P1086: EFFICACY AND SAFETY OF ZANDELISIB ADMINISTERED BY INTERMITTENT DOSING IN JAPANESE PATIENTS WITH RELAPSED/REFRACTORY INDOLENT B-CELL NON-HODGKIN’S LYMPHOMA: PRIMARY ANALYSIS OF THE PHASE 2 STUDY MIRAGE

Topic: 18. Indolent and mantle-cell non-Hodgkin lymphoma - Clinical Background: Zandelisib is an oral selective PI3Kδ inhibitor. Results from a prior phase 1b trial in B-cell malignancies demonstrated that the immune-related adverse events (AEs) associated with continuous PI3Kδ inhibition were mitigated with intermittent dosing (ID) (Pagel et al. Lancet Oncol 2022). In the global phase 2 TIDAL trial, zandelisib on ID showed an objective response rate (ORR) of 72.7% and complete response (CR) rate of 38.0% assessed by independent review, and a low rate of discontinuation due to any treatment-related AEs (13.2%), at a median follow-up of 14.3 months in 121 patients with relapsed/refractory (r/r) follicular lymphoma (FL) after failure of ≥2 prior therapies (Zelenetz et al. ASH 2022). Aims: This MIRAGE trial (NCT04533581) was conducted as a multicenter, open-label, single-arm, phase 2 study to evaluate the efficacy and safety of zandelisib in Japanese patients with r/r indolent B-cell non-Hodgkin’s lymphoma (B-NHL). Methods: Eligible patients were aged ≥ 20 years with r/r indolent B-NHL of either FL grade 1-3A or marginal zone lymphoma (MZL) and had received ≥ 2 prior therapies including an anti-CD20 antibody and/or chemotherapy. Patients treated with prior PI3Kδ or BTK inhibitors were excluded. In the first two 28-day cycles, zandelisib was administered at 60mg daily, and then by ID on Days 1 to 7 of a 28-day cycle, until progressive disease or unacceptable toxicity. Prophylaxis against pneumocystis jiroveci pneumonia was mandatory. The primary endpoint was ORR according to the response criteria of the Lugano Classification assessed by independent review. Results: Between October 2020 and November 2021, 73 patients consented to be screened, and 61 patients (FL [n=57], MZL [n=4]) were treated with zandelisib. The median age was 70 years (range, 46-80) and the median number of previous therapies was 3 (range 2-9). Thirty patients (49.2%) were refractory to their last therapy, 16 (26.2%) had bulky diseases (≥ 5 cm) and 29 (47.5%) were POD24. The ORR was 75.4% (46/61, 95% CI 62.7, 85.4), with 15 patients achieving CR (24.6%, 95% CI 14.5, 37.3). A median time to response was 58 days, and 43 of 46 responders obtained their first response including 13.1% (n=8) of CRs at week 8 which increased to 21.3% (n=13) by week 24. For patients who were refractory to their last therapy, had bulky disease and were POD24, ORR was 70.0%, 62.5% and 65.5%, respectively. The most common AE was neutrophil count decreased (n=26, 42.6%). Grade ≥ 3 AEs of special interest included AST and ALT elevation (n=5, 8.2%), cutaneous reactions (n=2, 3.3%) and diarrhea, colitis, and lung infection (one patient each, 1.6%). At a median follow-up of 9.5 months, discontinuation rate due to any treatment emergent AEs was 14.8%. There were no treatment-related or COVID-19-related deaths. Summary/Conclusion: This study demonstrates a favorable efficacy and safety profiles consistent with TIDAL study in a Japanese patient population with r/r indolent B-NHL. Zandelisib is being investigated as a potential new therapeutic option for these patients. Keywords: PI3K, Marginal zone, Indolent non-Hodgkin’s lymphoma, Follicular lymphoma