Nine years after wowing the audience at the American Society of Clinical Oncology annual meeting with a CAR-T candidate that would become Carvykti—now the world’s most successful cell therapy—Legend Biotech’s scientific founder, Frank Fan, M.D., Ph.D., is returning to the spotlight with an entirely new playbook.
This time, Fan isn’t showcasing an autologous product engineered with each individual patient’s cells. Instead, with his new venture, Wondercel Therapeutics, Fan hopes an off-the-shelf universal CAR-T platform can tackle two bottlenecks of the cell therapy industry: massive production scalability and the pitfalls associated with gene editing.
“If this approach proves successful, the critical thing is that we can achieve linear scalability in CAR-T production capacity that can match traditional biologics,” Fan said in an interview with Fierce.
Producing off-the-shelf CAR-T at scale
Limited manufacturing capacity blunted the early launch trajectory of Carvykti in multiple myeloma. After years of capacity expansions, Legend finally claimed that it and its partner Johnson & Johnson built out enough infrastructure to meet ongoing global demand at the end of 2025. Even today, the two companies are scaling their capacity to cope with future demand.
Allogenic CAR-T represents an improvement in that regard since it doesn’t require tailored manipulation of each patient’s cells. Existing allogeneic CAR-Ts may be able to produce enough doses for hundreds of patients—depending on the dosage levels—in one batch, Fan noted.
With its platform, called Revo-U, Fan claims Wondecel can yield up to 3,000 patient doses from a single healthy donor batch. That significantly improved productivity can directly translate into lower costs in the future, he said.
Additionally, the development of allogeneic CAR-T therapies has so far been reliant on gene editing. Because the therapeutic cells do not come from the patients themselves, to prevent potentially dangerous Graft-versus-Host disease (GvHD) or graft rejection, developers typically use tools like CRISPR or TALEN to knock out several genes on donor T cells. This creates safety concerns, such as translocations where severed chromosomes attach incorrectly, potentially leading to uncontrolled cell proliferation and hence tumor formation.
The other problem with gene editing is that it requires electroporation, a method to introduce genetic material into T cells using short electrical pulses. The throughput of electroporation is limited as it may kill a lot of cells with poor batch-to-batch consistency and transfection efficiency. The entire gene-editing process is complex and needs a lot of costly GMP-level materials. All of these factors could constrain product yield, Fan said.
Wondercel’s Revo-U approach bypasses the genome-editing path through a proprietary, protein-directed degradation technology to eliminate the TCR-CD3 complex. It utilizes antibody epitopes for targeted recognition and trafficking to protein-degradation organelles for clearance, Fan said.
“It avoids the cellular stress of electroporation, meaning the cells remain healthy, highly functional, and biologically more natural,” he said.
Compared with other non-gene-editing approaches, Revo-U can achieve more gene-editing-like outcomes while circumventing issues related to residual proteins, which may affect the fitness of the therapeutic cells, he explained.
In vivo CAR-Ts represent another potential major class of competitors to Wondercel. But Fan argued that as an emerging technology, in vivo CAR-T also has a long way to go to prove itself, especially in terms of safety. Besides, in vivo CAR-T requires more pricey viral vectors than Revo-U.
In early clinical data reported from AstraZeneca’s BCMA-directed in vivo candidate acquired from EsoBiotec, three of five multiple myeloma patients experienced grade 3 cytokine release syndrome (CRS), raising questions of insufficient viral vector purification. One patient died directly from extramedullary lesion-related spinal cord compression, but the timing raises the possibility of an indirect contribution from CAR-T.
Baby steps
While the potential is substantial, Revo-U faces a steep uphill climb as a new platform that clearly still needs refinement. Unlike Carvykti’s impressive debut at ASCO 2017, Revo-U’s initial data thus far are very immature.
In Wondercel’s ASCO 2026 data reveal, four patients in China with previously treated diffuse large B-cell lymphoma received one of three doses of the company’s CD19-directed CAR-T, Revo-UWD19, as part of an investigator-initiated trial.
The first patient received standard lymphodepletion and 100 million CAR-T cells. The investigators reported that the patient’s disease was under control 13 months post-infusion, with no adverse events recorded other than those related to lymphodepletion.
Patient 2 received 200 million CAR-T cells and achieved a complete response at their 1-month assessment—but the disease progressed at month 3. The patient had grade 1 cytokine release syndrome (CRS) with fever lasting five days.
The third patient, a 73-year-old male, received a high dose of lymphodepletion and 200 million CAR-T cells, leading to a major expansion of CAR-T, reaching over 19 million copies per milliliter of blood on day 15. The patient developed grade 3 CRS and hemophagocytic lymphohistiocytosis, a hyperinflammatory toxicity related to CAR-T treatment.
Despite aggressive management, the patient died of sepsis on day 52 after battling a serious bacterial infection and prolonged neutropenia.
The case was tragic, but to Fan, it also shows the potential that Revo-U can solve regarding a limitation of allogeneic therapies—premature clearance by the host immune system—while achieving CAR-T expansion at levels similar to that of autologous CAR-T.
After the fatal event, investigators drastically lowered the CAR-T dose to 50 million cells for the fourth patient, who had not yet shown an objective response.
In the future, Wondercel will need to refine its dosing, lymphodepletion and immune suppression management strategies to find the right balance between efficacy and safety, Fan acknowledged.
After abruptly leaving Legend in 2022, Fan formed Wondercel and got nearly 100 million Chinese yuan in angel funding from Yuanbio Venture Capital in 2023. Soon following his departure, Fan was dragged into a trade secret dispute with Legend’s then-parent GenScript. China’s high court sided with Fan in November.
While Fan is now trying to avoid crossing paths with Legend again, the Carvykti inventor hopes to replicate the commercial partnership model that Legend did with J&J.
“We also hope to partner with large companies with strong commercial capabilities,” Fan said. “This would allow us, as a biotech, to focus on developing foundational innovative technologies, which is our core strength.”
