Jen-Hao Wu (1,2), Kim P J Schellekens (1,2), Charlotte Rigaud (3), Reineke A Schoot (2), Gertruud Bakker (2), Maaike Boonstra-Schelfhorst (2), Francisco Bautista (2), Michael J. Hanley (4), Samantha Lampron (4), Yousif H. Matloub (4), Florin Vranceanu (4), C. Michel Zwaan (1,2), Alwin D.R. Huitema (2,5,6)
(1) Department of Pediatric Oncology, Erasmus MC-Sophia Children’s Hospital, Rotterdam, the Netherlands; (2) Princess Máxima Center for Pediatric Oncology, Utrecht, the Netherlands; (3) Department of Children and Adolescent Oncology, Gustave Roussy Cancer Campus, Villejuif, France; (4) Takeda Development Center Americas, Lexington, MA, United States; (5) Department of Pharmacy & Pharmacology, Netherlands Cancer Institute, Amsterdam, the Netherlands; (6) Department of Clinical Pharmacy, University Medical Center Utrecht, Utrecht, The Netherlands
Objectives:
The primary aim of phase I clinical trials is to define the recommended phase II dose (RP2D) [1]. In pediatric oncology, matching the adult pharmacokinetic (PK) exposure is often a selection criterion bedsides toxicity. In many cases, non-compartmental analysis (NCA) is used to derive PK parameters in phase I trials; however, NCA may not adequately describe the PK depending on the number, timing of samples, and the number of patients (pts) [2–4]. Moreover, intense sampling design for NCA to accurately capture PK is undesirable in pediatric trials [5,6], while the small dataset often precludes developing a population PK (popPK) model solely on the pediatric data. Pediatric trials generally possess the possibility to utilize adult PK knowledge [7,8]. With adult PK information, an efficient modeling method can be applied in phase I trials to evaluate pediatric PK. Instead of NCA, pediatric PK can be determined based on a popPK model informed by the limited pediatric samples collected in the trial and adult PK using a Bayesian framework (PRIOR subroutine of NONMEM) [9,10], without accessing granular adult data. This study aimed to evaluate the feasibility of this modeling approach, estimate expected exposure by the popPK model and compare with the RP2D informed by NCA.
Methods:
The phase I part of ITCC-098 study is a brigatinib (tablet) trial in pediatric pts (age ≥1–<18y, weight (WT)>10kg) with ALK+ cancer. Two dose levels (DL) adjusted on WT bins (>10–<18kg; 18–40kg; >40kg) were tested (DL1: 30mg QD for 7 days then 60mg QD; 60mg then 120mg; 90mg then 180mg. DL2: 30mg then 90mg; 60mg then 150mg; 90mg then 240mg). PK samples were collected on Day1 and 22. On day22, three steady-state samples were collected for pts <18kg (pre-dose, 2 and 4h post dose); six for pts >18kg (pre-dose, 0.5, 1, 2, 4, 6h post dose). The RP2D is defined by the incidence of dose-limiting toxicities and the DL leading to an equivalent (± 20%) target area under concentration-time curve at steady-state over a dosing interval (AUCss,tau) of 20.3 h*mg/L for the approved adult dosing (90mg then 180mg QD). NCA was used to calculate AUCss,tau to select RP2D.
In this study, we described pediatric brigatinib PK by building a popPK model informed by the pediatric data collected in the trial and the adult PK using Bayesian framework (PRIOR subroutine of NONMEM). Expected AUCss,tau for the two DLs were informed by the popPK model clearance estimate in a simulated children dataset.
Results:
PK data from 10 pts (95 PK observations) were analyzed. NCA estimated AUCss,tau on DL1 (n = 4) was 11.7 [64.3%] (geometric mean [%CV]); on DL2 (n = 6) was 19.0 [29.5%] which was within the target range.
Brigatinib PK in children was well described by a three-compartment model with a two transit compartment input informed by the adult model [11] and fixed allometric scaling. Albumin was included as a covariate on apparent clearance (CL/F), and inter-occasion variability was included on the apparent central volume of distribution (V4/F). Non-informative priors (i.e. predominately informed by pediatric data) were used for the central PK parameters (CL/F, V4/F), apparent volume of the 1st peripheral compartment and mean transit time. No prior was used on residual error. Informative priors were used for the other parameters and inter-individual variability. The pediatric population CL/F estimate was 14.0 L/h (RSE: 14.6%), for a WT of 70 kg and an albumin of 38 g/dL. For simulated pediatric pts treated on DL1 (n = 6000), 11.4% (>10–<18kg), 48.4% (18–40kg), 22.0% (>40kg) and 30.1% (overall) of the expected AUCss,tau informed by population CL/F estimate from the popPK model fell within the target range; on DL2 (n = 6000) were 78.6%, 66.5%, 76.7% and 73.1%, respectively.
Conclusions:
This study demonstrated the feasibility to build a popPK model informed by the limited data collected and adult PK to determine pediatric PK. Majority of expected AUCss,tau of children treated on DL2 fell within the target range, which aligns with the RP2D based on NCA. Yet, unlike NCA which relied on few steady-state samples, the popPK approach accounted for all collected data. This approach can be further valuable in simulating exposure to help deciding to include more pts in a certain DL or to open others. We propose the modeling approach as an efficient and accurate way to evaluate PK of a drug in a vulnerable population in an early phase trial, where intense PK sampling design is undesirable.
References:
[1] Le Tourneau C, Lee JJ, Siu LL. Dose Escalation Methods in Phase I Cancer Clinical Trials. JNCI: Journal of the National Cancer Institute 2009;101:708–20. https://doi.org/10.1093/jnci/djp079.
[2] Straughn AB. Limitations of Noncompartmental Pharmacokinetic Analysis of Biotech Drugs. Pharmacokinetics and Pharmacodynamics of Biotech Drugs, Wiley; 2006, p. 181–8. https://doi.org/10.1002/9783527609628.ch7.
[3] Gabrielsson J, Weiner D. Non-compartmental Analysis, 2012, p. 377–89. https://doi.org/10.1007/978-1-62703-050-2_16.
[4] Patel K, M.J. Kirkpatrick C. Pharmacokinetic Concepts Revisited – Basic and Applied. Curr Pharm Biotechnol 2011;12:1983–90. https://doi.org/10.2174/138920111798808400.
[5] European Medicines Agency. Guideline on the role of pharmacokinetics in the development of medicinal products in the paediatric population n.d.
[6] Batchelor HK, Marriott JF. Paediatric pharmacokinetics: key considerations. Br J Clin Pharmacol 2015;79:395–404. https://doi.org/10.1111/bcp.12267.
[7] Doussau A, Geoerger B, Jiménez I, Paoletti X. Innovations for phase I dose-finding designs in pediatric oncology clinical trials. Contemp Clin Trials 2016;47:217–27. https://doi.org/10.1016/j.cct.2016.01.009.
[8] Paoletti X, Geoerger B, Doz F, Baruchel A, Lokiec F, Le Tourneau C. A comparative analysis of paediatric dose-finding trials of molecularly targeted agent with adults’ trials. Eur J Cancer 2013;49:2392–402. https://doi.org/10.1016/j.ejca.2013.02.028.
[9] Gisleskog PO, Karlsson MO, Beal SL. Use of prior information to stabilize a population data analysis. J Pharmacokinet Pharmacodyn 2002;29:473–505. https://doi.org/10.1023/A:1022972420004.
[10] Chan Kwong AH-XP, Calvier EAM, Fabre D, Gattacceca F, Khier S. Prior information for population pharmacokinetic and pharmacokinetic/pharmacodynamic analysis: overview and guidance with a focus on the NONMEM PRIOR subroutine. J Pharmacokinet Pharmacodyn 2020;47:431–46. https://doi.org/10.1007/s10928-020-09695-z.
[11] Gupta N, Wang X, Offman E, Prohn M, Narasimhan N, Kerstein D, et al. Population Pharmacokinetics of Brigatinib in Healthy Volunteers and Patients With Cancer. Clin Pharmacokinet 2021;60:235–47. https://doi.org/10.1007/s40262-020-00929-4.
Reference: PAGE 32 (2024) Abstr 10918 [www.page-meeting.org/?abstract=10918]
Poster: Drug/Disease Modelling - Paediatrics