Abstract
Outcomes of relapsed/refractory (R/R) diffuse large B cell lymphoma（DLBCL）are poor. Drug resistance and patient unfitness are key factors limiting therapeutic responses in R/R DLBCL. Although CAR-T cell therapy has improved the efficacy in R/R DLBCL, treatment related adverse events limit its application in unfit patients. Bispecific antibodies like glofitamab showed better tolerance but lower CR rate compared with CAR-T cell therapy. Therefore, combinatory strategies to improve the CR rate of glofitamab are warranted. As reported, immunosuppressive tumor microenvironment (TME) and T-cell dysfunction are associated with poor response to glofitamab (S. Tumuluru, Blood 2025). Since previous studies suggest that radiotherapy and Granulocyte-macrophage colony-stimulating factors (GM-CSF) can heat up the TME and boost anti-tumor immune response (Y. Kong, Front Immunol 2022), while lenalidomide can modulate the TME and increase T cell function (H. Guo, Front Immunol 2022), we designed a chemo-free TME-reshaping regimen with radiotherapy, GM-CSF and lenalidomide to enhance the efficacy of Glofitamab in R/R DLBCL and minimize the treatment related adverse events.Methods: This is a single center, open label, investigator-initiated phase 2 trial planning to enroll 22 patients (NCT06651853). Key inclusion criteria include DLBCL patients with early relapsed disease (< 12 months), or late relapsed disease but are transplant-ineligible after 1 line of therapy, or relapsed disease after 2+ lines of therapy. Patients with central nervous system involvement could be included. Glofitamab is administered on 21-day cycles for a total of 6 cycles. In Cycle (C) 1, large fraction radiotherapy is administered at 5Gy/day for 3 days, GM-CSF is administered at 400µg per day for 3 days (day 1-3) and Lenalidomide is administered at 25mg per day for 14 days (day 1-14) starting on the first day after the end of radiotherapy. Pretreatment with obinutuzumab (1000 mg) is administered intravenously on the first day after the end of radiotherapy (day 1). Glofitamab is then administered intravenously as step-up doses on day 8 (2.5 mg) and day 15 (10 mg) of cycle 1. For cycles 2-6, GM-CSF is administered for day 1-3, Glofitamab is administered at 30mg on day 4. Lenalidomide is administered at 25mg/day on day 1-14. Treatment consists of 6 cycles or until disease progression, death, intolerable toxicity, withdrawal of informed consent. All patients receive assessments by FDG PET-CT after C3 and at the end of treatment (EOT). The primary endpoint is CR rate by Lugano 2014 criteria. In every cycle, peripheral blood is collected the day before lenalidomide and GM-CSF received and the day of glofitamab. T cell phenotype and function are detected by flowcytometry.Results: At data cutoff (July 22, 2025), 11 patients had been enrolled. Median age of all pts was 70.0 (range: 46-88) years, 45% were male, 54% had Ann Arbor stage III/IV disease, and 91% had an ECOG score ≥2 at study entry. 7 pts were with early relapsed disease; 2 pts were with primary central nervous system lymphoma. The median follow-up was 3.0 months (range: 0.5-9.0), and 9 patients underwent at least one post-baseline tumor assessment, with 100% achieving CR after C3 and remaining in CR by the data cut-off. All patients were evaluated for safety. Five pts (45%) experienced CRS events, and 1 pt experienced a Gr 3 CRS event. One pt experienced Gr 2 ICANS. All CRS and ICANS events are resolved by data cut-off. GM-CSF related infusion-related reaction occurred in 3 pts, and lenalidomide related rash occurred in 2 pts. One patient experienced Gr 2 thrombocytopenia. One patient experienced COVID-19. Due to the lack of co-stimulatory signals such as CD28 would result in T-cell anergy during glofitamab treatment, we further explored whether the pretreatment by lenalidomide and GM-CSF could bypass CD28 co-stimulation to enhance T cell function. Our results demonstrated that the Ki-67 expression of T cells was significantly upregulated after lenalidomide and GM-CSF treatment. Notably, the upregulation of Ki-67 was also observed in CD28-negative T cells. Additionally, GM-CSF could inhibit polarization of macrophages to M2 phenotype in vitro.Conclusions: Chemo-free Tumor Microenvironment (TME) reshaping regimen could enhance T cell function and inhibit M2 polarization, subsequently improving the response of glofitamab and inducing high CR rates with good tolerance in pts with R/R DLBCL.
