Iris K. Minichmayr, Siv Jönsson
Department of Pharmaceutical Biosciences, Uppsala University, Uppsala, Sweden
Objectives: The topoisomerase inhibitor irinotecan (CPT-11) is a valuable treatment option primarily for colorectal cancer [1]. It is bioactivated by hydrolysis into its more potent metabolite SN-38 and further metabolised by uridine diphosphate-glucuronosyltransferase 1A1 (UGT1A1) to the inactive SN-38-glucuronide (SN-38-G). Neutropenia is a common adverse reaction associated with irinotecan, particularly in patients with genetic variants involving reduced UGT enzyme activity. To avoid complications from neutropenia, a reduction of the initial dose by 30% (if >250 mg/m2) has been recommended for poor metabolisers homozygous for the UGT1A1*28 allele [2]. The objective of this analysis was to integrate several model sources to (i) illustrate the impact of genetic variants of UGT1A1 on neutropenia following conventional and pharmacogenomics (PGx)-based dosing of irinotecan and to (ii) evaluate alternative study designs for their power to show potential superiority of PGx-based dosing with respect to the occurrence of neutropenia.
Methods: A multicompartmental PK model [3] was used to predict plasma concentrations of CPT-11 (number of individuals nID=109), SN-38 (nID=109) and SN-38-G (nID=83) in patients with solid tumours. The model was extended with a covariate effect of UGT1A1 genotype on the clearance of SN-38 (CL -35.7% in poor metabolisers) [4]. Furthermore, body surface area (BSA) was included as a covariate on the distribution and elimination parameters of all three entities. SN-38-induced neutropenia was modelled using a semi-mechanistic PK/pharmacodynamic (PD) model for myelosuppression [5] (nID=20). Stochastic simulations (n=500; BSA 1.3-2.5 m2) were performed to assess the impact of UGT1A1 genotype and dosing (350 mg/m2 versus 245 mg/m2, 90 minutes i.v.) on the occurrence of grade 4 neutropenia (neutrophil counts <0.5·109 cells/L according to CTCAE criteria) during a dosing interval (21 days; assuming 1 measurement of neutrophils/week). Diverse study designs were simulated (sample size n=50-400/treatment arm, parallel vs. crossover, varied values of CLSN-38/magnitude of exposure-response relationship/associated interindividual variability IIV) and the power to show a difference in grade 4 neutropenia after the two dosing regimens was assessed (χ2 test and McNemar’s test, α=0.05). Population modelling and simulations were executed by PsN v4.8.11 (https://uupharmacometrics.github.io/PsN) using NONMEM7.3 [6, 7]. Statistical and graphical analyses were conducted using R 3.4.3 (CRAN.R-project.org).
Results: The semi-mechanistic PK/PD model adequately predicted neutropenia driven by SN-38. The drug effect on neutrophils was modelled using a linear function (slope·conc; slope=26.8 µM-1). System-related parameters were in the range of previously published values (neutrophil baseline: 5.58·109 cells/L, mean transit time: 92.8 h) [5]. Simulations suggested a marked effect of patients’ genotype on irinotecan toxicity: Following a single dose of 350 mg/m2 irinotecan, 17% (median) of patients with wildtype-genotype displayed grade 4 neutropenia (95% prediction interval PI95=12-22%). The corresponding proportion in patients with UGT1A1*28 genotype was 25% (PI95=20-31%) and decreased given a reduced dose of 245 mg/m2 (14%; PI95=9.5-19%). To achieve a power >80%, 200 patients per treatment arm would be required for a study with parallel design and n=100 for a crossover study. Changes in the strength of the UGT1A1 covariate and drug effect relationships had a high influence on the power.
Conclusions: Simulations with a population PK/PD model describing the PK of irinotecan and its metabolites as well as myelosuppression suggested that genetic variants associated with reduced UGT1A1 enzyme activity increase the risk of severe neutropenia in agreement with clinical observations. Power analyses indicated a high influence of the study design (i.e. crossover vs. parallel), the extent of CL reduction in poor metabolisers and the myelosuppressive drug effect on study power. The model will be extended to include also the risk of diarrhoea. Besides, model-based power analysis may serve to further inform the design of future clinical trials investigating the benefit of PGx-based dosing to improve the safety and efficacy of irinotecan therapy, particularly in patients with aberrant genotypes.
The research leading to these results has received funding from the European Community’s Horizon 2020 Programme under grant agreement No. 668353 (U-PGx).
References:
[1] Pfizer. Camptosar® (Irinotecan) Injection, intravenous infusion – Summary of product characteristics (2014)
[2] Swen JJ, Nijenhuis M, de Boer A, Grandia L, Maitland-van der Zee AH, Mulder H, Rongen GA, van Schaik RH, Schalekamp T, Touw DJ, van der Weide J, Wilffert B, Deneer VH, Guchelaar HJ. Pharmacogenetics: from bench to byte–an update of guidelines. Clin Pharmacol Ther, 2011; 89(5):662-73.
[3] Xie R, Mathijssen RH, Sparreboom A, Verweij J, Karlsson MO. Clinical pharmacokinetics of irinotecan and its metabolites in relation with diarrhea. Clin Pharmacol Ther, 2002; 72(3):265-75.
[4] Valenzuela Jiménez B, González Sales M, Escudero Ortiz V, Martínez Navarro E, Pérez Ruixo C, Rebollo Liceaga J, González Manzano R, Pérez Ruixo JJ. [Influence of genetic polymorphisms in UGT1A1, UGT1A7 and UGT1A9 on the pharmacokynetics of irinotecan, SN-38 and SN-38G]. Farm Hosp, 2013; 37(2):111-27.
[5] Friberg LE, Henningsson A, Maas H, Nguyen L, Karlsson MO. Model of chemotherapy-induced myelosuppression with parameter consistency across drugs. J Clin Oncol, 2002; 20(24):4713-21.
[6] Keizer RJ, Karlsson MO, Hooker A. Modeling and Simulation Workbench for NONMEM: Tutorial on Pirana, PsN, and Xpose. CPT Pharmacometrics Syst Pharmacol, 2013;2:e50.
[7] Beal S, Sheiner LB, Boeckmann A, Bauer RJ. NONMEM User’s Guides (1989-2009) [Internet]. Icon Development Solutions, Ellicott City, MD, USA; 2009. Available from: http://www.iconplc.com/innovation/nonmem
Reference: PAGE 28 (2019) Abstr 9185 [www.page-meeting.org/?abstract=9185]
Poster: Drug/Disease Modelling - Other Topics