Short answer: STARGLO was an open-label phase 3 trial of 274 transplant-ineligible patients with relapsed or refractory diffuse large B-cell lymphoma (DLBCL). Glofitamab plus gemcitabine and oxaliplatin (Glofit-GemOx) improved overall survival over rituximab plus GemOx (R-GemOx): at the primary analysis, HR 0.59, and at the updated analysis, median OS 25.5 vs 12.9 months (HR 0.62). The complete response rate was 58.5% vs 25.3%. The FDA nonetheless declined to approve the combination in July 2025, citing doubts about whether the results applied to the U.S. population, and the regional subgroup results are the main teaching point.
Author and disclosure: Roupen Odabashian, MD, FRCPC, Oncologist and founder of MeDucation AI. I am the founder of MeDucation AI, which sells hematology and oncology board preparation. I have no other financial relationships relevant to the studies discussed here. This article is educational and is not a substitute for individualized clinical judgment or tumor board discussion.
What was the STARGLO trial design?
STARGLO randomized 274 patients 2:1 between February 2021 and March 2023 to Glofit-GemOx (n = 183) or R-GemOx (n = 91). Eligible patients had relapsed or refractory DLBCL after at least one prior line of therapy and were not candidates for autologous stem cell transplant. Patients with only one prior line had to be transplant-ineligible for reasons such as age, organ dysfunction, performance status or comorbidity. The median age was 68. Glofitamab, a CD20 by CD3 bispecific antibody, was given after obinutuzumab pretreatment to reduce cytokine release syndrome, and was combined with GemOx for eight cycles followed by glofitamab alone for up to four more cycles. The primary endpoint was overall survival. The trial was sponsored by Roche and published in The Lancet by Abramson and colleagues in 2024. Sites were in 13 countries across Asia, Australia, Europe and North America.
What were the survival and response results?
Outcome | Glofit-GemOx (n = 183) | R-GemOx (n = 91) |
|---|---|---|
Median OS, primary analysis (median follow-up 11.3 months) | Not estimable | 9.0 months (HR 0.59, 95% CI 0.40 to 0.89, P = 0.011) |
Median OS, updated analysis (median follow-up 20.7 months) | 25.5 months | 12.9 months (HR 0.62, 95% CI 0.43 to 0.88) |
Complete response rate | 58.5% | 25.3% |
Median progression-free survival | 13.8 months | 3.6 months (HR 0.40) |
The magnitude of the PFS and response differences is large, and the survival advantage persisted with longer follow-up. A later analysis at a median follow-up of 24.7 months reported a PFS hazard ratio of 0.41 and the same complete response rates.
What toxicities did STARGLO report?
Safety measure | Glofit-GemOx | R-GemOx |
|---|---|---|
Cytokine release syndrome (any grade) | 44% (grade 3 in 2%) | Not applicable |
ICANS (any grade) | 2% (grade 3 in 1%) | Not applicable |
Grade 3 or higher adverse events | 78% | 41% |
Serious adverse events | 54% | 17% |
Discontinuation of any study drug for adverse events | 27% | 13% |
Adverse event related deaths | 8% | 5% |
Cytokine release syndrome was mostly low grade and concentrated around the first doses. Patients in the glofitamab arm also received many more cycles of treatment (median 11 vs 4 in the interim report), so part of the higher toxicity burden reflects longer exposure. Fatal adverse events in the interim report were largely from COVID-19.
Why did the FDA decline to approve Glofit-GemOx?
The regional subgroup analysis is the core issue. In an updated analysis, the OS hazard ratio favored glofitamab in the rest of the world (HR 0.41, 95% CI 0.27 to 0.64) but not clearly in Europe (HR 1.09, 95% CI 0.54 to 2.18) or North America (HR 2.62, 95% CI 0.56 to 12.34). The North American subgroup was small, with 15 patients in the glofitamab arm and 10 in the control arm, so the confidence intervals are wide and the estimate is unstable. Even so, the FDA Oncologic Drugs Advisory Committee voted 8 to 1 in May 2025 against the applicability of the results to the U.S. population, and the agency issued a complete response letter on July 18, 2025. Glofitamab retained its existing accelerated approval for relapsed or refractory DLBCL after at least two lines of systemic therapy. The sponsor said it was discussing the phase 3 SKYGLO study with the FDA as a future confirmatory study. Regulatory status changes, so check current labeling before using this in practice.
What should you remember for the boards?
Glofitamab is a bispecific antibody that engages CD20 on the lymphoma cell and CD3 on T cells, and its signature toxicity is cytokine release syndrome, which is mitigated with obinutuzumab pretreatment and step up dosing.
STARGLO tested a bispecific plus chemotherapy in transplant-ineligible patients, which is a different question from the use of single-agent glofitamab after two or more lines.
A positive overall result with heterogeneous regional subgroups is a classic reason regulators ask whether a trial population reflects local practice, such as differences in prior rituximab exposure and in the availability of later-line therapies.
Subgroup hazard ratios from small samples, such as the North American cohort here, carry very wide confidence intervals and should not be over-interpreted in either direction.
To practice this kind of question, see Oncology Board Review Question Banks Compared or the broader plan in How to Pass the ABIM Medical Oncology Boards. For the full exam approach, read Oncology Board Review 2026.
Frequently asked questions
What did the STARGLO trial show?
Glofitamab plus GemOx improved overall survival, progression-free survival and complete response rate compared with rituximab plus GemOx in transplant-ineligible relapsed or refractory DLBCL, at the cost of more serious adverse events.
Is glofitamab plus GemOx FDA approved?
No. The FDA issued a complete response letter for the combination in July 2025. Glofitamab remains approved under accelerated approval for relapsed or refractory DLBCL after at least two lines of systemic therapy.
Why was the STARGLO result questioned?
Regional subgroups were not consistent, with a clear survival benefit in the rest of the world but not in Europe or North America, and the advisory committee questioned whether the results applied to U.S. patients.
How common was cytokine release syndrome in STARGLO?
Cytokine release syndrome occurred in 44% of patients on glofitamab, with grade 3 events in 2%.
What is the difference between STARGLO and the single-agent glofitamab approval?
STARGLO combined glofitamab with GemOx in a randomized phase 3 trial after at least one prior line, whereas the existing approval is for single-agent glofitamab after at least two prior lines.
References
The ASCO Post. Glofitamab combination therapy shows survival benefit in relapsed or refractory diffuse large B-cell lymphoma. August 2024. ascopost.com
The ASCO Post. Glofitamab or rituximab plus gemcitabine and oxaliplatin in relapsed or refractory DLBCL (Abramson et al, The Lancet). November 2024. ascopost.com
OncLive. Updated STARGLO data show maintained OS benefit with glofitamab plus chemo in R/R DLBCL. onclive.com
OncLive. FDA issues CRL for glofitamab plus chemo in R/R DLBCL. July 2025. onclive.com
Genentech. Update on supplemental biologics license application for Columvi combination in relapsed or refractory DLBCL. July 18, 2025. gene.com
Roche. Investor update, July 18, 2025. roche.com
ClinicalTrials.gov. STARGLO, NCT04408638. clinicaltrials.gov
