Impact of fludarabine exposure on CAR T-cell outcomes in patients with large Bcell lymphoma

The impact of fludarabine pharmacokinetics (PK) on chimeric antigen receptor (CAR) T-cell outcomes is not fully elucidated. We hypothesized that, in relapsed/refractory (R/R) large B-cell lymphoma (LBCL) patients, measured fludarabine exposure during lymphodepletion would correlate with CAR Tcell results. We conducted a prospective, single-center study including R/R LBCL patients treated with commercial, CD19-targeted CAR T-cell products. Up to seven plasma samples per patient were collected at predefined intervals during lymphodepletion. Fludarabine exposure (area under the curve, AUC) was estimated using a population PK model. An iterative cubic spline approach identified optimal AUC exposure ranges specific to each CAR-T product. The primary endpoint was progression-free survival (PFS). Among 54 patients with a median follow-up of 24.5 months, the median fludarabine AUC was 18.12 mg*h/L (IQR 14.56-20.55). Optimal exposure windows were 17.5-21.5 mg*h/L for axicabtagene ciloleucel (axi-cel) and 14.0-18.0 mg*h/L for tisagenlecleucel (tisa-cel), with 24 patients (44%) inside these ranges. Median PFS after CAR-T infusion was not reached vs 13.3 months in the optimal vs other cohort (p=0.03). In the multivariable analysis, optimal exposure remained independently associated with improved PFS (p=0.03), together with axi-cel treatment (p=0.03). CAR T-cell expansion and rates of any grade CRS, ICANS and hematotoxicity were similar between groups. In conclusion, prospective measurement of fludarabine exposure identified an optimal AUC window associated with prolonged PFS in R/R LBCL patients receiving CAR T-cell therapy. These findings support the integration of PK-guided fludarabine dosing to enhance therapeutic outcomes.