Thank you for your interest in CCO content. As a guest, please complete the following information fields. These data help ensure our continued delivery of impactful education.
Become a member (or login)? Member benefits include accreditation certificates, downloadable slides, and decision support tools.
Division of Medical Oncology
Department of Medicine
Harvard Medical School
Dana-Farber Cancer Institute
Caron A. Jacobson, MD, has disclosed that she has received consulting fees from Celgene, Humanigen, Kite, Novartis, Pfizer, and Precision BioSciences.
With the development of CD19-directed CAR T-cells, clinicians can now offer potentially curative therapy to approximately 40% to 50% of patients with B-cell acute lymphoblastic leukemia and aggressive B-cell non-Hodgkin lymphoma who previously had no curative option. However, approximately 20% of patients do not respond to CAR T-cell therapy, and up to 40% of responders will relapse. Fortunately, promising clinical data are emerging on novel strategies combining CAR T-cells with targeted agents in these nonresponding and relapsing patients.
CAR T-Cell Combination Therapy: Rationale
The mechanisms underlying innate and acquired resistance to CAR T-cell therapies remain to be fully determined. Data indicate that CAR T-cells upregulate markers of T-cell activation and immunomodulatory receptors upon tumor antigen engagement and activation. Upregulated markers include markers of exhaustion (including PD-1, LAG-3, and TIM-3), activating molecules (ICOS and 4-1BB), and apoptosis markers, such as cleaved caspase 3—consistent with T-cell exhaustion and cell death. These findings suggest that combining CAR T-cells with agents targeting T-cell exhaustion pathways may improve remission durability.
This approach was further supported by a case report of diffuse large B-cell lymphoma (DLBCL) that progressed following anti-CD19 CAR T-cell therapy. The patient had a clinically significant response when subsequently treated with an anti–PD-1 antibody, in parallel with CAR T-cell expansion and decreased PD-1 levels on the CAR T-cells.
Multiple phase I/II trials are investigating the use of anti–PD-1/PD-L1 therapy after CAR T-cell therapy in relapsed/refractory lymphomas, such as pembrolizumab following failure or relapse with tisagenlecleucel, pembrolizumab consolidation after infusion of the investigational CD19/CD22-directed AUTO3 (ALEXANDER), and durvalumab in combination with lisocabtagene maraleucel, formerly JCAR017 (PLATFORM). Phase I data are now available from the ZUMA-6 trial evaluating atezolizumab after axicabtagene ciloleucel (axi-cel).
Phase I Results for ZUMA-6 Trial: Axi-cel Followed by Atezolizumab
At ASH 2018, my colleagues and I presented end of phase I results for the ongoing phase I/II ZUMA-6 trial in refractory DLBCL. Following axi-cel infusion, the 12 patients in phase I received 4 doses of the anti–PD-L1 antibody atezolizumab every 3 weeks, with the first atezolizumab dose administered 3 weeks (cohort 1, n = 3), 2 weeks (cohort 2, n = 3), or 1 day (cohort 3, n = 6) after infusion. Enrollment in each cohort followed a 3 x 3 design. At baseline, all but 1 of 10 patients tested had PD-L1 staining in the tumor or its microenvironment.
We observed 1 dose-limiting toxicity (DLT) of prolonged grade 4 cytopenias in cohort 3; this cohort was expanded to 6 patients without further DLTs. All 12 patients experienced cytokine-release syndrome (CRS), but this only reached grade 3 in 3 patients. Neurotoxicity occurred in 9 patients, including 5 grade 3 cases and 1 grade 4 case. The ORR was 92%, with 58% achieving a CR. Of interest, there were 3 late conversions from an earlier stable disease/PR to CR from Months 6-9. Based on these results, we selected the cohort 3 dosing schedule for the phase II dose expansion.
We also tracked axi-cel blood levels over the first year. Compared with patients with DLBCL who received axi-cel monotherapy on the ZUMA-1 trial, axi-cel expansion was greater and more persistent in those treated with combination therapy. Furthermore, we found similar safety profiles and no significant differences in levels of key cytokines associated with CRS and neurotoxicity in ZUMA-6 compared with ZUMA-1.
Although it is premature to draw conclusions from this small number of patients in our ZUMA-6 analysis, combining PD-L1 blockade with axi-cel appears to have favorable efficacy and a similar safety profile compared with axi-cel alone. The phase II dose expansion will provide important confirmatory data on safety and efficacy, including whether we again observe deepening responses and greater persistence of CAR T-cells over time. Results from ZUMA-6 may also shed light on pretreatment biomarkers that predict response to this combination or other PD-1/PD-L1–targeting strategies, such as using CRISPR/Cas9 technology to knockout the gene encoding PD-1 during CAR T-cell manufacturing. Clinical data will help elucidate the key mechanisms and markers of resistance to CAR T-cell therapies, ultimately improving individualized care of our patients experiencing resistance or relapse.
To read about the latest clinical data on CAR T-cell therapy in relapsed/refractory multiple myeloma, please view the first commentary in this series by Noopur Raje, MD.