Population Pharmacokinetics of Ledipasvir/Sofosbuvir in Pediatric Patients: Impact of Acute Lymphoblastic Leukemia

Purpose: The pharmacokinetic (PK) profile of direct-acting antivirals, namely ledipasvir/sofosbuvir (LDV/SOF), might be altered in patients with acute lymphoblastic leukemia (ALL), affecting the optimum dose needed for hepatitis C virus treatment. Limited data are available evaluating the population PK of LDV/SOF and SOF metabolite GS-331007. We aimed to study whether ALL could affect population PK parameters of LDV, SOF, and the SOF major metabolite GS-331007 in hepatitis C virus-infected children, develop and validate a predictive PK model of LDV/SOF disposition in this special population, and identify their explained and unexplained sources of variability.

Methods: Population PK modeling was performed using MonolixSuite software using the non-linear mixed effect modeling approach. Different compartmental models, absorption models, and lag times for absorption parameters were tested to find out the best-fitting base model. For final model development, data-driven systematic covariate analysis using conditional sampling for the stepwise approach based on the correlation tests method has been performed. The final models were then evaluated using internal validation approaches.

Findings: The PK results of 22 fully compliant patients were included in the population PK analysis. LDV and SOF were best described by a 1-compartment model with zero-order absorption and lag time, while the 2-compartment model with first-order absorption and lag time was the best-fitting model for the SOF metabolite. The internal validation approach confirmed the good predictive power of the selected models. The patients' weight explained the variability in the volume of distribution of LDV and the systemic clearance of SOF and LDV. The final SOF model also included a statistically significant covariate of steatosis stage on its volume of distribution, while the final GS-331007 model included mean corpuscular volume values on GS-331007 central compartment volume, packed cell volume, and direct bilirubin values on metabolite intercompartmental clearance.

Implications: The presence of ALL did not explain any variability in the developed population PK models for SOF, LDV, and GS-331007. Despite weight being a significant covariate in the final models suggesting that weight-based dosing of LDV/SOF is better than fixed dosing, the fixed dosing (45/200 mg LDV/SOF) is more practical in terms of simplicity in dosing children at home besides the proved efficacy and safety through both the clinical outcomes and PK exposure results. Weight-based dosing is still hindered due to the absence of exposure-response analysis, and the unavailability of dose-flexible formulas in the market. Future studies are required to support these findings.

Gov identifier: NCT03903185.