Towards Personalized Lymphodepletion: A Population Pharmacokinetic Fludarabine Model in Patients Receiving CAR T-Cell...

Background/Objectives: Optimal fludarabine dosing in the conditioning regimen based on population pharmacokinetic analysis (popPK) can predict outcomes in patients receiving hematopoietic stem cell transplantation. To date, there is no popPK tailored for patients receiving fludarabine as part of the lymphodepleting regimen before chimeric antigen receptor (CAR) T-cell infusion. The objective of this study was to develop a PopPK model of fludarabine in patients receiving CAR T-cell therapy. Methods: A prospective study was conducted at a tertiary hospital, from January 2021 to July 2022. Demographic, clinical, and analytical variables were collected. Blood samples were obtained on days 1 and 3 of the lymphodepleting regimen at 1.5, 2, 7 and 24 h post-fludarabine doses, and 30 min prior to CART-cell infusion. Fludarabine levels were analyzed through an ultra-performance liquid chromatography tandem mass spectrometry assay based on liquid-liquid extraction. Population pharmacokinetic analysis modeling was performed using nonlinear mixed-effects models (NONMEM). Results: Fifty-six patients (59% male) with a median age of 59 years (range 23-82) received CAR T-cell therapy (38 [68%] axicabtagene ciloleucel, 18 [32%] tisagenlecleucel) for relapsed/refractory large B-cell lymphoma. A total of 348 samples were collected for model development. A three-compartment model with first-order elimination best described the data. Body size, as represented by weight (WGT) with allometric scaling, was a significant predictor of all pharmacokinetic parameters (p < 0.05). Estimated glomerular filtration rate (eGFR) and the CAR T-cell construct type also showed statistical significance for fludarabine clearance (CL) (p < 0.05). Clearance was differentiated into non-renal and renal components. Estimates of V1, V2 and V3 volumes (the apparent volume of distribution of the central, shallow and deep compartments) were 41.2, 14.5 and 10.8 L, respectively. Conclusions: WGT, eGFR and type of CAR-T were predictors of fludarabine pharmacokinetics. This model offers a step toward precision-guided lymphodepletion and might support individualized dosing to optimize efficacy and minimize toxicity.