Liu Xiao: Hello, Dr. Zhang. We are all aware of your deep research experience in the laboratory of Academician Carl June, the "Father of CAR-T." A fundamental question: Does the emergence of In Vivo CAR-T signify the end of Ex Vivo CAR-T?

Zhang Changfeng: This is not an end but more like a "reunion of old friends." The earliest concept of cell therapy was, in fact, In Vivo. Back in the 1990s, scientists attempted to directly inject plasmids carrying CAR genes into patients' bodies, using electroporation to facilitate cellular uptake. However, limited by technology, the effects were minimal and the process was painful. This led the industry to shift toward the Ex Vivo approach—modifying cells outside the body and then reinfusing them.

Over the past two decades, the field of CAR-T has evolved, and gene delivery systems (such as LNPs and lentiviruses) have also advanced rapidly. Their convergence today is a reunion after years of independent progress. If some view In Vivo CAR-T as an "unwelcome intruder," I would say, from a historical perspective, Ex Vivo CAR-T was actually the latecomer.

Liu Xiao: In June of this year, the CDE released the "Points to Consider for Clinical Trial Applications of Cell and Gene Therapy Products," which categorizes CAR-T as a cell therapy. However, from a technical standpoint, In Vivo CAR-T seems more akin to gene therapy. What is your perspective on this classification ambiguity?

Zhang Changfeng: From a strict technical definition, In Vivo CAR-T is a standard gene therapy product. It uses vectors (viruses or LNPs) to genetically modify cells within the body. However, in practical regulatory terms, particularly in China, classifications often rely on "established conventions" and "regulatory inertia." Due to its "familial relationship" with CAR-T, it is currently managed under the CAR-T umbrella, thus falling into the category of cell therapy. This is more for the sake of regulatory continuity and the convenience of transferring existing knowledge frameworks, rather than a purely scientific definition.

Liu Xiao: What unique challenges does this "hybrid" identity in definition pose for regulation?

Zhang Changfeng: When regulators face new technologies, the first question they ask is whether it belongs to a "new category." When Ex Vivo CAR-T first emerged, it represented an entirely new drug category, posing significant challenges. The "fortune" of In Vivo CAR-T lies in the fact that Ex Vivo CAR-T has paved the way for it. It is viewed as a new variant within the broader CAR-T category, rather than an entirely new class, which significantly lowers the cognitive barrier for regulators.

Specifically, the core regulatory concerns always revolve around safety, focusing on three key points:

1. Non-target cell infection: Does the vector infect only the intended cells (e.g., T cells)?

2. Genotoxicity and tumorigenicity: Could genetic insertion potentially lead to tumor formation?

3. Risk of residual impurities: Are impurities introduced during the manufacturing process controllable?

These concerns are not unique to In Vivo CAR-T but are common requirements for gene therapies and cell therapies. In Vivo CAR-T can be understood as a "prodrug"—the entity itself (e.g., lentivirus) has no therapeutic effect, but after "metabolism" within the body (transduction of T cells), it becomes active CAR-T cells. Therefore, regulatory assessment is divided into two parts: the safety of the "prodrug" (the vector, including immunogenicity and insertion risks) and the safety of the "active drug" (the CAR-T cells generated in vivo, such as CRS, neurotoxicity, and off-target effects).

Liu Xiao: In terms of quality standards, which specific indicators are regulators particularly concerned about? How do these differ from those for Ex Vivo CAR-T?

Zhang Changfeng: Quality standards can be divided into two main categories:

Category 1: Standards for the vector itself as a drug. Regardless of what is packaged inside, if you are producing lentivirus or LNPs, you must meet the general standards applicable to such drugs. For example:

• Lentivirus: Focus on empty capsid ratio, aggregation level, various process residuals (e.g., host cell proteins, DNA), sterility, endotoxins, etc.

• LNPs: Focus on encapsulation efficiency, particle size distribution, related impurities, etc.

This category represents mature, non-negotiable baseline requirements.

Category 2: Standards related to the unique mechanism of action of In Vivo CAR-T. This is where the core regulatory differences lie, primarily revolving around the three key safety concerns mentioned earlier. Specifically, this includes:

1. Targeting specificity validation: Must demonstrate through in vitro experiments that your vector efficiently and specifically transduces the target cells (e.g., CD8+ T cells) and not other non-target cells.

2. Functional efficacy validation: Must demonstrate that the CAR-T cells generated in vivo through transduction possess the intended functions, such as cell-killing ability and cytokine release.

3. Control of insertional mutagenesis risk: Requires analysis of vector integration sites and control of vector copy number (VCN) per cell within a reasonable range (e.g., 3–5 copies) to avoid risks associated with excessively high copy numbers.

4. Upgraded impurity standards: When lentivirus transitions from its role as a "raw material" in Ex Vivo processes to a direct "drug product" in In Vivo applications, the residual limits for many impurities (e.g., Bovine Serum Albumin, BSA) must be significantly tightened, potentially aligning with the stringent standards applied to vaccines or recombinant proteins (e.g., BSA residues may need to be controlled below 50 ng per dose).

Liu Xiao: Recently, an overseas In Vivo CAR-T company delayed the release of its IIT data in China, citing enrollment issues. In your opinion, what are the main challenges for overseas companies conducting IIT (Investigator-Initiated Trials) in China?

Zhang Changfeng: Based on our observations, the challenges for overseas companies conducting IITs in China rarely stem directly from product safety or patient enrollment speed. More often, the issue is a "lack of adaptation to the local environment":

1. Misalignment of perceptions and expectations: They may have heard that IITs are "fast" but do not fully understand China's complete IIT system, ethics review processes, and hospital collaboration models. This can lead to internal disagreements and concerns, slowing down decision-making and execution.

2. Challenges in operational localization: Establishing a legal entity, building a team, and communicating with domestic hospitals (which often involves very different approaches compared to the West) can present unexpected difficulties. Even a seemingly straightforward ethics approval process can become a significant hurdle for them.

Liu Xiao: Indeed, IITs represent a distinctive "fast track" within China's innovative drug R&D ecosystem, but they are by no means synonymous with "low standards or low quality." The scientific and regulatory requirements are equally rigorous. They provide a valuable opportunity for companies—both domestic and those international firms that truly understand China—to rapidly obtain proof-of-concept data in humans.

Liu Xiao: Beyond regulatory challenges, what do you think is the greatest challenge facing In Vivo CAR-T, and indeed the entire innovative drug industry?

Zhang Changfeng: I believe the most profound challenge lies in the Chinese pharmaceutical industry's still-insufficient understanding and preparation for the essence of "innovative drugs." In the past, our industry primarily focused on generics, traditional Chinese medicine injections, and similar products. The logic of generics is "how to manufacture it," while the logic of innovative drugs includes the possibility that "it may simply not work at all."

Take CAR-T as an example. Before the success of Emily Whitehead (the first leukemia girl cured by CAR-T), the University of Pennsylvania team went through 26 INDs (Investigational New Drug applications), with only one ultimately achieving commercial success. This is a process of "accumulating strength for a breakthrough," or even "brutal competition." However, many domestic companies and investors still hold expectations like "every project will succeed" or "one out of two or three will succeed," underestimating the inherent high failure rates and long-term risks of innovation. This is the greatest challenge to truly innovative technologies.

Liu Xiao: I couldn’t agree more. China’s success rate in innovative drugs was once artificially inflated. As we delve deeper into truly cutting-edge technologies like CAR-T, ADCs, and gene therapies, we are learning to accept and respect the scientific realities of drug development. IND approval is only the first step of a long journey.

Liu Xiao: In Vivo CAR-T currently has relatively few targets. Do you think future variants—like dual-targeting, armored designs, or applications to other immune cells—will emerge, similar to Ex Vivo CAR-T?

Zhang Changfeng: Almost certainly. The development path of In Vivo CAR-T is likely to mirror the exploratory journey of Ex Vivo CAR-T:

1. Vector optimization: Developing "switchable" or "universal" vectors that can target different cell surface antigens.

2. Diversification of CAR structures: From single-target to dual-target, and even designs incorporating immune-modulating factors (armoring).

3. Expansion of cell types: From CAR-T to In Vivo TCR-T, CAR-M, CAR-NK, and more.
   This is a natural extension of technological evolution and a testament to the vitality of the industry.

Liu Xiao: A critical question: Does the success of Ex Vivo CAR-T in hematological cancers mean its technological platform can be easily transferred to the In Vivo pathway and achieve similar success in hematological or even solid tumors?

Zhang Changfeng: It’s not that straightforward. Successfully generating CAR-T cells in vivo and achieving clinical success in a specific indication are two different concepts. In Vivo CAR-T has its own advantages (e.g., cost, accessibility) and limitations (e.g., transduction efficiency, targeting specificity, in vivo expansion dynamics). Just as Ex Vivo CAR-T succeeded in hematological cancers but struggled in solid tumors, In Vivo CAR-T needs to find its own "main battlefield." It may first prove its value in leukemia, lymphoma, or autoimmune diseases, establishing a foothold for the technology and the industry to "survive," before gradually tackling more challenging areas like solid tumors. Survival is the top priority for technological evolution.

Liu Xiao: Speaking of commercialization, Ex Vivo CAR-T faces challenges like high costs, complex manufacturing, and reimbursement hurdles. Do you truly believe In Vivo CAR-T has commercial value?

Zhang Changfeng: I firmly believe it has immense commercial potential for two reasons:

1. Addressing the core pain points of Ex Vivo CAR-T commercialization: When Ex Vivo CAR-T emerged, it was so "new" that it required rebuilding the entire ecosystem—production, logistics, sales, and reimbursement—while educating every stakeholder, resulting in extremely high costs. In contrast, In Vivo CAR-T resembles an "innovative core packaged in a traditional form." It encapsulates the innovative concept of CAR-T within a relatively mature "drug" format (e.g., LNP formulations or viral vectors), making it easier to integrate into existing drug R&D, manufacturing, supply chain, and reimbursement systems. Knowledge, resources, and experience can be transferred, which is a significant advantage for commercialization.

2. Potential for a cost revolution: Its production costs could be 10 to 100 times lower than those of Ex Vivo CAR-T, which is critical for patient accessibility and healthcare reimbursement.

Liu Xiao: So, is this kind of innovation more likely to come from large pharmaceutical companies or small biotechs?

Zhang Changfeng: Based on my experience (having worked in both startups and now at a large pharmaceutical company like Shanghai Pharma), disruptive innovation often originates from small companies. While large companies have resources, their long decision-making processes and high risk aversion make internal incubation of entirely new technologies challenging. They often rely on business development (BD) to acquire innovation. A unique feature of the cell therapy industry is that it hasn’t yet formed solid technological barriers or monopolies, and the landscape changes rapidly. This gives small companies the advantage of agility, rapid iteration, and seizing opportunities. Open innovation collaborations between large companies and small biotechs will be the mainstream model.

Liu Xiao: Finally, a soul-searching question: Is In Vivo CAR-T truly a competitor to Ex Vivo CAR-T?

Zhang Changfeng: My perspective is clear: They are not competitors but collaborators in a synergistic relationship. The greatest threat to the CAR-T industry is not the divergence of internal technological pathways but the stagnation of the entire field, losing indications and attention to other therapeutic modalities (such as bispecific antibodies, ADCs, radiopharmaceuticals), leading to industry contraction and a "loss of relevance."

The emergence of In Vivo CAR-T has injected new vitality, imagination, and capital attention into the entire cell therapy field. It has brought the concept of "in vivo engineered immune cell therapy" back to the center stage, attracting more talent and resources into the arena. This is akin to "keeping the value within the pot"—ultimately enriching the scale and influence of the entire "cell-based immunotherapy" domain. Together, they compete for physicians' prescribing preferences and patient benefits—against disease, not against each other.

Liu Xiao: A brilliant summary. In today's dialogue, Dr. Zhang has outlined a grand and dynamic landscape for In Vivo CAR-T from multiple perspectives: regulatory science, technological history, and industrial logic. It represents both the revival of an old technology and the challenges of a new environment. Its success depends not only on the technology itself but also on the industry's respect for the principles of innovation, its focus on patient needs, and the collaborative progress of the entire industrial chain.

Thank you, Dr. Zhang Changfeng, for your candid and profound insights, and thank you to all our colleagues for joining us. The story of China's innovative drug development is being written by scientists like Dr. Zhang and every member of the industry. Until next time.