Lung Cancer 2015 Nov 12;90(2):307-13. Epub 2015 Aug 12.
Department of Thoracic Oncology, National Cancer Center Hospital East, Japan.
Background: Many patients are forced to discontinue treatment with EGFR tyrosine kinase inhibitors (TKIs), particularly gefitinib, due to severe hepatotoxicity. Here, we investigated the association between the rate of severe hepatotoxicity and single nucleotide polymorphisms (SNPs) in metabolic enzymes.
Materials And Methods: Multi-SNP analyses were performed in 60 patients with EGFR-mutated non-small cell lung cancer using blood samples obtained prior to starting treatment with gefitinib. The poor metabolizer (PM) phenotype was defined as homozygosity or double heterozygosity for variant alleles that confer reduced enzyme activities. Associated enzymes were screened using univariate logistic regression analyses adjusted for multiplicity and were further evaluated using multivariate logistic regression analyses to determine the influence of these enzymes on severe hepatotoxicity.
Results: Severe hepatotoxicity was detected in 19 (32%) of the 60 patients. Patient phenotypes consisted of CYP3A5, PM/non-PM (31/29) and CYP2D6, PM/non-PM (5/55). In multivariate logistic regression analyses, the rate of severe hepatotoxicity was significantly higher among patients with PM phenotypes than those without (CYP3A5 PM vs. non-PM: 48.4% vs. 13.8%, P=0.0069; CYP2D6 PM vs. non-PM: 80.0% vs. 27.3%, P=0.0364). Of the 9 patients switched from gefitinib to erlotinib due to severe hepatotoxicity, 8 had a PM phenotype for CYP2D6 or CYP3A5. All cases were successfully managed without exacerbation of severe hepatotoxicity.
Conclusions: Evaluation of SNPs in CYP3A5 and CYP2D6 can effectively predict severe hepatotoxicity induced by gefitinib. Erlotinib can be used as an alternative treatment for patients who develop gefitinib-induced severe hepatotoxicity.