Enhanced Costimulatory Signaling Improves CAR T-cell Effector Responses in CLL.

CD19-redirected chimeric antigen receptor (CAR) T cells have shown remarkable activity against B-cell cancers. While second-generation CARs induce complete remission in >80% of patients with acute lymphoblastic leukemia, similar monotherapy induces long-term remissions in only 26% of patients with chronic lymphocytic leukemia (CLL). This disparity is attributed to cell-intrinsic effector defects in autologous CLL-derived T cells. However, the mechanisms by which leukemic cells impact CAR T-cell potency are poorly understood. Herein we describe an in vitro assay that recapitulates endogenous CLL-mediated T-cell defects in healthy donor CAR T cells. Contact with CLL cells insufficiently activates, but does not irreversibly impair, CAR T-cell function. This state is rescuable by strong antigenic stimulation or IL2, and is not driven by immune suppression. Rather, this activation defect is attributable to low levels of costimulatory molecules on CLL cells, and exogenous costimulation enhanced CAR T-cell activation. We next assessed the stimulatory phenotype of CLL cells derived from different niches within the same patient. Lymph node (LN)-derived CLL cells had a strong costimulatory phenotype and promoted better CAR T-cell degranulation and cytokine production than matched peripheral blood CLL cells. Finally, in vitro CD40L-activated CLL cells acquired a costimulatory phenotype similar to the LN-derived tumor and stimulated improved CAR T-cell proliferation, cytokine production, and cytotoxicity. Together, these data identify insufficient activation as a driver of poor CAR T-cell responses in CLL. The costimulatory phenotype of CLL cells drives differential CAR T-cell responses, and can be augmented by improving costimulatory signaling.
Significance: CLL cells insufficiently activate CAR T cells, driven by low levels of costimulatory molecules on the tumor. LN-derived CLL cells are more costimulatory and mediate enhanced CAR T-cell killing. This costimulatory phenotype can be modeled via CD40 L activation, and the activated tumor promotes stronger CAR T-cell responses.
Competing Interests: M.A. Collins reports grants from Parker Institute for Cancer Immunotherapy during the conduct of the study. A.P. Kater reports grants and other from Abbvie, AstraZeneca, BMS; and other from Janssen and LAVA outside the submitted work; in addition, A.P. Kater has a patent to Janssen pending and a patent to LAVA pending. C. Sun reports grants from Genmab outside the submitted work. A. Wiestner reports grants from Pharmacyclics, Acerta, Merck, Nurix, Verastem, and Genmab outside the submitted work. J.J. Melenhorst reports grants from Parker Institute for Cancer Immunotherapy during the conduct of the study; personal fees from IASO Biotherapeutics, Poseida Therapeutics, and Kite Pharma outside the submitted work; in addition, J.J. Melenhorst has a patent to Methods for improving the efficacy and expansion of immune cells issued, a patent to Biomarkers predictive of therapeutic responsiveness to chimeric antigen: issued, a patent to Methods of making chimeric antigen receptor - expressing cells issued, a patent to Methods for improving the efficacy and expansion of chimeric antigen receptor: issued, a patent to CAR T-cell therapies with enhanced efficacy issued, and a patent to Biomarkers predictive of cytokine release syndrome issued. No disclosures were reported by the other authors.
(© 2022 The Authors; Published by the American Association for Cancer Research.)