1272P - Impact of single nucleotide polymorphisms on severe hepatotoxicity induced by gefitinib or erlotinib in non-small cell lung cancer patients harbori...

Date 27 September 2014
Event ESMO 2014
Session Poster Display session
Topics Anti-Cancer Agents & Biologic Therapy
Complications of Treatment
Non-Small-Cell Lung Cancer, Metastatic
Presenter Eri Sugiyama
Citation Annals of Oncology (2014) 25 (suppl_4): iv426-iv470. 10.1093/annonc/mdu349
Authors E. Sugiyama1, S. Umemura1, S. Nomura2, S. Matsumoto1, K. Yoh1, S. Niho1, H. Ohmatsu1, Y. Ohe1, M. Tsuboi3, K. Goto1
  • 1Division Of Thoracic Oncology, National Cancer Center Hospital East, 277-8577 - Kashiwa/JP
  • 2Resarch Center For Innovative Oncology, Clinical Trial Section, National Cancer Center Hospital, 277-8577 - Kashiwa/JP
  • 3Division Of Thoracic Surgery, National Cancer Center Hospital East, 2778577 - Kashiwa/JP

Abstract

Aim

Although EGFR-tyrosine kinase inhibitors (TKIs) are essential drugs for NSCLC patients harboring EGFR mutations, hepatotoxicity is a frequent adverse event resulting in treatment discontinuation. The purpose of this study was to investigate the correlation between severe hepatotoxicity rate and single nucleotide polymorphisms (SNPs) in enzymes related to the metabolism of EGFR-TKIs.

Methods

Among 650 NSCLC patients treated with EGFR-TKIs between 2009 and 2012 at our institution, 66 patients harboring EGFR mutations and whose blood samples before treatment were available were included. Multi-SNP analyses for representative 7 enzymes were performed using the DigiTag2 assay. Severe hepatotoxicity rate was defined as proportion of patients who had experienced ≥ Grade 3 transaminase/ total-bilirubin (T-bil) elevation or ≥ Grade 2 T-bil elevation with any grade of transaminase elevation.

Results

Among the 66 patients, 60 received gefitinib (Group A) and 27 received erlotinib (Group B). 21 were treated with both gefitinib and erlotinib at different times. Severe hepatotoxicity was detected in 19 (32%) from Group A and 6 (22%) from Group B. Poor metabolizer (PM) was defined as homozygous of variant allele (SNPs) which reduced metabolic enzyme activity. The phenotypes were as follows: CYP3A5, PM/ non-PM = 34/ 32; CYP2D6, PM/ non-PM = 5/ 61; and VKORC1, PM/ non-PM = 55/ 11. In Group A, severe hepatotoxicity rate was significantly higher among patients with PM for CYP3A5 (PM vs. non-PM: 48% vs. 14%; p < 0.01), PM for CYP2D6 (PM vs. non-PM: 80% vs. 27%; p = 0.04) and non-PM for VKORC1 (PM vs. non-PM: 26% vs. 67%; p = 0.02). In Group B, any types of SNPs were not correlated with severe hepatotoxicity rate. Among 8 patients whose treatments were changed from gefitinib to erlotinib because of severe hepatotoxicity at the initial EGFR-TKI treatment, 7 patients exhibited an improvement of hepatotoxicity.

Conclusions

The impact of SNPs on severe hepatotoxicity was different between gefitinib and erlotinib. The evaluation of SNPs in metabolic enzyme, such as CYP3A5, CYP2D6 and VKORC1 may be useful for predicting severe hepatotoxicity induced by gefitinib.

Disclosure

K. Yoh: I have a consultant or advisory relationship to disclosure during the past two years: AstraZeneca I have honoraria to disclose during the past two years: AstraZeneca, Chugai; S. Niho: I have honoraria to disclose during the past two years: AstraZeneca, Chugai; Y. Ohe: I have a research funding to disclose :AstraZeneca, Chugai I have honoraria to disclose during the past two years: AstraZeneca, Chugai; K. Goto: I have a consultant or advisory relationship to disclosure during the past two years: Chugai I have honoraria to disclose during the past two years: AstraZeneca, Chugai I have research funding to disclose: AstraZeneca, Chugai. All other authors have declared no conflicts of interest.