Early clinical data from Legend Biotech Corporation (Nasdaq: LEGN) suggest that LB2501, its investigational in vivo CD19/CD20 dual-targeting CAR-T cell therapy, can generate functional CAR-T cells directly within the patient and produce deep responses in relapsed or refractory B-cell non-Hodgkin lymphoma without lymphodepletion — a combination that, if confirmed in larger cohorts, would distinguish it from every approved CAR-T product currently on the market. Presented at the European Hematology Association 2026 Congress as a late-breaking oral, the Phase I data from NCT07002112 enrolled 12 heavily pretreated patients and reported a 100% objective response rate and an 83.3% complete response rate at the higher of two dose levels tested — results that, while preliminary and drawn from six patients, are difficult to ignore given the absence of any serious adverse events, ICANS, or deaths.
The significance lies not just in the response numbers but in the mechanism. All approved CAR-T therapies — including Bristol Myers Squibb's Breyanzi (lisocabtagene maraleucel), Gilead's Yescarta (axicabtagene ciloleucel), and Legend's own Carvykti (ciltacabtagene autoleucel) for multiple myeloma — require ex vivo manufacturing: a patient's T cells are collected, genetically engineered outside the body, expanded over weeks, and reinfused. This process is logistically complex, expensive, and time-consuming, and it excludes patients whose disease progresses too rapidly or whose T cells are too compromised to manufacture. LB2501 takes a different approach: a single intravenous infusion of a lentiviral vector, built on Legend's proprietary TaVec platform, is designed to reprogram the patient's own T cells in vivo, eliminating the manufacturing step entirely.
The study provides one of the clearest demonstrations to date that functional CAR-T cells can be generated directly inside patients and produce meaningful anti-tumor responses without traditional manufacturing or lymphodepletion.
Trial data
The open-label Phase I dose-escalation study enrolled 12 adults with relapsed or refractory B-cell non-Hodgkin lymphoma across two dose levels and was conducted without lymphodepleting chemotherapy. At the higher dose level (DL2), LB2501 achieved a 100% objective response rate (6/6) and an 83.3% complete response rate (5/6), with responses observed across diffuse large B-cell lymphoma, mantle cell lymphoma, and follicular lymphoma. All responses remained ongoing at the data cutoff, while pharmacokinetic analyses confirmed dose-dependent in vivo CAR-T expansion in all DL2 patients.
Safety was notable for the absence of dose-limiting toxicities, serious adverse events, ICANS, or deaths. Cytokine release syndrome occurred in 66.7% of patients and infusion-related reactions in 75%, but all events were Grade 1–2. Translational analyses showed no evidence of off-target cellular transduction, while vector integration patterns were reported as highly polyclonal, supporting the feasibility of in vivo T-cell engineering.