Preclinical development of anti-CD21 chimeric antigen receptor T cells to treat T cell acute lymphoblastic leukemia.

Patients with relapsed/refractory (r/r) T cell acute lymphoblastic leukemia (T-ALL) have a dismal prognosis, highlighting the urgent need for effective therapies. Chimeric antigen receptor (CAR)-T cell approaches targeting pan-T cell antigens may be limited by T cell aplasia and fratricide, necessitating "rescue" allogeneic hematopoietic stem cell transplantation. In this study, we identify CD21, a pan-B cell marker, as a promising target for T-ALL immunotherapy. CD21 is expressed in 50% of T-ALL cases at diagnosis but in fewer than 10% of mature T cells. We observed that CAR-T cells targeting membrane-distal CD21 epitopes were ineffective, likely because of the bulky, glycosylated nature of the antigen. However, when we engineered CAR-T cells to target membrane-proximal CD21 epitopes using an antigen-binding fragment (Fab)-CAR design, we demonstrated robust activity against T-ALL cell lines, primary tumors, and patient-derived xenografts in both in vitro and in vivo models. The enhanced efficacy of this Fab-CAR design was driven by its high stability and reduced surface expression, addressing limitations of traditional CAR constructs. In addition, pharmacological inhibition of the phosphatidylinositol 3-kinase axis up-regulated CD21 expression in T-ALL, further enhancing the potency of anti-CD21 CAR-T cells in vitro and in a patient-derived xenograft in vivo model. This study establishes CD21 as a viable CAR-T target and highlights advances in CAR design for bulky antigens, as well as the potential for pharmacological strategies to augment target expression. Anti-CD21 CAR-T cells represent a promising therapeutic option for improving outcomes for patients with T-ALL.