Using Polatuzumab Plus Chemotherapy for Patients with Relapsed/Refractory Diffuse Large B-Cell Lymphoma: A Real-World Study from the Grupo De Estudio De Linfoproliferativos De Latinoamerica (GELL)

Background. Diffuse large B-cell lymphoma (DLBCL) is the most common subtype of non-Hodgkin lymphoma. Although DLBCL is curable in earlier stages, outcomes in relapsed/refractory (RR) disease are dismal, as shown in the SCHOLAR-1 study (Crump, Blood, 2017). The combination of the anti-CD79b antibody-drug conjugates (ADC) polatuzumab vedotin with rituximab and bendamustine (PRB) has been shown to be safe and effective in heavily pre-treated patients ineligible for transplant (Sehn, Blood 2022). However, there is little to no information on the use of polatuzumab-containing regimens in RR DLBCL in Latin America (LATAM) due to a delay in regulatory approvals in LATAM countries. Here, we present the first analysis of a cohort of RR DLBCL patients from LATAM treated with polatuzumab-containing regimens. Aim. To present real-world data (RWD) of a cohort of patients with RR DLBCL in LATAM treated with polatuzumab-containing regimens. Methods. This is a retrospective study of patients treated with polatuzumab-chemotherapy from January 2019 to April 2024 in Mexico, Uruguay, Ecuador, Cuba, Paraguay, Argentina, Venezuela, Peru, and Brazil. Patients aged ≥18 years were eligible if they had histologically confirmed RR DLBCL (WHO 2016 classification), cell of origin (COO) assessed (all patients were evaluated) and received ≥1 prior line of therapy. We described the clinical characteristics of the cohort and analyzed response, progression-free survival (PFS), and overall survival (OS) using Kaplan- Meier Curves and the possible variables that influenced survival PFS or OS through proportional-hazard Cox regression models. Results. The cohort consisted of 58 patients. Frontline treatments included R-CHOP (90%), R-miniCHOP (5%), and R-EPOCH (5%). Of these 58 cases, the COO was determined in 93% (GC, 52%; non-GC, 41%), and 7% were unclassified; the Hans' algorithm was used for COO assessment in 99%). The median age at diagnosis was 61 years, and at initiation of the polatuzumab-chemotherapy regimen was 65 (p=0.03); 52% were women. 17% were older than 75 at diagnosis, and 24% at the initiation of the polatuzumab-chemotherapy regimen. 90% of patients had an Ann Arbor III/IV stage at pola-chemotherapy initiation. Low, intermediate, and high-risk IPI scores were seen in 5%, 59%, and 36%, respectively. 81% were refractory to the last prior therapy. PRB was used in 90% of patients, and polatuzumab with other agents (obinutuzumab, lenalidomide, ICE, or DHAP) in 10%. We observed a 61% overall response rate (ORR) with 42.5% complete response (CR), and 18.5% partial response (PR) rates and failure was observed in 39%. The median PFS of 54 patients was 12 months (95% CI 7 to 28); four patients are ongoing treatment and were not evaluable for PFS. The median OS of the 58 patients was 13 months (95% CI 11-28). In multivariate Cox models for PFS, non-GC COO subtype (HR 2.26; 95% CI 1.06 to 4.78) and refractory disease before using polatuzumab-chemotherapy (HR 4.41; 95% CI 1.04 to 18.5) were associated with an inferior PFS. The only variable associated to OS was response obtained after polatuzumab-chemotherapy (HR 2.21; 95% CI 1.56 to 3.12). Conclusion. Using polatuzumab-containing regimens in our highly refractory cohort was associated with 61% responses, of which 42.5% were complete. Median PFS and OS were similar to previous reports and consistent with the poor outcomes these patients have. The non-GC COO subgroup might be associated with inferior PFS and OS in LATAM patients.