700 employees and about one million square feet of space. Different modalities are processed at different sites, but quite a few sites are multimodality. We want to have almost everything under one roof that is needed to go from viral vector, for example, all the way to manufacturing GMP materials for early-phase clinical trials.

Manufacturing for Unique Modalities
Seymour asked the panelists to talk about the challenges that they are experiencing in development and manufacturing of their modalities.

Skoble: Our allogeneic CAR T-cell therapies require GMP production of a number of critical starting materials before we can manufacture the product itself. Logistics require a broad number of specialized contract manufacturing organizations to work in concert making the critical materials that we need. The challenge is in accommodating long lead times and coordinating work with a number of CMOs to meet our GMP manufacturing timelines. Supply-chain issues are real, but some of our materials are so specialized that we simply cannot bring them all under one roof.

Fowler: We have encountered many of the same manufacturing challenges and the need for greater understanding of the science and MoA so that we can ensure quality in our manufacturing. We’re delivering with AAVs and are still using the transient-transfection platform. So there is a lead time associated with getting plasmids before we can begin our AAV process. Unlike in the MAb/protein therapeutics world, we don’t have solid, dependable, cell lines yet for stable production. We are still working mostly with transient/transfection processes, which limit our ability to scale up. It keeps me up at night being able to keep the cost of goods reasonable to supply a potential commercial market.

Managing the supply chain for raw materials and equipment is difficult right now. We are exploring novel capsids or engineered capsids. Will they work in a production platform developed for an AAV? The chemistry of some of those capsids may not align with some of the commercially available resins. There’s not a lot of clear regulatory guidance about how we enrich for full capsids.

Abraham: The need for vectors is huge and drives costs up. How do we address that from a cell-line perspective? How do we increase titers, reduce downstream losses, and get more full and empty capsids? All of those elements are going to be critical to advancing the field. We are hearing about patients who are unable to receive even commercial CAR T-cell therapies because of the cost components and a lack of manufacturing capacity. Scaling these therapies is critical.

Manufacturing Innovations and Data Analytics
Seymour: Eytan, your CEO has been quoted many times saying that the unrealized power of personalized medicine is attributable to the lack of innovation in biomanufacturing. How do you introduce innovation to a manufacturing process to manage some of the regulatory risks — especially because not only are the modalities new, but introducing new manufacturing technologies can compound the risks?

Abraham: There are several ways that we’re trying to increase the availability and reduce the cost of our therapies. We are stitching together the best off-the-shelf components in a way that is most effective to enable efficiencies and close processes. We also can drop in many of the existing therapies that companies are coming up with and decrease the time it takes for process development. Another interesting element has to do with parent–child INDs: How do you keep the same process, the same reagents, but drop in a different gene of interest to accelerate interactions with regulators? And how do we further evolve fully closed, automated systems for autologous cell therapy manufacturing?

Skoble: In our first-in-human clinical study, we were thrilled at the level of activity we were seeing with the low dose (40 million CAR-positive T cells), so we are escalating to twice the dose at 80 million CAR-positive T cells. We hope to see a higher level of T-cell engagement. We don’t know yet whether it will be safe or effective, but we do think that it could provide longer durability of response basically by just eliminating the tumor from the get-go. Other strategies aim to enhance on a molecular level the persistence of each cell therapy as it is manufactured. But with just our CP-010 technology, we think that a higher dose could put us in the competitive space with autologous therapies.

Seymour: Eytan, how are you going to have the data analytics and the power of the information not just to increase understanding, but also to make decisions going forward and actually learn from that and other types of AI applications?

Abraham: First, I wouldn’t underestimate the value of electronic batch records. Anyone in QC or regulatory will tell you that doing everything on paper is cumbersome. Implementation of fully electronic batch records is advantageous. But how do you implement that across sites? How do you implement it across modalities and equipment? On a deeper level, 4.0 is not practiced a lot. We’re generating a fair amount of data from the processes that we’re running, but we’re not always collecting that information, and we’re almost never analyzing it. So how do we do that functionally in terms of collection and analysis? We need to start aggregating those data and analyzing them more robustly across companies and clinical centers. How do we collect those data, analyze them, and start to find correlations between starting materials, processes, release criteria, and clinical outcomes? It’s not simple. Once we start generating and analyzing those data, information will pop out. And we’re not doing this work in isolation, but rather, through creating collaborations with companies, partners, and clinical centers.

Seymour: What would be the attributes of a perfect CMO/CDMO?

Fowler: Beyond obvious things —capacity, technological expertise, regulatory history, price, location, and so on — I would look for a CDMO that is willing to partner with us and to invest in a relationship long term for mutual benefit despite ups and downs, delays and fall out. We need people to work with us through that ride and help us get there, but with the understanding that plans, timelines, and needs are going to change. Such commitment and flexibility are what I look for in a partner. I also would want a program or product to stay with that CMO for the full life cycle — from preclinical, first in human, and all the way through late-stage and commercial manufacturing. Moving a process from one manufacturing site to another is challenging enough, even in the world of MAbs and other modalities that we understand very well. But doing that with gene therapies is even a bigger challenge.

At Locanabio we’re taking a sort of hybrid approach. We don’t want to be wholly dependent on a CDMO to everything that we need to do, so we’ve built our own development laboratories, processes, and analytical/formulation capabilities. We want partners that are willing to work with us in that mode. Finally, it is easy to get a CDMO’s attention when you have a well-known company name, but it is not that easy when you work for a smaller start up or a private company. So I also look for CDMO partners who are willing to work with us even though we’re not one of the big companies.

Skoble: The word partnership is important. Obviously, it’s a transaction, but it works much more efficiently if everyone is aligned in understanding goals and being as flexible as possible. One way in which we’ve mitigated some difficulties is to secure a dedicated space in which we maintain control over our own schedule, even though we are using a CDMO. In that way, we’re not juggling competing priorities with other companies that have their own timelines.

We also would like to see more continuity and less turnover in this space. We should be able to incentivize people to stick with some of these programs through to product launcht.

Fowler: Having common teams that we work with across our multiple pipeline programs can be a significant benefit. We get to know each other, build relationships, and work together more efficiently — even if we can’t always keep the same team intact for years.

We are trying to build significant development capability in house, but leveraging the experience and expertise of our partner and continuing to bridge that knowledge and expertise across multiple programs.

Competition for Talent
Seymour concluded the panel with remarks about the competition for talent. Because therapeutic nnovators are hiring the same people that the CMOs need, the current struggle for all is to find qualified people. She encouraged audience members to talk with their own children and other young people about careers in the biopharmaceutical industry to help build that essential talent pool. A lively Q&A session completed the panel discussion. Questions concerned CDMO requirements for electronic batch records and how to choose whether to manufacture in-house or to outsource.

Moderator Patricia Seymour is managing director, BPTG, BDO USA. Justin Skoble is vice president of technical operations, Caribou Biosciences, Inc.; Adam Fowler is senior vice president of technical operations at Locanabio; and Eytan Abraham is vice president, head of cell therapy, National Resilience.