Daniel Centanni 1, Anne Mols Krarup 2,3, Merete Dam 2,3, Birgitte Klug Albertsen 2,3, Mats Karlsson 1, Lena Friberg 1
1 Department of Pharmacy, Uppsala University (Uppsala, Sweden), 2 Department of Clinical Medicine, Aarhus University (Aarhus, Denmark), 3 Department of Paediatrics and Adolescent Medicine, Aarhus University Hospital (Aarhus, Denmark)
Introduction: Acute lymphoblastic leukemia (ALL) is the most common childhood malignancy, with an estimated 103,727 new cases reported globally in 2021. [1,2]. Asparaginase-based therapy is a cornerstone of ALL treatment, and pegylated asparaginase (peg-asparaginase) is the most used formulation in contemporary protocols. However, peg-asparaginase treatment is complicated by substantial variability in asparaginase enzyme activity (AEA) both between patients and across dosing occasions. Asparaginase treatment is also associated with several toxicities, such as hypersensitivity reactions, which remain a significant clinical concern [2,3].
Objectives: This study aimed to characterize the pharmacokinetics (PK) of peg-asparaginase following intravenous (IV) and intramuscular (IM) dosing by integrating pediatric and adult PK data across multiple protocols. This one-stage meta-analysis approach increases sample size and population diversity, potentially enabling a more robust and comprehensive characterization of peg-asparaginase PK than single-protocol analyses. Specifically, we sought to expand the relatively limited population pharmacokinetic knowledge regarding absorption variability following IM dosing and to further delineate the apparent time-dependence in peg-asparaginase PK after repeated IV and IM administrations. An additional objective was to identify significant covariates influencing PK within a meta-analytical framework.
Methods: Data from three NOPHO treatment protocols (ALL2008, ALLTogether Pilot, and ALLTogether1) were pooled for population PK modeling, comprising patients aged 0–45 years receiving peg-asparaginase IV or IM according to protocol-specific age-based dosing regimens (1000–1500 IU/m², typically every 2 weeks in continuous schedules, 4–11 doses). AEA was quantified using a validated L-aspartic acid β-hydroximate assay, with samples obtained according to TDM schedules that included peak, day 1, 4, 7,11, and 14 sampling, depending on the protocol and dose number [4-6]. Data handling and visualization were performed in R. Model development was performed in NONMEM® 7.5.1, supported by PsN and Pirana using IMPMAP estimation. Concentrations below the LLOQ (5 IU/L) and above the ULOQ (1000 IU/L), were analyzed using M4 and M3 methods, respectively. Interindividual and inter-occasion variabilities were tested on parameters, including logit-transformed bioavailability. Model selection was based on likelihood ratio tests, Akaike information criterion (AIC) for non-nested comparisons, goodness of fit diagnostics, and visual predictive checks. Final model performance was evaluated using external data from an additional cohort of ALLTogether1 patients.
Results: A one‑compartment model with a hill function based on time after last dose best described the within‑occasion increase in drug clearance (T50 = 23 days) providing an improved fit compared with the previously evaluated transit‑compartment structure (ΔAIC = –65.75; +2 parameters) [7,8]. For the former, the structural model included parallel compartments to adequately describe time-after-dose–specific clearance trends related to de-pegylation for each administration. For IM dosing, a sequential zero‑order followed by first‑order absorption process significantly improved the fit over first‑order alone (ΔOFV = –143.73; OFV_extended − OFV_base). Beyond the within-dose increase, clearance decreased across administrations and was modeled as a constant baseline plus a monoexponentially declining component, which accounted for 58% of total clearance at treatment start and declined with a rate constant of 0.052 day⁻¹ (ΔOFV = –1518.43). Finally, asparaginase drug inactivation was also modeled through mixture modeling that identified a subpopulation with accelerated clearance over successive doses, in contrast to the decrease observed in the main population, and markedly improved the overall fit (ΔOFV = –6066.95). Overall, the final model captured distinct clearance trajectories and adequately described PK profiles across IV and IM dosing. Clearance and volume were allometrically scaled to body weight; substituting with body surface area did not improve the fit (ΔOFV = +0.34). High-risk ALL status was associated with a 43% higher baseline clearance than in standard-risk patients (ΔOFV = –64.98). IM bioavailability was 34% lower in high-risk patients (ΔOFV = –333.11) and 51% lower in those with BMI >25 kg/m² (ΔOFV = –12.21). Discontinuous treatment and high-risk status were associated with higher likelihood of asparaginase inactivation (ΔOFV = –157.73 and –70.53).
Conclusion: Using an integrated multi-protocol population PK approach, we developed a model that adequately described the PK of peg-asparaginase following IV and IM dosing. The model captured key clearance dynamics, including the periodic increase in clearance due to (i) de-pegylation within a dosing interval, (ii) hypersensitivity development, and (iii) a time-dependent reduction in CL. Beyond allometric scaling, several patient characteristics were found to influence typical asparaginase activity and may warrant tailored dosing strategies to optimize drug exposure.
References:
References:
1. Fan, H., et al. (2025). Global burden of acute lymphoblastic leukemia following the COVID-19 pandemic. Annals of Hematology, 104(9), 4713–4727.
2. Brown, P., et al. (2020). Pediatric acute lymphoblastic leukemia, version 2.2020, NCCN clinical practice guidelines in oncology. Journal of the National Comprehensive Cancer Network, 18(1), 81–112.
3. Kloos, R. Q. H., et al. (2020). Individualized asparaginase dosing in childhood acute lymphoblastic leukemia. Journal of Clinical Oncology, 38(7), 715–724.
4. Dam, M., et al. (2024). Increase in peg-asparaginase clearance as a predictor for inactivation in patients with acute lymphoblastic leukemia. Leukemia, 38, 712–719.
5. Dam, M., et al. (2025). Early prediction of subsequent peg-asparaginase inactivation in acute lymphoblastic leukaemia patients: A NOPHO ALL2008 study. British Journal of Haematology, 207(5), 1920–1929.
6. Heyman, M., et al. (2019). A treatment protocol for participants 0–45 years with acute lymphoblastic leukaemia (Clinical Trial No. NCT03911128). ClinicalTrials.gov.
7. Würthwein, G., et al. (2020). Therapeutic drug monitoring of asparaginase: Intra-individual variability and predictivity in children with acute lymphoblastic leukemia treated with peg-asparaginase in the AIEOP-BFM acute lymphoblastic leukemia 2009 study. Therapeutic Drug Monitoring, 42(3), 435–444.
8. Würthwein, G., et al. (2017). Population pharmacokinetics to model the time-varying clearance of the PEGylated asparaginase Oncaspar in children with acute lymphoblastic leukemia. European Journal of Drug Metabolism and Pharmacokinetics, 42(6), 955–963.
Reference: PAGE 34 (2026) Abstr 12068 [www.page-meeting.org/?abstract=12068]
Poster: Drug/Disease Modelling - Oncology