David Belada, Gregor Verhoef, Xuejun Chen, Claudia Corrado, Wanying Zhao, Morton Coleman, René-Olivier Casasnovas, Javier Munoz, Keith Fay, Hui-Peng Lee, Remy Gressin, Douglas J DeMarini, and Amitkumar Mehta
e19038 Background: Parsaclisib, a potent, highly selective, next-generation PI3Kδ inhibitor, showed preliminary efficacy as monotherapy for relapsed or refractory non-Hodgkin lymphoma, including DLBCL (Abstract 410, ASH 2017), in a phase 1/2 study. This phase 2 study further assessed parsaclisib in patients (pts) with relapsed or refractory DLBCL (NCT02998476). Methods: Pts enrolled into 2 groups (A, Bruton tyrosine kinase [BTK] inhibitor naïve; B, BTK inhibitor experienced) and received oral parsaclisib 20 mg QD for 8 wks, then 20 mg QW. In a planned interim futility analysis conducted in the first 40 pts treated in Group A, if ≤13 (≤32.5%) responded by IRC assessment, Group A was to be terminated. Results: At data cutoff (22 Jun 2018), 60 pts (Group A, n = 55; Group B, n = 5) were treated (median age, 71 y [range, 36—94]; men, 63.3%; ≥3 prior systemic therapies, 60%). At the planned interim analysis in Group A, ORR (by PET) was 25% (10/40 pts; 5 CMR, 5 PMR); the futility boundary was crossed. At data cutoff, ORR in Group A was 25.5% (14/55 pts; 8 CMR, 6 PMR); median PFS was 2.2 mo (95% CI: 2.0‒4.1); median DOR was 4.5 mo (95% CI: 2.1‒5.1). ORs were observed in germinal center B-cell (GCB) and non-GCB subtypes. ORR in Group B was 20% (1/5 pts; 1 CMR). The most common non-hematologic treatment-emergent adverse events (TEAEs) occurring in > 10% of all pts (all grade [Gr]; Gr 3/4) were rash events (21.7%; 1.7%), colitis/diarrhea events (16.7%; 5%), nausea (16.7%; 0%), cough (15%; 0%), and pyrexia (15%; 8.3%). Gr 3/4 AST and ALT elevations occurred in 5% and 1.7% of pts, respectively; Gr 3/4 neutropenia and anemia occurred in 5% of pts each. The most frequent ( > 5%) serious TEAEs were pyrexia (8.3%), general physical health deterioration (6.7%), and hypercalcemia (6.7%). TEAEs led to therapy discontinuation in 7 pts (2 treatment-related), dose interruption in 20 pts (10 treatment-related), and dose reduction in 3 pts (all treatment-related). Median duration of therapy was 57.5 d (range, 11–318). Conclusions: Parsaclisib monotherapy using a QD followed by QW dosing regimen was well tolerated with no new safety signals reported. Further evaluation of parsaclisib in all subtypes of DLBCL is ongoing in a combination study Clinical trial information: NCT02998476.