Polatuzumab Vedotin Plus Bendamustine and Rituximab in Relapsed/Refractory Diffuse Large B-Cell Lymphoma (R/R DLBCL): Final Results of a Phase Ib/II Randomized Study and Single-Arm Extension (Ext) Study

Background: Polatuzumab vedotin (Pola) is a CD79b-targeting antibody–drug conjugate approved for the treatment of R/R DLBCL based on findings from the GO29365 randomized Phase lb/ll study (NCT02257567), which assessed the efficacy of Pola plus bendamustine and rituximab (Pola+BR) vs BR alone in patients (pts) with R/R DLBCL. The primary endpoint was met, with an independent review committee (IRC)-assessed complete response (CR) rate of 40.0% with Pola+BR vs 17.5% with BR (Sehn LH, et al. J Clin Oncol 2020). A single-arm Ext cohort was later enrolled to receive Pola+BR. Efficacy was consistent with the randomized phase, with a CR rate of 38.7% (Sehn LH, et al. Blood Adv 2022). We report final results from the GO29365 study with extended follow-up of up to 5 years, including updated results from the Phase lb safety run-in, Phase ll randomized, and Ext cohorts.

Methods: This Phase lb/ll, open-label, multicenter, randomized study included pts with R/R DLBCL, who were ≥18 years old, were stem cell transplant ineligible, had an Eastern Cooperative Oncology Group performance status of 0–2 and had received ≥1 prior line of therapy. Pts with Grade (Gr) >1 peripheral neuropathy (PN) were excluded. Pts received Pola 1.8mg/kg with each cycle of BR (full methods have been described; Sehn LH, et al. J Clin Oncol 2020). The primary endpoint was IRC-assessed CR at primary response assessment (PRA) measured by positron emission tomography/computed tomography. Secondary endpoints included safety, best objective response (BoR), best complete response (BCR), duration of response (DoR) and progression-free survival (PFS) assessed by IRC, and overall survival (OS).

Results: At final data cut-off (October 21, 2021), median duration of follow-up for pts treated with Pola+BR was 77.4 months (mo), 59.9 mo, and 29.2 mo, in the safety run-in (N=6), randomized (N=40), and Ext (N=106) cohorts, respectively. Baseline characteristics were similar between cohorts (Sehn LH, et al. Blood Adv 2022). In the randomized cohort (Pola+BR vs BR), final CR rate at PRA was 42.5% (n=17; 95% confidence interval [CI]: 27.0–59.1) vs 17.5% (n=7; 95% CI: 7.3–32.8), BoR rate was 62.5% (n=25; 95% CI: 45.8–77.3) vs 25.0% (n=10; 95% CI: 12.7–41.2), and BCR rate was 52.5% (n=21; 95% CI: 36.1–68.5) vs 22.5% (n=9; 95% CI: 10.8–38.5). Median (95% CI) DoR in the randomized cohort (Pola+BR vs BR) was 10.9 (5.7–40.7) vs 10.6 (4.0–19.7) mo (hazard ratio [HR], 0.6; 95% CI: 0.3–1.4), and PFS was 9.2 (6.0–13.9) vs 3.7 (2.1–4.5) mo (HR, 0.4; 95% CI: 0.2–0.7). The risk of responders progressing or dying was reduced by 40.0% (HR, 0.6; 95% CI: 0.3–1.4) with Pola+BR vs BR alone. Median OS (95% CI) in the randomized cohort was 12.4 (9.0–32.0) vs 4.5 (3.7–6.0) mo (HR, 0.4; 95% CI: 0.2–0.7) with Pola+BR vs BR alone. In the Ext cohort, CR rate at PRA for Pola+BR was 39.6% (n=42; 95% CI: 30.3–49.6), consistent with the randomized cohort. BoR rate was 57.5% (n=61; 95% CI: 47.6–67.1), and BCR rate was 53.8% (n=57; 95% CI: 43.8–63.5). Median DoR, PFS, and OS were 13.4 (95% CI: 8.6–20.0) mo, 7.0 (95% CI: 5.1–9.8) mo, and 12.3 (95% CI: 8.3–17.0) mo, respectively.

No new safety signals were reported; results were consistent with previous analyses. In pts receiving Pola+BR in the safety run-in, randomized, and Ext cohorts, respectively, Gr 3–4 adverse events (AEs) occurred in 5/6 (83.3%), 34/39 (87.2%), and 83/106 (78.3%) pts, and serious AEs occurred in 4/6 (66.7%), 27/39 (69.2%), and 57/106 (53.8%) pts, respectively. The most common Gr 3–4 AEs were neutropenia and thrombocytopenia in both the randomized and Ext cohorts, and for serious AEs were infections, pyrexia, and febrile neutropenia in the randomized cohort, and infections in the Ext cohort. In the randomized and Ext cohorts, respectively, any-grade PN occurred in 17 (43.6%) and 29 (27.4%) pts, and Gr ≥2 PN occurred in 6 (15.4%) and 16 (15.1%) pts. Median time to PN resolution was 0.3 and 2.9 mo in the randomized and Ext cohorts, respectively.

Conclusions: In this planned final analysis, with extended follow up of 5 years in the randomized cohort and 2 years in the Ext cohort, the improved efficacy of Pola+BR persisted. Median DoR was longer with increased follow-up, indicating Pola+BR leads to a durable benefit in responding pts. Median PFS and OS were longer with Pola+BR compared with BR, consistent with previous results. No new safety signals were identified. These findings provide further evidence of the effectiveness of Pola+BR for the treatment of R/R DLBCL.