Impact of fludarabine exposure on overall survival in patients receiving fludarabine/cyclophosphamide reduced intensity conditioning hematopoietic cell transplantation for hematologic malignancies

Abstract

Abstract Allogeneic hematopoietic cell transplantation (alloHCT) is a potentially curative treatment for many hematologic malignancies. Reduced-intensity conditioning (RIC) has expanded transplant options, especially for older and frail patients, reducing transplant-related mortality (TRM). Fludarabine (Flu) is the most common conditioning agent used. It is typically dosed based on body surface area (BSA). However, BSA-based dosing is associated with significant variability in drug exposure. Dose optimization using population pharmacokinetic (popPK) modeling has been evaluated to better calculate and predict patient-specific Flu exposure. Langenhorst et al. utilized a popPK model to determine an optimal Flu exposure in patients undergoing myeloablative conditioning; over- or under-exposure of Flu was associated with increased TRM or graft failure. Increasing evidence now suggests that non-optimal Flu exposure is also associated with poor outcomes in patients receiving RIC. To our knowledge, no studies have yet validated the popPK-based dosing thresholds proposed by Langenhorst et al. in adult patients undergoing RIC. We conducted a single-center retrospective study of adult patients who underwent HLA-matched (8/8) or mismatched (7/8) unrelated donor peripheral blood alloHCT after RIC for hematologic malignancies on two prospective clinical trials (NCT02356159; NCT00520130). The RIC regimen consisted of concurrent cyclophosphamide (1200 mg/m2/day) and Flu (30 mg/m2/day) administered for 4 days (days −6 to −3). Patients with renal impairment (creatinine clearance <70 mL/min/1.73 m2) received a 20% dose reduction. Graft-versus-host disease (GVHD) prophylaxis consisted of either alemtuzumab and cyclosporine (AC) or tacrolimus, methotrexate, and sirolimus (TMS). Flu PK was simulated using a popPK model adapted from Langenhorst et al., which demonstrated body weight and eGFR to be strong predictors of drug exposure (i.e., cumulative area under the curve over 4 Flu doses [cAUC]). Previously established drug exposure thresholds were applied (optimal Flu cAUC: 20±5 mg*h/L). The primary endpoint was 1-year overall survival (OS), adjusted for previously identified risk factors in this population. Patients were censored at one year for statistical analysis. We included 114 patients undergoing alloHCT between 10/2007 and 11/2022. The median age was 47.9 years (range: 20.1–71.1); 72 (63.1%) were males. 76 patients (66.7%) had HCT-CI ≥2 and 25 (21.9%) were ≥60 years. The most common indication for alloHCT was B-NHL (26.3%) followed by CLL (14.9%), MDS/AML (14.9%), and ALL (13.1%). 97 patients (85.1%) received MUD grafts. GVHD prophylaxis consisted of TMS in 70 patients (61.4%), while the remaining patients received AC. The average simulated Flu cAUC was 17.9 mg*h/L (range: 12.1–28.18). For 84 patients (73.7%), the cAUC was within the optimal range, while 26 patients were underexposed and only 4 patients were overexposed. Seven patients received renally adjusted dosing. Patients were followed for a median of 172 months (95% CI: 154–257). At one year, 76 patients (66.7%) were alive: 63.1% (53/84) in the optimal-exposure group and 76.7% (23/30) in the non-optimal group. However, the difference in 1-year OS between the two groups was not statistically significant (log-rank p=0.20), even when adjusted for GVHD prophylaxis, HLA matching, CMV positivity, age >60 years, and HCT-CI (HR=2.03 [95% CI: 0.87-4.77], p=0.10). In multivariate Cox regression, factors significantly associated with improved 1-year OS included GVHD prophylaxis with TMS (HR=0.52 [0.27-0.99], p=0.05) and use of MUD graft (HR=0.39 [0.18-0.81], p=0.01). In contrast, CMV positivity (HR=2.40 [1.14-5.05], p=0.02) and HCT-CI ≥2 (HR=2.81 [1.17-6.74], p=0.02) were associated with worse OS at one year. In conclusion, cumulative 120 mg/m2 Flu dosing with renal adjustment achieved an optimal Flu cAUC of 20±5 mg*h/L in the majority of patients (73.7%) per the popPK model simulation. In this analysis, achieving optimal Flu exposure was not associated with improved OS at 1-year in patients undergoing alloHCT with RIC. While our findings suggest that the current dosing approach is effective in preventing Flu overexposure, larger studies are needed to adequately assess the impact of Flu exposure on survival outcomes. Future prospective research is warranted to refine and validate optimal Flu exposure in the RIC alloHCT setting.