Feasibility and tolerability of a CD4 re-directed chimeric antigen receptor T cell therapy (CD4CAR) in hematologic malignancies: Initial Results from multi-institutional first-in-human trials

Abstract

Abstract Background CD4 is expressed across several hematopoietic malignancies, including T-cell malignancies (TCM), chronic myelomonocytic leukemia (CMML), and subsets of acute myeloid leukemias (AML), but not on normal hematopoietic stem cells or other tissues. In addition, CD4 is expressed on immune-suppressive cell populations, including regulatory T cells (Tregs), myeloid-derived suppressor cells (MDSCs), and monocytic leukemia stem cells (LSCs), including those associated with venetoclax resistance. These features collectively identify CD4 as a promising but underexplored therapeutic target. Methods We initiated multi-center, first-in-human (FIH) phase I clinical trials to evaluate the safety, feasibility and potential immune modulating effects of CD4-targeted chimeric antigen receptor (CD4CAR) T cell therapy in patients with relapsed or refractory CD4-positive hematologic malignancies. The CD4CAR construct is a third-generation lentiviral vector–encoded receptor that incorporates both CD28 and 4-1BB costimulatory domains to enhance T cell activation, persistence, and antitumor function. Autologous T cells were transduced and expanded ex vivo and infused following lymphodepletion with either fludarabine/cyclophosphamide or bendamustine. A standard 3+3 dose-escalation design was employed to determine the recommended phase 2 dose, and treatment-related toxicities, pharmacodynamics, and immunologic biomarkers were systematically assessed. Adverse events were graded per CTCAE v5.0. Results Fourteen patients were treated: CD4+ AML (n=2), CMML (n=3), and TCM (n=9). Median age was 65.5 years (range 18-78), and patients received a median of 2 prior therapies (range 1–5). CD4CAR was infused at doses ranging from 8.0×10⁴ to 1.6×10⁵ CAR+ cells/kg (Cohorts 1 and 2). Cell products were successfully manufactured for all patients without the need for CD4+ cell, negative selection. CD4CAR was well tolerated, with no dose-limiting toxicities. Grade ≥3 lymphopenia resolved by day 30 without DLT qualifying infections. Grade 1 cytokine release syndrome (CRS) occurred in two patients and resolved without steroids or tocilizumab. No immune effector cell - associated neurotoxicity was observed. The low rate of CRS may reflect may be attributable to the final product composition, which consisted predominantly of CD8+ CAR+ T cells with minimal or absent CD4+ CAR+ T cells. CD4CAR expansion was observed in blood and marrow, with a median time to peak CAR T cell count of 14.5 days (range 7-120) and a median peak of CAR+/CD3+ percentage at 9.5 days (range 7-150). CD4CAR Persistence extended beyond day 90 in most patients and up to day 365 in one case with evidence of in vivo Tcm phenotype. Updated timepoint analyses continue to demonstrate reductions in Tregs and MDSCs, particularly monocytic subsets (CD45+/CD11b+/CD193−/CD84−/CD16+14+ or 16−), across AML, CMML, and TCM. LSCs were detected at baseline in several AML and CMML patients and showed early and durable clearance post-infusion. Among evaluable patients (n=7), the overall response rate was 71.4%, including 57.1% complete responses. All complete responders remain in remission at last follow-up (median 5.7 months). During manufacturing, fratricide-related T cell exhaustion remained minimal despite sustained target (CD4+ T cels) CAR engagement. Conclusions CD4CAR therapy is feasible and safe, with evidence of biologic activity, immune modulation, as well as early clinical responses across CD4-expressing hematologic malignancies. The predominance of CD8+ CAR T cells in the infused product did not impair in vivo expansion or persistence. These findings support the continued development of CD4-targeted CAR T cell therapies.