Premium
A genomic and clinicopathological study of non‐small‐cell lung cancers with discordant ROS 1 gene status by fluorescence in‐situ hybridisation and immunohistochemical analysis
Author(s) -
Zhao Jing,
Chen Xiaotong,
Zheng Jing,
Kong Mei,
Wang Bo,
Ding Wei
Publication year - 2018
Publication title -
histopathology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.626
H-Index - 124
eISSN - 1365-2559
pISSN - 0309-0167
DOI - 10.1111/his.13492
Subject(s) - kras , immunohistochemistry , biology , lung cancer , pathology , fluorescence in situ hybridization , cancer research , ros1 , carcinogenesis , adenocarcinoma , cancer , medicine , gene , colorectal cancer , genetics , chromosome
Aims ROS 1 immunohistochemistry ( IHC ) using D4D6 antibody is a useful tool for screening patients with non‐small‐cell lung cancer ( NSCLC ) who may be suitable for targeted therapy. Many studies and our data have identified cases that express the ROS 1 protein strongly but are negative for ROS 1 by fluorescence in‐situ hybridisation ( FISH ). The present study investigated the driver mutation and clinicopathological characteristics of 26 discordant cases ( ROS 1 IHC ‐positive but FISH ‐negative) to find new clues for distinguishing real ROS 1 ‐rearranged cases. Methods and results Tumours from 26 discordant cases were analysed for clinicopathological characteristics, mutations in EGFR , KRAS , ERBB 2 , BRAF and PIK 3 CA ; fusions in ALK and RET ; and amplifications in MET , ERBB 2 and ROS 1 . ROS 1‐rearranged NSCLC s were significantly more likely to be found in younger patients and at an advanced stage; they showed cribriform features, extracellular mucus and psammoma bodies, whereas ROS 1‐discordant cases were found in older patients at a relatively early tumour–node–metastasis ( TNM ) stage and showed a lepidic growth pattern (all P < 0.001). Most ROS 1 ‐rearranged NSCLC s had no concurrent mutation, whereas 73% of discordant cases harboured genetic aberrations, including EGFR and ERBB 2 . Compared with general lung adenocarcinomas, ERBB ‐2 abnormality was disproportionately high in ROS 1‐discordant cases. Moreover, we optimised the scoring criteria for ROS 1 IHC as ‘H score > 150 and no concurrent mutations’; the specificity was then increased to 81.6%. Conclusions Compared with ROS 1 ‐rearranged cases, ROS 1‐discordant patients showed distinct clinical and morphological features and often harboured another oncogenic driver alteration. The use of optimised screening criteria will increase the specificity of ROS 1 antibody.