Paper Details

ABSTRACT

Background: In the Phase III SUNMO study (NCT05171647), mosunetuzumab plus polatuzumab vedotin (Mosun-Pola) demonstrated superior efficacy vs rituximab, gemcitabine and oxaliplatin (R-GemOx), with a manageable safety profile and improved health-related quality of life in patients (pts) with relapsed/refractory (R/R) large B-cell lymphoma (LBCL; Budde, et al. J Clin Oncol 2025). Establishing the potential for circulating tumor DNA (ctDNA) clearance by Mosun-Pola, and evaluating efficacy in subgroups of interest by baseline tumor burden and lymphoma microenvironment, remain important research topics. Here, we report updated data from SUNMO after 3 years of follow-up, alongside a comprehensive biomarker assessment of ctDNA dynamics and the prognostic impact of baseline ctDNA, total metabolic tumor volume (TMTV), and RNA-seq-based lymphoma microenvironment archetype profiles (LymphoMAP).

Methods: Transplant-ineligible pts with R/R LBCL were randomized 2:1 to Mosun (subcutaneous)-Pola or R-GemOx. Dual primary endpoints were IRC-assessed progression-free survival (PFS) and objective response rate (ORR); secondary endpoints included complete response (CR) rate, duration of CR (DOCR), and safety. Overall survival analysis remains preliminary due to insufficient accumulated events. ctDNA was measured at baseline and longitudinally using the AVENIO Oncology Assay (research use only) and a customized linked somatic variant method. Baseline TMTV and LymphoMAP archetypes were assessed as previously described (Jemaa, et al. J Digit Imaging 2020; Li, et al. Cancer Cell 2025).

Results: At the data cut-off (May 8, 2026), 208 pts with R/R LBCL were enrolled (Mosun-Pola, n=138; R-GemOx, n=70). Overall, 44% of pts had 1 and 56% had ≥2 prior lines of therapy, 19% had high-grade B-cell lymphoma, 58% had primary refractory disease, and 70% were refractory to their last therapy. With a median follow-up of 37.1 months (mos), Mosun-Pola continued to demonstrate superior efficacy over R-GemOx: median PFS, 11.6 vs 3.8 mos (HR: 0.41; 95% CI: 0.28–0.60); ORR, 70% vs 40%; CR rate, 51% vs 24%, respectively. The 24-month PFS rate with Mosun-Pola was 38.6% (95% CI: 29.8–47.5) and median DOCR was 28.7 mos (95% CI: 28.7–not evaluable); 64.1% of pts with a CR remained in remission at 24 mos after initial CR.

Mosun-Pola led to a higher proportion of pts with undetectable ctDNA at Cycle (C)4 Day (D)1 versus R-GemOx, 55.6% (35/63) vs 12.5% (3/24), respectively. Achieving undetectable ctDNA at C4D1 was associated with prolonged PFS; pts with undetectable ctDNA at C4 had a 24-month PFS estimate of 65.2% (95% CI: 51.0–83.3). At C8D1, undetectable ctDNA was achieved in 91.2% (31/34) vs 28.6% (2/7) of pts assessed, for Mosun-Pola and R-GemOx, respectively.

Exploratory analyses correlated clinical efficacy with baseline biomarkers. High baseline ctDNA (≥141 mutant molecule per mL, median) was associated with inferior PFS vs lower levels (HR: 1.9; 95% CI: 1.16–3.10). Mosun-Pola maintained a PFS benefit over R-GemOx across both low- and high-ctDNA subgroups, with HRs of 0.34 (95% CI: 0.15–0.74) and 0.51 (95% CI: 0.26?1.01), respectively. Among 203 TMTV-evaluable pts, the median baseline TMTV was 173cm³; high baseline TMTV was prognostic for inferior PFS. Mosun-Pola showed improved PFS over R-GemOx in both high-TMTV (≥173cm³; HR: 0.59; 95% CI: 0.35?0.99), and low-TMTV (<173cm³; HR: 0.28; 95% CI: 0.16?0.50) subgroups. Overall, 168 pts were evaluable for LymphoMAP archetype classification, with 38.1% classified as fibroblast/macrophage-enriched (FMAC), 34.5% as lymph node-like (LN), and 27.4% as T-cell exhausted (TEX). PFS favored Mosun-Pola over R-GemOx across all LymphoMAP subtypes, with HRs of 0.75 (95% CI: 0.40?1.43) for FMAC, 0.34 (95% CI: 0.17–0.71) for LN, and 0.53 (95% CI: 0.19–1.47) for TEX. ORRs with Mosun-Pola were 60.5% for FMAC, 69.0% for LN, and 85.3% for TEX.

No new safety signals were identified with extended follow-up. B cells recovered by 9–12 mos after the end of Mosun-Pola treatment while IgG remained stable throughout treatment.

Conclusions: With 3 years of follow-up, Mosun-Pola continues to deliver sustained clinical benefit and a consistent safety profile in transplant-ineligible pts with R/R LBCL. The consistent PFS benefit with Mosun-Pola across all baseline tumor burden and microenvironment subgroups reinforces its use as a highly effective and versatile therapeutic option, achieving deep, durable responses and high rates of undetectable ctDNA.