Abstract
Background: Patients with relapsed or refractory (R/R) large B-cell lymphoma (LBCL) have poor prognoses, particularly those who have experienced multiple lines of treatment failure. Glofitamab, a bispecific antibody targeting CD20 and CD3, has shown promising efficacy in R/R DLBCL in the NP30179 and STARGLO studies and has been approved for its treatment. However, real-world patients often present with more complex conditions, such as poorer performance status or more comorbidities. To address this gap, we conducted this prospective, multicenter, real-world study to evaluate the treatment patterns, effectiveness, and safety of glofitamab in R/R LBCL patients in real-world settings (NCT06656234).
Methods: This study involves 20 hematology centers across China and aims to enroll 200 R/R LBCL patients. Eligible patients included those with a pathological diagnosis of CD20-positive R/R DLBCL, transformed lymphoma, high-grade B-cell lymphoma (HGBCL), primary mediastinal large B-cell lymphoma (PMBCL), or primary central nervous system lymphoma (PCNSL) who have received at least one prior line of systemic therapy. Treatment patterns, effectiveness, and safety, including cytokine release syndrome (CRS), ICANS, and infections, were analyzed.
Results: As of July 10, 2025, 74 patients were enrolled: 57 (77%) DLBCL NOS, 7 (9.5%) transformed lymphoma, 2 (2.7%) PCNSL, 4 (5.4%) PMBCL, and 4 (5.4%) HGBCL. Median age was 57.5 years (range: 19–81), and 47.3% had ECOG PS ≥2. Additionally, 81.1% were at Ann Arbor stage III–IV, 28.4% had bulky disease, and 45.9% had double-expressor lymphoma. IPI scores were 3 in 23.0% and 4–5 in 40.5%. Patients were heavily pretreated: median prior lines were 2 (range: 1–5), 71.6% had ≥2 prior lines, 5.4% had ASCT, and 21.6% had CAR-T therapy.
Among 74 patients, 19 (25.7%) received monotherapy, 36 (48.6%) combined with CAR-T, 7 (9.5%) with chemotherapy/radiotherapy, 6 (8.1%) with anti-PD-1 therapy, and 6 (8.1%) with ADC/ small molecule targeted therapies. Of 22 patients completing treatment, 3 completed 12 cycles, 15 bridged to CAR-T, and 4 bridged to ASCT. Treatment was discontinued in 21 patients (4 PD, 4 deaths, 2 CD20 antigen loss, 11 personal reasons).
Among 65 evaluable patients after 2–3 cycles, ORR was 76.9%, with CR 32.3%, PR 44.6%, SD 10.8%, and PD 12.3%. Results align with clinical trials, showing glofitamab induces early remission. ORR in 49 DLBCL NOS patients was 77.6% (CR 32.7%). For transformed lymphoma, ORR was 66.7% (PR 4/6). Among HGBCL patients, 75.0% achieved CR (3/4). For PMBCL, 25.0% achieved CR, 50.0% PR. All 4 CNS-involved patients responded (2 CR, 2 PR). ECOG PS ≥2 patients (n=35) are ineligible for NP30179. Among 31 evaluable patients, ORR was 77.4%, with CR 29.0%.
The treatment sequence of glofitamab and CAR-T is one of the key topics of clinical interest. Among 12 evaluable patients who had previously received CAR-T therapy, the ORR was 75.0%, with a CR rate of 33.3%, suggesting that glofitamab can serve as an effective salvage therapy after CAR-T failure. Additionally, among 33 evaluable patients receiving glofitamab in combination with CAR-T, following 2–3 cycles of glofitamab treatment, the ORR was 75.8%, indicating that glofitamab can be an effective bridging therapy prior to CAR-T cell treatment.
Among the 15 evaluable patients who received glofitamab monotherapy, the ORR was 60.0%, with a CR rate of 40.0%. Notably, six patients in this study received combination therapy with anti-PD-1, all of whom achieved a response, including four achieving CR. This outcome is superior to the efficacy observed in previous clinical studies. Additionally, four patients who had undergone multiple prior lines of therapy (range: 1–4) achieved remission (3 CR, 1 PR) after glofitamab treatment and subsequently underwent ASCT, suggesting that glofitamab may enable younger patients to regain eligibility for ASCT in later-line settings.
Safety was favorable, with CRS in 43.2% (≥Grade 3: 2.7%). CRS mostly occurred in cycle 1 (any grade: 35.1%; ≥Grade 3: 1.4%) and only one Grade 1 event after cycle 5. No Grade 5 CRS events were reported. ICANS occurred in 4.1% (≥Grade 3: 1.4%) with no Grade 5 events.
Conclusion: In real-world settings, glofitamab, as monotherapy or combination therapy, demonstrated rapid responses and high remission rates in high-risk, heavily pretreated R/R LBCL patients, with manageable safety. Future studies will expand sample size and follow-up to confirm these findings.
