An epithelial-mesenchymal transition (EMT) gene signature predicts resistance to erlotinib and PI3K pathway inhibitors and identifies Axl as a novel EMT marker in non-small cell lung cancer

Epithelial/mesenchymal transition (EMT) is associated with loss of cell adhesion molecules, such as E-cadherin, and increased invasion, migration, and proliferation in epithelial cancers. In non-small cell lung cancer (NSCLC), EMT is associated with greater resistance to EGFR inhibitors. However, its potential to predict response to other targeted drugs or chemotherapy has not been well characterized. The goal of this study was to develop a robust, platform-independent EMT gene expression signature and to investigate the association of EMT and drug response in NSCLC. A 76-gene EMT signature was derived in 54 DNA-fingerprinted NSCLC cell lines and tested in an independent set of cell lines and in NSCLC patients from the BATTLE clinical trial. The signature classified cell lines as epithelial or mesenchymal independent of the microarray platform and correlated strongly with E-cadherin protein levels, as measured by reverse phase protein array. Higher protein expression of Rab25 (in epithelial lines) and Axl (in mesenchymal lines), two signature genes associated with in EMT in other cancer types, was also confirmed. Mesenchymal cell lines demonstrated significantly greater resistance to EGFR inhibition, independent of EGFR mutation status and were more resistant to drugs targeting the PI3K/Akt pathway. We observed no association between EMT and response to cytotoxic chemotherapies, including cisplatin, pemetrexed, and docetaxel monotherapy and/or doublets (p-values ≥0.2). In NSCLC patients, the EMT signature predicted 8-week disease control in the erlotinib arm, but not in other treatment arms. In conclusion, we have developed a robust EMT signature that predicts resistance to EGFR inhibitors and PI3K/Akt pathway inhibitors. Gene expression profiles were measured in 131 core biopsies from patients with refractory non-small cell lung cancer in the Biomarker-integrated Approaches of Targeted Therapy for Lung Cancer Elimination (BATTLE) trial. We used the BATTLE dataset to test an EMT gene expression signature trained in cell lines and independant of the microarray platform.

上皮间质转化（Epithelial/Mesenchymal Transition, EMT）与上皮源性癌症中细胞黏附分子（如E-钙粘蛋白（E-cadherin））的丢失、侵袭、迁移与增殖能力增强密切相关。在非小细胞肺癌（Non-Small Cell Lung Cancer, NSCLC）中，EMT与EGFR抑制剂（EGFR inhibitor）耐药性升高存在关联。然而，其在预测其他靶向药物或化疗应答方面的潜力尚未得到充分阐明。本研究旨在构建一种稳定且不受微阵列平台限制的EMT基因表达特征，并探究NSCLC中EMT与药物应答的相关性。本研究团队在54株经DNA指纹分型（DNA-fingerprinted）的NSCLC细胞系中构建了含76个基因的EMT特征模型，并在一组独立细胞系以及来自肺癌靶向治疗消除生物标志物整合方案（Biomarker-integrated Approaches of Targeted Therapy for Lung Cancer Elimination, BATTLE）临床试验的NSCLC患者队列中对该模型进行了验证。该特征模型可不受微阵列平台影响，将细胞系准确分为上皮型或间质型，且与通过反相蛋白阵列（reverse phase protein array）检测得到的E-钙粘蛋白蛋白水平呈显著相关。此外，研究证实了另外两个在其他癌种中与EMT相关的特征基因的蛋白表达差异：上皮型细胞系中Rab25蛋白表达升高，间质型细胞系中Axl蛋白表达升高。间质型细胞系对EGFR抑制剂的耐药性显著更强，且该特性不受EGFR突变状态影响；同时，间质型细胞系对PI3K/Akt通路靶向药物的耐药性也更高。本研究未观察到EMT与细胞毒性化疗（包括顺铂、培美曲塞、多西他赛单药及联合方案）应答之间存在关联（p值≥0.2）。在NSCLC患者中，EMT特征模型可预测厄洛替尼组患者的8周疾病控制情况，但无法预测其他治疗组的疾病控制情况。综上，本研究构建了一种稳定的EMT基因表达特征模型，该模型可预测EGFR抑制剂及PI3K/Akt通路抑制剂的耐药性。本研究针对肺癌靶向治疗消除生物标志物整合方案（BATTLE）临床试验中131例难治性非小细胞肺癌患者的核心活检样本进行了基因表达谱检测，并利用该BATTLE数据集，对在细胞系中构建且不受微阵列平台限制的EMT基因表达特征模型进行了验证。