Biomarker predicts poor response to CAR-T, points way to personalized, gene-edited therapy
Click Here to Manage Email Alerts
Researchers have identified CD58 status as a predictive marker for durable responses to CD19-directed chimeric antigen receptor T-cell therapy in patients with large B-cell lymphoma.
The results, presented at the virtual ASH Annual Meeting and Exposition, showed that nearly all patients with CD58 alterations had disease progression after receiving CD19-directed CAR T-cell therapy. Researchers have used this knowledge to develop an investigational CAR that they hope will point the way toward use of gene editing to personalize CAR T-cell therapy and maximize durability of response to the treatment.
“CD19 CAR T cells have revolutionized treatment of large B-cell lymphoma, resulting in a durable complete response rate in 40% to 50% of patients,” Robbie G. Majzner, MD, assistant professor of pediatrics at Stanford University School of Medicine, said during a presentation. “This has nicely held up in real-world experience at centers using commercial [CAR T cells],” which he said have demonstrated similar complete response rates.
Majzner said his group wanted to focus on patients who are unable to achieve durable complete responses to CAR T-cell therapy because of the poor outcomes they typically face.
“We know that CAR T cells are able to induce long-term remissions in some patients, so we felt that if we could increase the initial response rate and find the mechanisms of resistance and engineer around them, then we could cure more patients,” he added.
Majzner and colleagues evaluated the CD58 status of 51 patients with relapsed or refractory large B-cell lymphoma who received commercially available axicabtagene ciloleucel (Yescarta; Kite Pharma/Gilead) at Stanford University.
The researchers assessed CD58 status using immunohistochemistry of tumor biopsies and/or deep sequencing of circulating tumor DNA using a proprietary analysis method known as CAPP-Seq. Researchers defined CD58 aberrations as lack of CD58 expression using immunohistochemistry or evidence of a mutation with CAPP-Seq.
Results showed 12 patients (24%) in the study group had a CD58 aberration. Median PFS was significantly shorter among these patients (3 months vs. not yet reached; P < .0001).
One patient with a CD58 aberration had a durable, complete response to axicabtagene ciloleucel. The reaming 11 patients with CD58 aberrations had an initial response to therapy but eventually experienced disease progression.
Majzner noted that despite a near-lack of complete response to therapy, partial responses were more common among patients with CD58 aberrations compared with those with intact CD58 (58% vs. 10%; P <. 001).
The investigators then developed a preclinical model of leukemia in which they knocked out CD58 in mice using CRISPR-CAS gene editing. They found that mice with CD58 knocked out had only partial responses to therapy, and all of them had disease progression and death.
“CAR T cells lose their ability to kill a tumor cell when CD58 is knocked out,” Majzner said. This was true for mouse models treated with any of the three CAR T-cell therapies they tested, which included axicabtagene ciloleucel.
“Patients with a CD58 alteration rarely have long-term durable responses to treatment with axicabtagene ciloleucel,” he added.
Majzner and colleagues constructed a new CAR molecule that integrated a CD2 costimulatory domain and an additional CD2 receptor. This molecule was able to overcome CD58 loss in preclinical models and showed antitumor activity in large B-cell lymphoma.
“CD58-CD2 is a novel axis of CAR resistance that we have uncovered through our deep correlative studies,” Majzner said. “Loss and/or mutations in CD58 portend a poor response to CD19 CAR T cells, and we will see more of this as more CAR T cells are used in the clinic.”
Majzner said the results demonstrate an ability to engineer CAR T cells to integrate CD2 signaling — which is crucial to CAR activation — in such a manner that it is possible to reestablish its efficacy regardless of the presence of CD58 on the tumor cell.
“We think this will be important in other malignancies because CD58 is common in other cancers, including myeloma, Hodgkin lymphoma and some solid tumors,” he added.
Perspective
Back to Top
Catherine Bollard, MD
These results suggest there is an opportunity to personalize immunotherapy for large B-cell lymphoma. Majzner and colleagues created a modified version of the immunotherapy axicabtagene ciloleucel, focusing on the role of a protein called CD58 to overcome the fact that approximately one-quarter of patients do not respond well to this therapy. The results show it is possible to improve our capacity to administer effective T-cell therapies.
The George Washington University
Perspective
Back to Top
Samantha Bucktrout, PhD
This study, which bypassed a resistance mechanism of large B-cell lymphoma to CAR T-cell therapy, has broader implications for improving T-cell therapy in hematological and solid tumor types. Moving the location of the costimulatory signaling domain, CD2, from the CAR to an independent membrane receptor, mirroring the natural scenario, resulted in superior CAR T-cell activity in preclinical models.
These data support deeper exploration into the positioning of other costimulatory signaling domains that regulate key aspects of T-cell biology relevant for cellular therapy, such as expansion, effector function, differentiation, and in vivo distribution and durability. Further research in this area would open up infinitely more engineering strategies to be tested in the T-cell therapy space.
Collapse
Majzner RG, et al. Abstract 556. Presented at: ASH Annual Meeting and Exposition (virtual meeting); Dec. 5-8, 2020.
Click Here to Manage Email Alerts