P149 | CD19CAR-T CD8+ CELL EXPANSION AND CD4+/CD8+ RATIO REVERSAL ARE EARLY PREDICTIVE FACTORS OF RESPONSE TO CAR-T THERAPY.

Chimeric antigen receptor T-cell (CAR-T) therapy has proven to be very effective in B-cell malignancies, but a subset of patients does not benefit from this therapy. The aims of our study were to monitor the expansion and exhaustion of CD19CAR-T CD4+ (CAR-T CD4+) and CD19CAR-T CD8+ (CAR-T CD8+) lymphocytes after CAR-T infusion and to investigate their possible correlation with the clinical response. Our cohort included 22 patients (13 LBCL, 3 MCL, 2 PMBL and 4 ALL) undergoing CAR-T therapy between June 2021 and July 2024. Patients received Axi-cel (n=11), Tisa-cel (n=4) and Brexu-cel (n=7). Flow-cytometry was perfomed on the washing fluid of the CAR-T bags after the end of infusion and subsequently on peripheral blood samples (PB) collected once a week during the first month and then every 3 months for two year post-infusion. At median follow-up of 12,5 months (range 1-46) after CAR-T therapy, 13 patients achieved CR (responder) and 9 patients showed refractoriness or relapse (non responder). Bags analysis showed the presence of 65,35% CAR-T CD4+ and 33,20% CAR-T CD8+ (%CD19CAR-T CD4+ responder vs non-responder 60,84% vs 70,43%, p:0,2599 - %CD19CAR-T CD8+ responder vs non-responder 37,76% vs 28,06%, p:0,2730). All patients showed an increase in CAR-T CD8+ cells and thus a reversal of the CD4+/CD8+ ratio from day +7. However, this trend remained significantly stable in responder patients during the first month, while we observed an early reduction of CAR-T CD8+ cells in non-responder patients at day +21 (%CD19CAR-T CD4+ responder vs non-responder 21,59% vs 53,50%, p:0,0115 - %CD19CAR-T CD8+ responder vs non-responder 75,35% vs 26,45%, p:0,0032). We also analyzed exhaustion biomarkers TIM-3, PD-1 and LAG-3 within CAR-T cells. No significant differences were found between responder and non-responder patients. Surprisingly, in responder patients, low levels of CAR-T CD8+ TEMRA was associated with a greater expansion of CAR-T CD8+. In non-Hodgkin Lymphoma patients, we compared the expansion of CD4+ and CD8+ lymphocytes with the pre-infusion tumor burden (TB) measured by PET-CT and we observed that patients with lower TB had a lower CD4+/CD8+ ratio (p:0,0415) due to an increase in CAR-T CD8+ cells. In conclusion, the early and persistent expansion of CAR-T CD8+ cells and CD4+/CD8+ ratio reversal could represent potential predictive factors of response to CAR-T therapy. Low levels of CAR-T CD8+ TEMRA result in a more marked and persistent expansion of functionally active CAR-T CD8+ cells in the early phases after infusion. Furthermore, the basal or early expression of exhaustion biomarkers does not represent a prognostic factor for response to therapy.