Long-term follow-up with sustained progression-free survival (PFS) benefit after subcutaneous (SC) mosunetuzumab in combination with polatuzumab vedotin compared with rituximab plus polatuzumab vedotin in patients with relapsed or refractory (R/R) B-

Abstract Background: A prior analysis of mosunetuzumab (Mosun) SC, a CD20xCD3 T-cell-engaging bispecific antibody, with polatuzumab vedotin (Pola; M-Pola), an anti-CD79b monoclonal antibody, in a Phase Ib/II study (NCT03671018) showed promising durability of response and manageable safety versus (vs) rituximab (R)-Pola in patients with R/R transplant-ineligible large B-cell lymphoma (LBCL; Chavez et al. ASH 2024). Here we report an updated analysis of ≥2 years' follow-up in the randomized cohort of the Phase Ib/II study. Methods: Patients with histologically confirmed R/R LBCL and ≥1 previous line of treatment, including anti-CD20 therapy, were randomized 1:1 to receive M-Pola or R-Pola. Patients previously treated with Pola were not included. Patients in the R-Pola arm with PD during or at end of treatment (EOT) or stable disease at EOT could cross over to receive M-Pola (up to 8 cycles of cumulative Pola). Primary endpoint was best independent review committee-assessed objective response rate (ORR) per Lugano 2014 criteria. Cytokine release syndrome (CRS) events were defined per ASTCT criteria. All p-values were descriptive. Results: As of November 15, 2024, 80 patients were enrolled (M-Pola, n=40; R-Pola, n=40) of which 54 patients (67.5%) were enrolled in the United States (US). Baseline characteristics were as previously reported, with 75% diffuse large B-cell lymphoma, 10% Grade 3b follicular lymphoma, 20% high-grade B-cell lymphoma, and 17.5% transformed follicular lymphoma (Chavez et al. ASH 2024). Median follow-up in the M-Pola vs R-Pola arms was 25.7 (range 1–35) months vs 27.2 (range 0‒34) months, respectively. For M-Pola vs R-Pola, best ORR and complete response (CR) rate were 78% (95% CI: 61.6–89.2) and 55% (95% CI: 38.5–70.7) vs 50% (95% CI: 33.8–66.2) and 35% (95% CI: 20.6–51.7), respectively. Median PFS was 25.4 months (95% CI: 9.2– not evaluable [NE]) for M-Pola and 6.4 months (95% CI: 4.7–18.6) for R-Pola (HR 0.47 [95% CI: 0.2–0.9] p=0.0287]). Median overall survival (OS) was not reached (NR) for M-Pola (95% CI: 17.6–NE) and 25.5 months (95% CI: 16.2–NE) for R-Pola (HR 0.78 [95% CI: 0.4‒1.5] p=0.4749). Median duration of response (DOR) and duration of CR were NR in the M-Pola arm (95% CI: 15.0–NE and 95% CI: 16.8–NE, respectively) and were 11.3 months (95% CI: 5.8–NE) and 12.7 months (95% CI: 3.6–NE) in the R-Pola arm, respectively. With a median follow-up of 18.7 (95% CI: 15.2–NE) months for 20 crossover patients, ORR was 50% (95% CI: 27.0–72.8) and the median DOR was 15.7 months (95% CI: 2.8–NE). In the safety population (n=79), no new safety signals were identified. The most common adverse events (AEs) for M-Pola vs R-Pola were diarrhea (48% vs 33%), fatigue (35% vs 31%), and nausea (25% vs 36%), respectively. Grade 3/4 AEs occurred in 58% of M-Pola vs 54% R-Pola patients; serious AEs (SAE) occurred in 33% of M-Pola vs 26% R-Pola patients. Grade 5 (fatal) AEs occurred in 5% of patients in the M-Pola arm (COVID-19 [n=1], COVID-19 pneumonia [n=1]) and 3% in the R-Pola arm (hepatic failure [n=1]). AEs leading to discontinuation of study treatment occurred in 7.5% of M-Pola patients (COVID-19 pneumonia [n=1], peripheral motor neuropathy [n=1], peripheral sensory neuropathy [n=1]) and in 7.7% of R-Pola patients (peripheral neuropathy [n=1], neutrophil count decreased [n=1], pain in extremity [n=1]). CRS events were reported in 13% (5/40) of patients in the M-Pola arm: all were low grade (Grade 1: 10%; Grade 2: 3%) and occurred during Cycle 1 Days 1–7, with median time to CRS onset of 2 days (range 1–3) and median duration of 2 days (range 1–5), and all CRS events were resolved. Other AEs of interest included serious infections, which occurred in 23% and 18% of patients in M-Pola and R-Pola arms, respectively. There were no investigator-reported events of immune effector-cell associated neurotoxicity syndrome. Neutropenia occurred in 40.0% of M-Pola patients (Grade 2: 7.5%; Grade 3/4: 32.5%); 1 patient experienced a Grade 3 event of febrile neutropenia. Conclusions: Fixed-duration outpatient M-Pola continues to demonstrate clinically meaningful and sustained improvements across all efficacy endpoints compared with R-Pola. These extended follow-up results reveal no new safety signals, reinforcing the treatment's favorable profile. The findings provide strong additional evidence for the long-term efficacy of the M-Pola regimen in the US patient population, supporting broad use in the second-line plus setting.