Initial experience with anti-CD19 CAR-T cell therapy in patients with Relapsed/Refractory diffuse large B-cell lymphoma in a brazilian cancer center

Abstract

Abstract INTRODUCTION: Chimeric antigen receptor T-cell (CAR-T) therapy targeting CD19 has emerged as a promising therapeutic modality for patients with relapsed or refractory (R/R) diffuse large B-cell lymphoma (DLBCL). Since the approval of axicabtagene ciloleucel (axi-cel) and tisagenlecleucel (tisa-cel) in Brazil in 2022, access to cellular therapies has significantly expanded. However, real-world data regarding their feasibility, outcomes, and associated toxicities remain limited. This study aims to describe the initial institutional experience with CAR-T cell therapy in patients with R/R DLBCL, comparing the two anti-CD19 products available in Brazil at the time. METHODS: Retrospective analysis of medical records from R/R DLBCL patients treated with tisa-cel or axi-cel at our institution (2022–2025). RESULTS: 23 patients were included, 11 treated with tisa-cel (47.8%) and 12 with axi-cel (52.2%). The median age was 56 (tisa-cel) and 52 (axi-cel), with a male predominance in both groups (14 vs 9 female). Functionality was assessed by ECOG: most axi-cel patients were ECOG 0 (91.7%), while tisa-cel patients were split between ECOG 0 and 1 (45.5% each). The main histology subtype was germinal center; few searched for double-hit mutations, with only 2 confirmed by FISH in the tisa-cel group. Advanced stage disease (Ann Arbor III/IV) and elevated lactate dehydrogenase (median value:321 U/L; Upper Limit:227U/L) were similar between groups. Bulky disease was more common in patients submitted to tisa-cel (36,4% vs. 16,7% axi-cel). Most patients received ≥3 prior therapy lines, with 4 lines predominant in axi-cel (50%) and 3 in tisa-cel (45.5%). At infusion, progressive disease (PD) was more frequent in tisa-cel (81.1%), while partial response (PR) predominated in axi-cel (50%). Median brain-to-vein time was 109 days (tisa-cel) vs. 122 days (axi-cel). Hematologic toxicity risk (CARHEMATOTOX >2) was higher in the tisa-cel group, with 18.2% scoring 6. Infections were more frequent in the tisa-cel group (90.9% vs. 66.7%). ICU admission was similar (63.6%), but stays were longer with axi-cel (16 vs. 4 days). Intubation was also more common with axi-cel (25% vs. 9.1%). Grade 2 CRS predominated in both groups; grade 3 occurred only with tisa-cel. Tocilizumab use was similar. ICANS was more frequent with axi-cel (75%), leading to greater Anakinra use (50%), while 72% of tisa-cel patients had no ICANS. At day 30, 34% achieved CR (75% tisa-cel, 25% axi-cel), 13% PR (66% axi-cel, 34% tisa-cel), and 21% PD (40% tisa-cel, 60% axi-cel). At day 60, CR was 30% (57% tisa-cel, 43% axi-cel), 1 PR (axi-cel), and 30% PD (28% tisa-cel, 72% axi-cel). At day 180, 13% were in CR (66% tisa-cel, 34% axi-cel), 1 PR (axi-cel), and 17% (n=4) in PD—all axi-cel. There were 7 deaths: one at day 30 (axi-cel), one at day 60 (tisa-cel), and three at day 180 (2 tisa-cel, 1 axi-cel). Follow-up was unavailable for 20% at day 30, 8% at day 60, and 30% at day 180. CONCLUSION: This Brazilian real-world analysis demonstrates that both CAR-T products are feasible but differ in efficacy and toxicity. Tisa-cel showed superior early responses despite higher-risk disease, while axi-cel was associated with more neurologic toxicity and early mortality. However, it is important to highlight the limitations of our data, considering the small sample size, missing follow-up registries and the heterogeneity in baseline characteristics, precluding a robust comparative analysis between products. Finally, real-world challenges, such as the prolonged brain-to-vein time and high ICU utilization, underscore logistical and toxicity management hurdles in middle-income countries like Brazil.