In a sponsor presented, non peer reviewed readout at the 2026 European Hematology Association meeting, the Hong Kong listed Chinese biotech said two off the shelf CAR T (engineered immune cell) candidates produced no severe cytokine storms, neurotoxicity, or graft versus host…
CARsgen Therapeutics reported on Sunday that two of its experimental donor-derived cell therapies posted a clean early safety profile in 21 heavily pretreated blood-cancer patients, a cautious positive signal in a category where prior candidates have stumbled on graft-versus-host disease and immune rejection.
According to a company press release presented at the 2026 European Hematology Association (EHA) congress, no patient in either cohort experienced grade 3 or higher cytokine release syndrome (CRS), the systemic inflammatory reaction common to CAR-T infusions. No patient developed immune effector cell-associated neurotoxicity syndrome (ICANS), and no patient developed graft-versus-host disease (GvHD) at the doses reported.
The two products are allogeneic CAR-T therapies. They are T cells collected from healthy donors, engineered in a lab to recognize a cancer target, and stored for off-the-shelf use, rather than manufactured one batch at a time from each individual patient as today's approved CAR-T products are.
CT0596, which targets the BCMA protein on plasma cells, was given to 8 patients with relapsed or refractory multiple myeloma (n=6) or primary plasma cell leukemia (n=2) at a single 4.5×10⁸ CAR+ T cell dose. The cohort was heavily pretreated, with a median of 3.5 prior lines of therapy (range 2 to 6), five patients at ISS Stage III, five with high-risk cytogenetics, and one with extramedullary disease. As of a May 10, 2026 data cut with median follow-up of 6.97 months, the company said all eight patients had treatment-emergent adverse events, mostly the hematologic toxicities typical of lymphodepletion and CAR-T, with the severe-immune-syndrome column empty.
CT1190B, a dual-targeting allogeneic CAR-T directed at both CD19 and CD20, was given to 13 patients with relapsed or refractory B-cell non-Hodgkin lymphoma. CARsgen did not provide the same granular safety breakdown for that cohort in the release.
The efficacy signals were the kind of early data an analyst would file away rather than celebrate. CARsgen said the patients who responded reached stringent complete response (sCR), the deepest measurable remission category, including cases of minimal residual disease (MRD) negativity at a 10⁻⁶ sensitivity threshold. The company also disclosed that one multiple-myeloma patient who failed to respond at the 3.0×10⁸ dose level was retreated at 4.5×10⁸, and that the heavier patient with primary plasma cell leukemia required full-dose lymphodepletion plus a higher 6.0×10⁸ cell dose to reach sCR. That detail matters: the dose and conditioning regimen are still being calibrated, and the safety profile could move as those variables change.
The structural bet CARsgen is making is the same one Allogene Therapeutics, Crispr Therapeutics, Caribou Biosciences, and Precision Biosciences have been making for the better part of a decade. Approved CAR-T products, including Kymriah, Yescarta, Tecartus, Breyanzi, Carvykti, and Abecma, are autologous, manufactured from each patient's own cells, with multi-week production times, complex logistics, and list prices north of $500,000 per course. An allogeneic platform that holds up clinically would let a hospital reach for a frozen vial the way it reaches for a unit of platelets.
That field has a documented history of early readouts that later required redesign or quietly disappeared. Allogene's ALLO-501 and ALLO-501A, Crispr's CTX110, Precision's PBCAR19, and Cellectis's UCART19 all reported encouraging Phase 1 data before running into GvHD, durability, or manufacturing issues at scale. The CARsgen dataset, presented as posters at EHA 2026, has not been peer-reviewed and rests on a single-arm, non-randomized, dose-finding design with the median patient followed for less than seven months.
What to watch next: the underlying EHA posters, peer-reviewed publication, longer follow-up, the higher dose-escalation cohorts, and whether any patient in either study eventually develops GvHD, ICANS, or late CRS as the n grows. The first independent confirmation will be whether the safety column stays empty as the cohorts expand.