Dynamic risk stratification using early CAR-T expansion in R/R LBCL treated with axicabtagene ciloleucel.

Abstract

The clinical utility of routine monitoring of chimeric antigen receptor (CAR) T-cell expansion post-infusion remains controversial. We conducted a retrospective analysis of a prospectively collected cohort of 176 patients with relapsed/refractory large B-cell lymphoma (LBCL) treated with axicabtagene ciloleucel (axi-cel). CAR T-cell expansion was measured by flow cytometry at days 7, 14 and 28 following infusion with real-time reporting to the clinical service in 111 patients. We examined the association between CAR T-cell expansion, efficacy (progression-free survival [PFS]), toxicity (cytokine release syndrome [CRS] and immune effector cell-associated neurotoxicity syndrome [ICANS]) and toxicity management, accounting for baseline risk factors. A multivariable model combining day 7 expansion with pre-lymphodepletion lactate dehydrogenase (LDH) and bridging response demonstrated that robust day 7 expansion (≥48 cells/μL) was independently associated with improved PFS (hazard ratio [HR], 0.43; 95% confidence interval [CI] 0.22-0.83; p = 0.01). Higher day 7 expansion was associated with increased CRS severity and ICANS incidence. Higher corticosteroid exposure reflected this toxicity burden but was not associated with inferior PFS. Thus, incorporating quantitative CAR-T enumeration into clinical practice may dynamically refine risk stratification post-infusion and guide individualised management, such as the frequency of monitoring for progressive disease in the era of effective subsequent therapies.