Even after a decade of treating patients with hematologic malignancies with chimeric antigen receptor (CAR) T-cell (CAR-T) therapies, researchers are still trying to understand why most patients eventually relapse. Equally puzzling to some scientists is the question of these cellular interventions cause lasting remission at all in many patients.
Given that typical cancers consist of diverse cells, including those that do not express the antigens targeted by CAR-T cells, one would expect relapses through antigen escape to be much more common than currently observed in practice, explained Joshua Brody, MD, director of the Lymphoma Immunotherapy Program at The Tisch Cancer Institute at Mount Sinai in New York, New York. “In CAR-T cells, maybe [only] 40% of patients have this antigen escape problem. It’s weird to us that it’s not 100%,” he said.
For instance, even though most patients with B-cell acute lymphocytic leukemia (ALL) possess over 1% of CD19-negative cells at diagnosis, the incidence of relapse after CD19-targeted therapy only approximates 20%.1,2 And, in patients with diffuse large B-cell lymphoma (DLBCL), similar response rates to CAR-T cell therapy have been observed regardless of tumoral expression of CD19.3 To Brody and colleagues, such observations suggest that CAR-T cells use mechanisms independent of antigen targeting to eliminate tumor cells.
A study published by Brody’s team and collaborators at Kite Pharma in Cancer Discovery in December 2020 offered one explanation.4 Experiments in animal models suggested that CAR-T cells can kill off-target cells that are in the vicinity of the cells they’re designed to target, offering the first in vivo proof of localized bystander killing. This off-target effect is mediated by the interaction between the protein Fas—a cell death receptor expressed on many cellular surfaces—and its ligand, which is present on T cells. In fact, tumoral expression of Fas was predictive of survival in patients with DLBCL who were treated with anti-CD19 CAR-T cell therapy in the phase 1/2 ZUMA-1 trial (ClinicalTrials.gov identifier: NCT02348216).
“I think it’s becoming more and more [clear] that the CAR-Ts, in addition to their CAR interaction with the tumor antigen, rely on the Fas-Fas ligand…interaction to exert their killing. The next question is, ‘how do we manipulate these pathways safely to make CARs more potent?”” said Saad J. Kenderian, MB, ChB, a consultant in the division of hematology in the department of internal medicine at the Mayo Clinic in Rochester, Minnesota, who was not involved in the study.
The new research was the result of a serendipitous observation made during a clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated protein 9 (Cas9) screen exploring the genes that tumor cells use to either resist or facilitate cytotoxic T cell killing. The team noticed in cell culture that T cells engineered to target a specific protein would not only kill the “on-target” lymphoma cells expressing that protein, but also the cells that did not.
Further experiments pointed to Fas as a mediator of this process. When the researchers experimentally removed the gene encoding Fas from cultured lymphoma cells, they noticed that this protected both on-target and off-target cells. This was the case even as T-cells emitted the cell death-inducing molecules granzyme and perforin, which are thought to represent the main method of killing.
T cells require an interaction with their target antigen to kill a cell via the perforin and granzyme mechanism, but as long as off-target cells are in the direct vicinity of on-target cells, Brody’s results suggest that T-cells can eliminate them via the Fas-dependent mechanism. Essentially, “you kill the target cell you’re going after and just to be safe, you [also] kill the cell next door,” he said.
Further cell culture and mouse experiments bolstered this hypothesis. In one experiment, mice treated with murine CD19 CD3ζ-CD28 CAR-T cells had similar rates of survival irrespective of whether their lymphoma tumors consisted of mixed antigen-positive and negative cells or only of antigen-expressing cells. “However, if we gave Fas ligand-blocking antibody, then those mice died sooner,” Brody added.