Checkpoint Inhibitor a New Approach to Jump-Start a Waning Response to CAR T-Cell Therapy

January/February 2019, Vol 10, No 1 | Payers’ Perspectives In Oncology: ASH 2018 Highlights

San Diego, CA—Although chimeric antigen receptor (CAR) T-cell therapy has achieved remarkable long-lasting remissions in B-cell malignancies, in approximately 60% of the cases, the initial response wanes over time because of “immune exhaustion.”

The use of a checkpoint inhibitor to boost immune response to CAR T-cell therapy is gaining traction as an attractive approach, according to 2 early studies presented at ASH 2018.

Both studies looked at the use of pembrolizumab (Keytruda) to augment CAR T-cell response—one study was in patients with relapsed or refractory non­-Hodgkin lymphoma (NHL) and one in pediatric patients with relapsed B-cell acute lymphoblastic leukemia (ALL).

Relapsed or Refractory B-Cell NHL

In a small, single-center study, 12 patients with NHL that progressed or relapsed who have previously received CD19 CAR T-cell therapy were given fixed-dose intravenous pembrolizumab 200 mg every 3 weeks until disease progression or toxicity. Pembrolizumab was the only treatment they received after the CAR T-cell infusion.

“The evidence suggests that immune exhaustion may contribute to progressive disease or relapse [with CAR T-cell], since two-thirds of patients at progression have PD-L1 expression on biopsies. The idea is to restart the immune system using a checkpoint inhibitor,” said lead investigator Elise A. Chong, MD, Hematology and Oncol­ogy Fellow, Lymphoma Program, Abramson Cancer Center, University of Pennsylvania, Philadelphia.

“We don’t know the best time to administer checkpoint inhibition relative to CAR T-cell, and several multicenter trials are studying this,” Dr Chong added.

Of the 12 patients included in this early trial, 11 had diffuse large B-cell lymphoma and 1 had follicular lymphoma. Of the 12 patients, 9 had progressive disease at a median of 3.3 months before they received pembrolizumab.

In the other 3 patients, the disease relapsed more than 1 year after the CAR T-cell therapy infusion, at which point they received pembrolizumab therapy.

The combination was safe in this early study, and CAR T-cells re-­expanded in 8 of 11 patients after pembrolizumab therapy, signaling a “boosting” of immune activity.

The only grade ≥3 adverse event considered possibly related to pembroliz­umab was neutropenia in 3 patients.

Among the 12 patients evaluable for response, the best overall response rate was 25%, with 1 complete response and 2 partial responses; 1 patient had stable disease, and 8 had progressive disease.

“The optimal timing of pembro­lizumab is a big question, and it appears that 1 year out is not optimal for most patients. Our study suggests that within 3 months is a better time frame, and we may be able to start pembrolizumab much sooner, or even before CAR T-cell. We treated some patients as early as 13 days after CAR T-cell infusion without toxicity,” Dr Chong said. “We need to determine the optimal window of time to give pembrolizumab, and balance that with toxicity.”

Several multicenter clinical trials are exploring the optimal timing of a checkpoint inhibitor in patients with lymphoma that progresses or relapses with CAR T-cell therapy in the third-line setting.

Those studies are studying different CAR T-cell therapy and checkpoint inhibitor combinations, such as tisagenlecleucel (Kymriah) and pembrolizu­mab; axicabtagene ciloleucel (Yescarta) and atezolizumab (Tecentriq); and JCAR017 and durvalumab (Imfinzi); as well as different intervals between CAR T-cell therapy and checkpoint inhibitor administration.

Relapsed or Refractory B-Cell ALL

A single-center study at the Children’s Hospital of Philadelphia enrolled 14 heavily pretreated pediatric patients, including 13 patients with relapsed B-cell ALL and 1 patient with B lymphoblastic lymphoma who had received CD19 CAR T-cell therapy and then a PD-1 inhibitor—pembrolizumab or nivolumab (Opdivo), respectively, which was started 14 days or later after CAR T-cell infusion and resolution of any cytokine release syndrome.

The 14-day interval was chosen because CAR T-cell levels typically decline by that time, and cytokine release syndrome tends to occur during that interval.

With a median of 13.3 months after receiving a checkpoint inhibitor, 50% of the enrolled patients maintained a partial or a complete response.

“We showed that the combination is safe, and that PD-1 checkpoint inhibitor may be used to improve CAR T-cell persistence. More than 18 months later, at least half of the patients exhibited response with a checkpoint inhibitor. The data suggest that it might be of benefit to patients with early B-cell recovery and bulky extramedullary disease,” said senior author Shannon L. Maude, MD, PhD, Attending Physician, Division of On­cology, Children’s Hospital of Philadelphia, PA.

Commenting on this study, Joseph C. Alvarnas, MD, Associate Clinical Professor, Department of Hematology & Hematopoietic Cell Transplantation, City of Hope, Duarte, CA, said, “When people don’t respond as well as they should, it may be that the T-cells are depleted and functionally exhausted. The mechanism of exhaustion is in part mediated by checkpoint-related killing, and by thwarting that process with pembrolizumab or nivo­lumab, this small study suggests that we can improve the quality and duration of response in patients we would expect not to do as well.”

In the study, 3 groups of patients received pembrolizumab—4 patients with partial or no response to CAR T-cell therapy (all 4 patients had disease progression with pembrolizumab); 6 patients with poor persistence of CAR T-cell therapy (pembrolizumab resulted in return of B-cell aplasia, a sign of CAR T-cell function) and sustained complete response; and 4 patients with bulky extramedullary disease (2 sustained complete response and 2 partial response).

No unexpected or fatal events were reported in the study. Within 2 days of initiating pembrolizumab therapy, cytokine release syndrome symptoms and fever were observed in 3 of the 14 patients; 4 patients had grade 3 or 4 cytopenia. No graft-versus-host disease flares were ­observed.

Related Articles