A Phase II Trial of Anakinra for the Prevention of CAR-T Cell Mediated Neurotoxicity

Introduction: Chimeric antigen receptor (CAR)-T cell therapy is limited in most cases to inpatient use due to risk of severe treatment-related toxicities. The two primary toxicities observed with CAR-T therapy, cytokine release syndrome (CRS) and neurotoxicity, are associated with increased circulating inflammatory cytokines such as IL-6 and IL-1. Targeting IL-6 with tocilizumab is effective for treating CRS but not neurotoxicity. Anakinra is an FDA-approved recombinant IL-1 receptor antagonist that competitively inhibits IL-1 receptor signaling and therefore blocks downstream production of inflammatory cytokines including IL-6. Leveraging support from Kite Pharma, we opened an investigator-initiated clinical trial (NCT04150913) with the hypothesis that anakinra could be administered prophylactically to prevent severe CRS and neurologic events (NE) in patients receiving axicabtagene ciloleucel (axi-cel). Here we report preliminary outcomes of this study.

Study Design and Methods: This is a phase II single center, open-label study for patients ≥18 years old with relapsed or refractory large cell lymphoma. Patients must have progressed after ≥2 lines of systemic therapy but could not have CNS disease or have been previously treated with CAR-T therapy. Following leukapheresis and manufacturing, patients received 3 days of lymphodepleting chemotherapy (LDC, cyclophosphamide 500mg/m ² and fludarabine 30 mg/m ²) and 200 mg of subcutaneously administered anakinra starting 4 hours prior to axi-cel infusion and daily thereafter for a total of 7 days. CRS and NE were graded based on the Lee 2013 criteria and the CTCAE 4.03 criteria, respectively, to enable direct comparison to the pivotal Zuma-1 cohorts. The primary endpoint is the rate and severity of NE within the first 30 days of infusion; secondary endpoints include the incidence and severity of CRS and disease response. CAR-T cell expansion, serum cytokines, and circulating biomarkers of toxicity were measured at baseline, day 3, 7, 14, 21, and 28 post CAR-T cell infusion.

Results: Interim analysis of the first 6 patients demonstrated a median age of 68 (range 59-72). Patients included a diverse group of histologies including double-hit lymphoma (n=2), transformed indolent NHL (n=3), and DLBCL NOS (n=1). Two patients were considered primary refractory at time of enrollment. Pre-LDC baseline characteristics included a median SPD of 2819 mm ² (range 1063-5802), median LDH of 415 (range 147-497) which were comparable to the pivotal ZUMA-1 cohorts. Baseline ferritin, CRP, SAA and IL-15 were similar to the pivotal ZUMA-1 cohorts. While low-grade CRS was observed in 5/6 patients, no patients experienced severe CRS and median onset occurred on day +8 (range 1-8). Four patients did not experience any NE, while two patients experienced grade 3 NE on days +6 till +9 (somnolence) and +12 (global aphasia only, for one day) respectively. With a median follow-up of 4 months, the day +28 overall response rate was 100% (4 CRs, 2 PRs), with 4/6 patients having an ongoing complete response at last disease assessment. One patient was re-infused at progression and remains in a CR 3 months from re-infusion. Responses were seen despite varying CAR-T peak level with most patients demonstrating expansion in the lower quartile of the historic ZUMA-1 cohort. Median post-infusion peak of CRP, ferritin, IL-2, GM-CSF, IFNγ, IL-10, IL-6 and SAA were lower than that observed in the pivotal ZUMA-1 cohorts. All patients remain alive at time of data analysis.

Conclusions: With a limited number of patients analyzed thus far, anakinra appears to provide benefit to the toxicity profile of axi-cel, presenting reduced and/or delayed CRS and NE and a decrease in post-infusion inflammatory analytes, when compared to ZUMA-1 pivotal cohorts. No severe CRS was observed in this initial analysis and 2/6 patients experienced grade 3 NE (somnolence and global aphasia) after day 6. Despite CAR-T expansion in the lower quartile of that of ZUMA-1, we observed a 100% ORR with 4 patients remaining in CR at a median follow-up of 4 months. Additional subjects will be assessed to investigate the role of prophylactic anakinra in the management of CRS and NE, which has potential for making axi-cel treatment an outpatient therapy.