Paper Details

ABSTRACT

Background: In the Phase III STARGLO trial (NCT04408638), glofitamab plus GemOx (Glofit-GemOx) showed significantly improved overall survival (OS) and progression-free survival (PFS) in transplant-ineligible patients with relapsed/refractory (R/R) diffuse large B-cell lymphoma (DLBCL) compared with rituximab plus GemOx (R-GemOx; Abramson, et al. Lancet 2024). We report 4-year results from the STARGLO trial and analyze clinical outcomes across LymphoMAP archetypes.

Methods: Patients were randomized 2:1 to Glofit-GemOx (8 cycles plus 4 cycles glofitamab monotherapy) or R-GemOx (8 cycles) and stratified by prior line of therapy (LOT; 1 vs ≥2) and refractoriness to last therapy. Primary endpoint was OS. Other endpoints included investigator-assessed PFS, complete response (CR) rate and duration of CR (DOCR). Scans were performed at screening, during treatment (Cycle [C]4, C8), at end of treatment (EOT), every 6 months for 2 years after EOT (protocol-specified), and thereafter as per standard of care. Landmark analyses in patients with a CR at the first tumor assessment (C4) were conducted. LymphoMAP subtypes were determined from global gene expression patterns using the LymphoMapR classifier.

Results: A total of 274 patients were enrolled (Glofit-GemOx, n=183; R-GemOx, n=91). At baseline, 172 (62.8%) had 1 prior LOT, 153 (55.8%) had primary refractory disease and 166 (60.6%) were refractory to their last therapy.

After a median follow-up of 47.4 months (clinical cut-off date: June 1, 2026), Glofit-GemOx demonstrated a sustained survival advantage over R-GemOx: median OS was 25.0 vs 12.5 months (hazard ratio [HR] 0.63, 95% confidence interval [CI]: 0.5–0.9); 48-month OS rate was 45.5% vs 27.1%. Median PFS was 15.8 vs 2.7 months (HR 0.35, 95% CI: 0.3–0.5), and CR rate was 57.9% vs 23.1% for Glofit-GemOx and R-Gemox, respectively. Median DOCR was not reached (95% CI: 28.9–not estimable [NE]) with Glofit-GemOx vs 18.9 months (95% CI: 6.6–NE) with R-GemOx; 36-month DOCR was 60.2% vs 48.8%, respectively. In the landmark analysis of patients achieving a CR at the first scheduled tumor assessment (C4; n=81), median OS was not reached (95% CI: 29.9–NE); median PFS was also not reached (95% CI: 27.2–NE).

In patients with 1 prior LOT, median OS was 50.1 months with Glofit-GemOx vs 14.4 months with R-GemOx, (HR 0.61, 95% CI: 0.4–1.0); 48-month OS estimates were 52.1% and 29.9%, respectively. Median PFS was 22.8 vs 3.3 months (HR 0.40, 95% CI: 0.3–0.6), and CR rate was 63.5% vs 24.6% in the Glofit-GemOx vs R-GemOx arms, respectively. Among patients with primary refractory disease or early relapse, median OS was 19.2 months (95% CI: 9.9–NE) with Glofit-GemOx vs 12.5 months (95% CI: 6.9–16.5) with R-GemOx; median PFS was 9.6 months (95% CI: 6.1–27.4) vs 2.5 months (95% CI: 1.6–3.3) with R-GemOx.

The safety profile was unchanged with extended follow-up. No new signals were identified, and no evidence of cumulative toxicity or new long-term adverse events were observed.

The prevalence of LymphoMAP archetypes in patients with available samples in the Glofit-GemOx (n=97) and R-GemOx (n=47) arms was 43.3% and 31.9% for high fibroblasts and macrophages (FMAC), 28.9% and 42.6% for high naïve and memory T cells (LN), and 27.8% and 25.5% for exhausted CD8+ T cells (TEX), respectively. Improved CR (33.3% vs 6.7% for FMAC, 85.7% vs 15.0% for LN, and 70.4% vs 50.0% for TEX) favoring Glofit-GemOx over R-GemOx were observed across all subtypes. PFS outcomes favored Glofit-GemOx vs R-GemOx: HR 0.40 (95% CI: 0.2–0.8) for FMAC, 0.21 (95% CI: 0.1–0.5) for LN, and 0.46 (95% CI: 0.2–1.1) for TEX. In the Glofit-GemOx arm, 24-month PFS was 19.8% for FMAC, 62.9 % for LN, and 43.7% for TEX.

Conclusion: This report represents the longest follow-up from a randomized controlled trial with a CD20xCD3 bispecific antibody in R/R DLBCL. Glofit-GemOx delivered durable survival and sustained remissions over R-GemOx in transplant-ineligible R/R DLBCL. These data highlight the durable long-term remissions in patients with one prior LOT and early relapsing/progressing disease. LymphoMAP profiling showed that Glofit-GemOx improved outcomes over R-GemOx across all three microenvironment archetypes and suggests that Glofit-GemOx drives superior clinical benefit in T-cell-enriched and T-cell-exhausted (LN/TEX) compared with the T-cell-depleted, stromal-rich (FMAC) archetype. Overall, these long-term data support the curative potential of this off-the-shelf therapy.