Prognostic subclass of intrahepatic cholangiocarcinoma by integrative molecular–clinical analysis and potential targeted approach
- 174 Downloads
Although molecular characterization of iCCA has been studied recently, integrative analysis of molecular and clinical characterization has not been fully established. If molecular features of iCCA can be predicted based on clinical findings, we can approach to distinguish targeted treatment. We analyzed RNA sequencing data annotated with clinicopathologic data to clarify molecular-specific clinical features and to evaluate potential therapies for molecular subtypes.
We performed next-generation RNA sequencing of 30 surgically resected iCCA from Korean patients and the clinicopathologic features were analyzed. The RNA sequences from 32 iCCA resected from US patients were used for validation.
Patients were grouped into two subclasses on the basis of unsupervised clustering, which showed a difference in 5-year survival rates (48.5% vs 14.2%, p = 0.007) and similar survival outcome in the US samples. In subclass B (poor prognosis), both data sets were similar in higher carcinoembryonic antigen and cancer antigen 19-9 levels, underlying cholangitis, and bile duct-type pathology; in subclass A (better prognosis), there was more frequent viral hepatitis and cholangiolar-type pathology. On pathway analysis, subclass A had enriched liver-related signatures. Subclass B had enriched inflammation-related and TP53 pathways, with more frequent KRAS mutations. CCA cell lines with similar gene expression patterns of subclass A were sensitive to gemcitabine.
Two molecular subtypes of iCCA with distinct clinicopathological differences were identified. Knowledge of clinical and pathologic characteristics can predict molecular subtypes, and knowledge of different subtype signaling pathways may lead to more rational, targeted approaches to treatment.
KeywordsCholangiocarcinoma RNA sequence Gene expression Pathway Mutation KRAS Target therapy
The authors thank Li Li, Dehai Wu, Katsuyuki Miyabe, Tao Song, Ning Zhang, Jianbo Huang, Shaoqing Wang, Lin Yang, and Amy S Mauer (Mayo clinic, Rochester, MN) for their help in the experiment. The biospecimens and data used for this study were provided by the Biobank of Keimyung University Dongsan Medical Center, member of the Korea Biobank Network. This work was supported by the National Research Foundation of Korea (NRF) Grant funded by the Korea government (MSIP) (No. 2018R1C1B3004435). This work was supported by the National Research Foundation of Korea (NRF) Grant funded by the Korea Government (MSIP) (No. 2014R1A5A2010008).
Compliance with ethical standards
Conflict of interest
Keun Soo Ahn, Daniel O’Brien, Yu Na Kang, Taofic Mounajjed4, Yong Hoon Kim, Tae-Seok Kim, Jean-Pierre A. Kocher, Loretta K. Allotey, Mitesh J. Borad, Lewis R. Roberts, and Koo Jeong Kang declare that they have no conflict of interest.
Availability of data and materials
RNA sequencing data of this manuscript is registered at Gene Expression Omnibus (http://www.ncbi.nlm.nih.gov/geo) with GEO accession number of GSE107943
This study was approved by institutional review boards at both institutions (Keimyung University Dongsan Medical Center: IRB No. 2014-12-066, Mayo clinic: IRB No. 16-007369) and included secondary use of human-derived materials.
Informed consent for the human-derived materials was obtained from all patients before surgery.