CAR-T cell therapy has not only revolutionized the treatment of blood cancers, but it has also shown unprecedented potential in the fight against autoimmune diseases. However, current autologous CAR-T therapies are personalized treatments, requiring cells to be collected from the patient, genetically modified and cultured in a laboratory, and then infused back into the patient. This process faces multiple challenges such as production, transportation, and high costs.
How can CAR-T therapy become easier to produce, more convenient to use, and more affordable? Companies such as Interius, Umoja, and Kelonia are exploring the concept of "in vivo CAR-T therapy." On July 9th, Interius announced that its first-in-class in vivo CAR-T candidate, INT2104, had received approval for its first Phase I clinical trial.
INT2104 uses a lentiviral vector to deliver CAR genes, enabling the generation of effector CAR-T cells and CAR-NK cells directly in the body. The generated CAR cells target CD20-positive B cells to treat B-cell malignancies. Unlike ex vivo CAR-T therapy, INT2104 is a ready-to-use, single-dose treatment that can be administered through a single intravenous infusion without the need for lymphodepletion or any special equipment or training.
"In vivo CAR-T generation is several weeks faster than the current ex vivo CAR-T production process, and the cost is expected to decrease significantly. Additionally, patients receiving ex vivo CAR-T therapy must undergo a strict chemotherapy regimen to deplete lymphocytes and make room for engineered CAR-T cells, but in in vivo CAR-T generation, no such depletion is needed, thus preserving the patient's intact immune system. Moreover, in vivo CAR-T cells are less likely to be exhausted compared to ex vivo CAR-T, making them potentially more active, leading to better anti-cancer effects," explained Interius CEO Phil Johnson.
The development of in vivo CAR-T involves various vectors. Companies like Interius, Umoja, and Kelonia use engineered lentiviral vectors, which can permanently alter immune cells. Capstan, Myeloid Therapeutics, Orbital, and Orna are using LNP (lipid nanoparticle) delivery systems to induce a transient surge of CAR expression in target cells. Pharmaceutical giants are also entering the in vivo CAR-T space. For instance, Sanofi disclosed last year that it had three in vivo CAR-T projects in preclinical development, while AbbVie announced a potential $1.44 billion collaboration with Umoja in January of this year for the development of in situ CAR-T therapies. Astellas also announced a collaboration with Kelonia in February to develop in vivo CAR-T...
Overall, multiple in vivo CAR-T candidates are either already in or about to enter clinical trials. "By 2026, we will learn more from the trial results of in vivo CAR-T as they are revealed," said Capstan’s CSO Adrian Bot.
Even before ex vivo CAR-T therapy entered clinical trials, researchers had been working on re-engineering lentiviral vectors to achieve in vivo reprogramming of immune cells.
Christian Buchholz, head of the molecular biotechnology and gene therapy research group at the Paul-Ehrlich-Institut, had early insights over 20 years ago. Based on a 2005 study published in Nature Biotechnology (which demonstrated the use of oncolytic virus, the measles virus, to introduce single-chain antibodies targeting tumor-specific CD38, EGFR, and EGFRvIII into tumor cells for directed anti-tumor effects), Buchholz wondered if a similar strategy could help lentiviral vectors target specific cell types. In 2008, he reported that lentiviral vectors targeting CD20 were preferentially taken up by B cells. In 2012, his research confirmed that CD8-targeted vectors could enter T cells.
However, turning lentiviral vectors into a drug required a powerful payload. When CAR-T therapy targeting B cells first entered clinical trials in 2010, Buchholz saw an opportunity. In 2018, he published his latest findings, showing that lentiviral vector CD8-LV (specifically targeting human CD8 cells) could generate human CD19-CAR T cells in vivo, leading to the depletion of CD19 B cells.
Some biotech companies began trying to develop this strategy, making adjustments, including choices of cell-targeting fragments and therapeutic payloads.
The team at Interius chose to incorporate anti-CD7 antibody fragments into the lentiviral vector to target both CD7-positive T cells and NK cells. In contrast, Umoja's strategy is to enable the viral vector to effectively target several different "signals" on T cells. Their "early version" vector only used a CD3-targeting antibody fragment, with reprogramming efficiency between 10% and 15%. By fusing the CD3 antibody fragment with co-stimulatory domains like CD80 and CD58, their VivoVec vector significantly increased reprogramming efficiency to about 50%.
A study published on August 24th in Blood demonstrated that in non-human primates, VivoVec particles injected without prior chemotherapy led to the generation of a large number of CD20 CAR-T cells, with B cells being completely depleted for over 10 weeks. These data support further clinical development of the VivoVec platform.
To obtain human data, Interius and Umoja are currently recruiting patients for Phase I cancer trials. On July 31st, Umoja announced that its in-situ generated CD19 CAR-T therapy, UB-VV111, received FDA approval for an Investigational New Drug (IND) application to treat hematologic malignancies. The Phase I study (NCT06528301) will recruit patients with relapsed/refractory large B-cell lymphoma (LBCL) and chronic lymphocytic leukemia (CLL) to evaluate the safety, tolerability, and clinical anti-tumor activity of UB-VV111. However, conducting in vivo CAR-T clinical trials in hematologic malignancies presents challenges. On one hand, ex vivo CAR-T therapies are already highly effective. On the other hand, if a patient does not respond or relapses after ex vivo CAR-T therapy, oncologists may be reluctant to repeat essentially the same therapeutic approach. “Our advisors (practicing physicians) told us they wouldn’t let patients who relapse after CD19 CAR-T treatment receive another CD19 CAR-T,” said Phil Johnson. Therefore, Interius is targeting CD20 as the first clinical target for its lead product. The related clinical trial (NCT06539338) is currently recruiting patients.
Umoja is more optimistic about the development prospects of in vivo CAR-T targeting CD19. However, to avoid the issue of "repeated" treatments, the company is also advancing an in vivo CAR-T targeting CD22. According to Buchholz, a key concern is the cell-targeting selectivity of the vectors. “A big question in clinical settings is whether the selectivity of these vectors is maintained,” he said. This is critical because "off-target" effects could lead to safety issues, including carcinogenic risks, which regulators will focus on. More directly, researchers will pay attention to cytokine release syndrome (CRS) and immune cell-related neurotoxicity syndrome (ICANS). Additionally, the immunogenicity of in vivo CAR-T must also be considered. Since in vivo CAR-T therapy does not require lymphodepletion and cannot deplete the immune system, the patient’s immune system remains intact, which could potentially lead to “rejection” of the vector. “I’m very excited about the upcoming trials, but I also have some concerns, and we will have to wait and see,” said Buchholz.
In addition to cancer, in vivo CAR-T therapy for autoimmune diseases is also progressing. This hope stems from preliminary evidence suggesting that ex vivo CAR-T can induce immune reset by killing B cells, potentially curing lupus. Interius is prioritizing the development of a CD19-targeted in vivo CAR-T for clinical studies in autoimmune diseases, with plans to begin in 2025. Umoja and its partner IASO Biotherapeutics aim to start testing a CD22-targeted in vivo CAR-T in autoimmune disease patients in China next year.
In addition to lentiviral reprogramming, in vivo CAR-T using nanoparticle-based delivery systems is also under development. In 2017, just months before Buchholz demonstrated that lentiviral vectors could drive B cell depletion in mice, another team reported similar breakthroughs using nanoparticle-based systems. The success of mRNA-based COVID-19 vaccines later boosted the prominence of lipid nanoparticle (LNP) systems, with biotech companies such as Capstan and Orbital seizing the opportunity in in vivo CAR-T.
On March 20, 2024, Capstan completed a $175 million Series B financing round, with major investors including Johnson & Johnson, BMS, Eli Lilly, Bayer, Novartis, and Pfizer (Capstan's founding team also includes CAR-T pioneers Carl H. June and Bruce Levine, mRNA pioneer Drew Weissman, Nobel laureate in Physiology or Medicine in 2023). The funds will be used to advance Capstan's leading in vivo CAR-T candidate CPTX2309 (which uses LNP to deliver mRNA encoding anti-CD19 CAR to CD8 T cells) for early clinical proof of concept in autoimmune diseases, and further develop its pipeline based on LNP targeting.
For Capstan’s CSO Adrian Bot, a significant advantage of mRNA-loaded LNPs is their adjustability. This therapy only makes T cells temporarily express B cell-killing CARs, so dosage adjustments and repeated treatments can maximize efficacy while minimizing long-term safety concerns. Moreover, LNPs are easier to produce than lentiviral vectors.
"In terms of cost and complexity, producing LNPs and mRNA is much easier than producing entire viruses," said Evgin.
He is also collaborating with NanoVation to develop CAR-T using LNPs in vivo. However, LNPs need to be optimized for in vivo reprogramming. The LNP vectors used for COVID-19 vaccines were designed to be reactive carriers, maximizing immune responses to the viral antigens they encoded. But for in vivo CAR-T development, immune reactogenicity must be minimized. It is reported that Capstan has not used Teflon-like particles (which have strong immune reactivity and can persist in the body), but instead employs a snowflake-like design. These particles will decompose unless rapidly absorbed through receptor-mediated endocytosis.
In addition to T cells, scientists also hope to use lentiviral and LNP vectors for reprogramming other types of cells. Particularly, LNPs are well-suited for reprogramming myeloid cells.
Myeloid cells can "absorb" these vectors without the need for any cell-targeting components. "Even when LNPs are equipped with T cell-targeting antibodies, a large portion of these vectors usually end up in myeloid cells," said Daniel Getts, CEO of Myeloid Therapeutics. The company already has two in vivo CAR products (MT-302 and MT-303) in clinical trials, targeting solid tumors that express TROP2 or GPC3.
Orna is taking a "pan CAR" approach. The company claims its main LNP (without any cell-targeting components) can effectively reprogram T cells, NK cells, and macrophages in non-human primates. They have prioritized this for various blood cancer and autoimmune indications.
Carisma and its partner Moderna are prioritizing the development of an LNP-based GPC3 CAR macrophage (CAR-M) reprogramming therapy for liver cancer. Additionally, last month, both companies announced a collaboration to develop two in vivo CAR-M therapies for autoimmune diseases.
Over the past two or three years, there has been substantial progress in the development of in vivo CAR-T therapies globally. In China, multiple companies, including Jiin Biotech, Yantai Biotech, Xianbo Biotech (in collaboration with Orna), Bosunji, Baite Biotech, and Ucardi, are also exploring in vivo CAR-T.
Notably, in October 2024, Jiin Biotech announced the successful treatment of a patient with relapsed/refractory acute B-lymphoblastic leukemia (B-ALL) using a CD19-targeted in vivo CAR-T therapy (lentiviral delivery) in an investigator-initiated clinical trial (IIT), achieving complete remission (CR). This is the first publicly disclosed clinical data on in vivo CAR-T treatment for relapsed/refractory B-ALL. Additionally, a patient with high tumor burden relapsed and refractory diffuse large B-cell lymphoma treated with the company’s in vivo CAR-T therapy also achieved CR and was discharged. These positive results will undoubtedly boost the confidence of in vivo CAR-T developers. We look forward to technological breakthroughs that will lead to the early approval of in vivo CAR-T and benefit more patients.