Page 35 - 2021_05-Haematologica-web

Individualized PEGasparaginase dosing guideline
apeutic drug monitoring (TDM). However, more insight into PEGasparaginase population (popPK) – and especially in characteristics explaining the variability – is needed to optimize individualized dosing. Recently, Hempel et al. described the body surface area (BSA) as an important fac- tor explaining variability in CL. However, the influence of other patient factors is still unknown.17 Therefore, the aim of this study was to describe the PK of PEGasparaginase in our cohort, to gain more insight into factors influencing the CL, and to develop a dosing guideline for PEGasparaginase therapy in children.
Methods
Patients and treatment protocol
Patients (1-18 years old) with newly diagnosed ALL between November 2014 and May 2017, treated according to the ALL-11 protocol in the Sophia Children’s Hospital–Erasmus MC, Rotterdam, were included. In these patients trough, top, week and other levels were prospectively measured. In addition, asparaginase levels measured after the last dose, as part of the TDM program from patients from other Dutch pediatric oncol- ogy centers, were included. The study was approved by the Institutional Review Board (clinicaltrials gov. Identifier [CCMO register]: NL50250.078.14). Informed consent was obtained from patients ≥12 and the parents in accordance with the Declaration of Helsinki.
Patients were stratified as standard (SR), medium (MR) and high risk (HR) after induction. Online Supplementary Table S1 describes the complete treatment for each risk group. Figure 1 shows the concomitant chemotherapy per treatment phase. All patients were treated with three doses of PEGasparaginase (1,500 IU/m2, biweekly) during induction (protocol 1A and B). For SR and MR patients, the subsequent dose(s) were individu- alized based on trough asparaginase activity levels (Online Supplementary Table S3). SR patients received one individualized dose after an interval of approximately 12 weeks (protocol IV); MR patients received 14 individualized doses, either directly after the first three doses or after an interval of approximately 12 weeks during intensification and maintenance as part of a ran- domized study. HR patients received another two to five doses with a fixed dose of 1,500 IU/m2. Patients were switched to Erwinia asparaginase in case of a neutralizing allergy or silent inactivation. PEGasparaginase was administered intravenously over 1 hour.
Measurements
Measurements of asparaginase levels and antibodies are described in the Online Supplementary Appendix.
Population pharmacokinetic analysis
The popPK were analyzed using non-linear mixed models (see the Online Supplementary Appendix). Several structural models were evaluated using one and two compartment models with linear, nonlinear, and time-dependent elimination. Since BSA is known to highly influence the PK of PEGasparaginase, BSA (cal- culated with the Mosteller formula19) was included in the struc- tural model to scale the volume of distribution (Vd) and CL.
Covariate analysis
Several demographic, clinical and therapy-related covariates were evaluated (Online Supplementary Table S2). Infections were defined as fever (>38°C) and hospital admission or prescription of antibiotics. Beside differences in the treatment phase and con-
comitant chemotherapy, several other supportive care drugs were evaluated. First, the covariates were explored with univari- ate analyses after which significant covariates (P<0.05) were evaluated using a stepwise forward inclusion, followed by back- ward elimination (P<0.001) in a multivariate analysis.
Model validation and development of dosing guidelines
The final model was validated using goodness of fit plots and visual predictive checks (VPC), see the Online Supplementary Appendix. An independent database, which was obtained by randomly selecting 25% of the population, was used for the external validation of the model.
In order to develop dosing guidelines, Monte Carlo simula- tions were performed. Starting doses were calculated targeting trough asparaginase activity levels >100 IU/L, >250 IU/L and >350 IU/L, taking into account the significant covariates.
Dosing guidelines were developed targeting a trough asparag- inase activity level of 100-250 IU/L or 250-400 IU/L based on either week levels or trough levels.
Results
Patients and samples
In total, 120 patients were included in the study. Ninety-two patients were included in the index dataset and 28 patients in the validation dataset. Online Supplementary Table S2 describes the patient characteris- tics of the two datasets. The patients in the validation database were older (median 8.0 years, interquartile range (IQR), 3.3–12.5 years) than in the main database (median 4.8 years, IQR, 3.3–8.2 years), and had a higher weight (median 28.0 kg, IQR, 16.6–47.9 kg vs. 19.2 kg, IQR, 14.9– 29.3 kg).
Table 1 shows the characteristics and distribution of the samples. In the index database, 816 samples were meas- ured in 92 patients. The majority of the levels were top, week or trough levels, and were measured during the intensification and maintenance phase of the MR group. Figure 2 shows all asparaginase activity levels plotted against the time after dose.
Structural model
The development of the structural PK model is described in the Online Supplementary Results. The esti- mated popPK parameters are stated in Table 2. Concentration-time profiles were best described by a one-compartment model. Adding a second compartment did not improve the model. Normalization of the CL by BSA reduced the (unexplained) inter-individual variability (IIV) of this parameter from 29.6% to 24.1%. Intra- patient variability (IOV) was 25.7%. As demonstrated in Figure 2, the elimination of PEGasparaginase was not lin- ear. Several models with time dependent CL were evalu- ated.20 The structural model without the effect of covari- ates best described the data with a CL of 0.075 L/day/m2 for the first 12.9 days after a dose, increasing with 0.079 L/day/m2 per day thereafter. Thus, during the first 12.9 days, the half-life of PEGasparaginase was 8.5 days and, thereafter, decreased to 4.1 days after 1 day, and 2.7 days after 2 days.
Covariate analysis
The univariate analysis resulted in 16 covariates signif- icantly correlated with PEGasparaginase CL (Table 3). In
haematologica | 2021; 106(5)
1255