Pharmacogenetically guided dosing of the anticancer agent indisulam normalizes the risk of severe hematological toxicity
Zandvliet, Anthe S.(1), Alwin D.R. Huitema (1), William Copalu (2), Yasuhide Yamada (3), Tomohide Tamura (3), Jos H. Beijnen (1,4) , Jan H.M. Schellens (4,5)
(1) Department of Pharmacy & Pharmacology, The Netherlands Cancer Institute/Slotervaart Hospital, Amsterdam, The Netherlands; (2) Eisai Ltd., London, UK; (3) Medical Oncology Division, National Cancer Center Hospital, Tokyo, Japan; (4) Department of Biomedical Analysis, Section of Drug Toxicology, Utrecht University, Utrecht, The Netherlands; (5) Department of Medical Oncology, The Netherlands Cancer Institute/Antoni van Leeuwenhoek Hospital, Amsterdam, The Netherlands
Introduction and objectives: The anticancer agent indisulam is metabolized by the cytochrome P450 enzymes CYP2C9 and CYP2C19. Pharmacokinetics of indisulam have previously been described with a physiologically-based population pharmacokinetic model, in which drug elimination was described by a linear and a saturable pathway. Polymorphisms of the CYP2C enzymes may affect the elimination rate and drug exposure by inhibition of either of these pathways. It was previously demonstrated that the occurrence of the major hematological side effects was highly related to indisulam exposure. A semi-physiological pharmacokinetic-pharmacodynamic model has previously been successfully applied to describe the time course and extent of neutropenia and thrombocytopenia after treatment with indisulam. Variant genotypes of CYPC9 genes may increase exposure to indisulam and may consequently be clinically relevant predictors for the risk of developing severe hematological toxicity. The objectives of this study were 1) to evaluate the effect of genetic variants of CYP2C9 and CYP2C19 on the pharmacokinetics of indisulam, 2) to determine the relative risk of dose limiting neutropenia and 3) to develop a pharmacogenetically guided dosing strategy.
Methods: Pharmacogenetic screening of CYP2C polymorphisms was performed in 67 patients treated with indisulam. Both Caucasian (n=46) and Japanese patients (n=21) were included, because mutant allele frequencies of CYP2C9 were expected to be high in Caucasians and mutant allele frequencies of CYP2C19 were expected to be high in Japanese populations. Pharmacokinetic data were analyzed with the population pharmacokinetic model of indisulam, which comprises a linear and a Michaelis Menten pathway for drug elimination. The relationships between allelic variants of CYP2C genes and each of the elimination pharmacokinetic parameters (k10, Vmax, Km) were verified in a multivariate analysis using non-linear mixed effects modeling (NONMEM).
Dose limiting neutropenia was defined as CTC grade 4 neutropenia (absolute neutrophil count <0.5*10^9/L) during at least seven days. To assess the relative risk of dose limiting neutropenia for variant genotypes, a simulation study was performed using the previously developed semi-physiological PK-PD model, which describes the relationship between indisulam pharmacokinetics and neutropenia. For each genotype, a population of 10,000 patients was simulated to receive the recommended indisulam dosage of 700 mg/m2 and the proportion of patients who experienced dose limiting neutropenia was assessed. A pharmacogenetically guided dosing strategy was developed by reiteration of the simulation study. Populations of 10,000 patients with one or more high risk mutations were simulated to receive a reduced dosage of indisulam and normalization of the relative risk of neutropenia was verified.
Results: The Michaelis Menten elimination rate (Vmax) was decreased by 24% (95% confidence interval: 9.6-36%) (P<0.0001) for heterozygous CYP2C9*3 mutants. The heterozygous *2 and *3 mutations in the CYP2C19 gene reduced the linear elimination rate (k10) by 44% (95% CI: 37-51%) (P=0.0002) and 70% (95% CI: 31-94%) (P<0.0001), respectively.
As expected, the risk of neutropenia could not be adequately assessed using clinical data due to small patient numbers. By means of the simulation study we succeeded to demonstrate that the risk of severe neutropenia was significantly increased for patients with CYP2C9*3, CYP2C19*2 and CYP2C19*3 mutations. For instance, the relative risk of dose limiting neutropenia was 2.1 for homozygous Caucasian CYP2C9*3 mutant patients and 4.0 for homozygous Japanese CYP2C19*2 mutant patients after administration of the recommended dosage of 700 mg/m2 indisulam. A dose adaptation score was calculated by assigning 3 points for each CYP2C19*3 allele, 2 points for CYP2C19*2 and 1 point for CYP2C9*3. The total score was multiplied by 50 mg/m2 to calculate the dose reduction. The recommended starting dose of indisulam was 700 mg/m2 minus the dose reduction. This strategy for dose adaptation resulted in normalization of the risk of severe neutropenia (relative risk 0.91-1.27) in patients with CYP2C polymorphisms.
Conclusions: CYP2C9*3, CYP2C19*2 and CYP2C19*3 polymorphisms resulted in a reduced elimination rate of indisulam. Patients with one or more of these CYP2C mutations have a higher risk of developing severe side effects. Screening for CYP2C polymorphisms before treatment with indisulam is recommended. The initial indisulam dosage should be reduced in patients with high risk mutations.
 Zandvliet AS, Yasuda S, Copalu W, Schellens JHM, Beijnen JH, Huitema ADR. A physiologically based population pharmacokinetic model describing the non-linear disposition and blood distribution of indisulam in Caucasian and Japanese patients. PAGE. Abstracts of the Annual Meeting of the Population Approach Group in Europe. 2005 No. 737. (oral presentation)
 van Kesteren C, Zandvliet AS, Karlsson MO, Mathot RAA, Punt CJA, Armand JP, Raymond E, Huitema ADR, Dittrich C, Dumez H, Roche H, Droz JP, Ravic M, Yule SM, Wanders J, Beijnen JH, Fumoleau P, Schellens JHM. Semi-physiological model describing the hematological toxicity of the anti-cancer agent indisulam. Invest New Drugs 2005; 23(3):225-234.