Two classes of intrahepatic cholangiocarcinoma defined by relative abundance of mutations and copy number alterations
Metrics: PDF 1708 views | HTML 1548 views | ?
Young-Ho Kim1, Eun-Kyung Hong2, Sun-Young Kong1,3,4, Sung-Sik Han2, Seoung-Hoon Kim2, Je-Keun Rhee5, Soo-Kyung Hwang1, Sang-Jae Park2, Tae-Min Kim5,6
1Translational Epidemiology Research Branch, Research Institute and Hospital, National Cancer Center, Goyang, Republic of Korea
2Center for Liver Cancer and Hospital, National Cancer Center, Goyang, Republic of Korea
3Department of Laboratory Medicine, Diagnostic Oncology Center and Hospital, National Cancer Center, Goyang, Republic of Korea
4Department of System Cancer Science, Graduate School of Cancer Science and Policy, National Cancer Center, Goyang, Republic of Korea
5Department of Medical Informatics, The Catholic University of Korea, Seoul, Republic of Korea
6Cancer Research Institute, College of Medicine, The Catholic University of Korea, Seoul, Republic of Korea
Tae-Min Kim, e-mail: firstname.lastname@example.org
Sang-Jae Park, e-mail: email@example.com
Keywords: cholangiocarcinoma, exome sequencing, transcriptome, somatic mutations
Received: August 24, 2015 Accepted: March 02, 2016 Published: March 18, 2016
Intrahepatic cholangiocarcinoma (ICC) is a biliary tree-origin epithelial malignancy in liver with unfavorable clinical outcomes. Systematic genome analyses may advance our understanding of ICC pathogenesis also improving current diagnostic and therapeutic modalities. In this study, we analyzed 17 ICC tumor-vs-matched normal pairs using either whole-exome (n = 7), transcriptome sequencing (n = 7) or both platforms (n = 3). For somatic mutations, we identified recurrent mutations of previously reported genes such as KRAS, TP53, APC as well as epigenetic regulators and those of TGFβ signaling pathway. According to the abundance of somatic mutations and DNA copy number alterations (CNA), ten ICC exome cases were distinguished into two classes as those primarily driven by either somatic mutations (M class) or CNAs (C class). Compared to M class ICCs (92–147 somatic mutations; n = 5) with a relative deficit of CNAs, C class ICCs (54–84 mutations; n = 5) harbor recurrent focal CNAs including deletions involving CDKN2A, ROBO1, ROBO2, RUNX3, and SMAD4. We also show that transcriptome sequencing can be used for expression-based ICC categorization but the somatic mutation calling from the transcriptome can be heavily influenced by the gene expression level and potentially, by posttranscriptional modification such as nonsense mediated decay. Along with a substantial level of mutational heterogeneity of ICC genomes, our study reveals previously unrecognized two ICC classes defined by relative abundance of somatic mutations over CNAs or vice versa, which should be considered in the selection of genotyping platforms and sensitive screening of targets for ICC therapeutics.
All site content, except where otherwise noted, is licensed under a Creative Commons Attribution 3.0 License.