P1174: pi3k inhibitor linperlisib combined with gemcitabine and oxalipatin (gemox) in relapsed and/or refractory dlbcl: analysis based on tumor biology

Topic: 19. Aggressive Non-Hodgkin lymphoma - Clinical Background: In YY-20394-008 study, PI3K inhibitor linperlisib combined with GEMOX (gemcitabine and oxalipatin) regimen led to a promising efficacy in heavily treated relapsed and/or refractory(R/R) diffuse large B cell lymphoma (DLBCL). However, the differences of outcomes stratified by cell origin subtypes are less well-defined. Aims: This post-hoc analysis was based on data from the phase Ib/II YY-20394-008 study (NCT04500561) and the association between different subtypes and clinical response to combined treatment with linperlisib plus GEMOX was evaluated. Methods: This was a phase Ib/II, single-arm, open-label study of linperlisib combined with GEMOX in patients with R/R DLBCL (NCT04500561). Patients were given one cycle (21 days) of linperlisib 80mg QD orally combined with 6 cycles of GEMOX (gemcitabine 1000 mg/m2, IV on day 1, Q3W, and oxaliplatin 100 mg/m2 IV on day2, Q3W), followed by linperlisib alone until disease progression, intolerable toxicity, or withdrawal. The endpoints for the assessment of clinical activity include objective response rate (ORR), disease control rate (DCR), progression-free survival (PFS), and overall survival (OS). Results: As of October 11, 2021, there were 39 R/R DLBCL patients enrolled at 6 clinical sites in China. Among them, 9 patients were germinal center B-cell-like (GCB) subtype while 25 were non-GCB subtype, and 5 patients were not validly assessed due to lack of adequate patient material and were deemed missing. The median age was 58, 28.2% of patients were older than 65 years, and 64.1% were male. The median number of prior therapies was two. There was no difference in baseline characteristics between these groups. The ORR was 66.7% (6/9) patients for GCB subtype and 40% (10/25) patients for non-GCB subtype (Table 1). The complete response (CR) rate was 20.0% for non-GCB subtype, while no patient of GCB subtype achieved CR. The DCR were: GCB subtype (77.8%), non-GCB (52.0%). Median PFS was 3.0 months and 5.3 months for patients with GCB and non-GCB subtypes, respectively (P=0.31). The median OS was not reached in both subtypes. Table 1. Tumor response stratified by cell origin - GCB (n=9) Non-GCB (n=25) Total (n=39) Best overall response, n (%) Complete response 0 (0.0) 5 (20.0) 6 (15.4) Partial response 6 (66.7) 5 (20.0) 15 (38.5) Stable disease 1 (11.1) 3(12.0) 4 (10.3) ORR, n (%) 6 (66.7) 10(40.0) 21 (53.8) (95%CI) (29.9-92.5) (21.1-61.3) (37.2-69.9) DCR, n (%) 7 (77.8) 13 (52.0) 25 (64.1) (95% CI) (40.0-97.2) (31.3-72.2) (47.2-78.8) Median PFS, month (95%CI) 3.0 (1.3-NE) 5.3 (1.2-6.7) 5.3 (1.8-6.9) Median OS, month (95%CI) NE (3.4-NE) NE (6.7-NE) NE (NE-NE) NE, not evaluable. Summary/Conclusion: The data from YY-20394-008 study suggest that there may be a trend towards a greater PFS benefit for non-GCB subtype compared to GCB subtype. The non-GCB subtype also showed a numerically higher percentage of complete responses compared to the GCB subtype, but neither was statistically significant different. The limited number of patients limits the ability to determine a meaningful result currently. Keywords: Oxaliplatin, PI3K, Gemcitabine, Diffuse large B cell lymphoma