Original ResearchFull Report: Clinical—Alimentary TractIdentification of Molecular Subtypes of Gastric Cancer With Different Responses to PI3-Kinase Inhibitors and 5-Fluorouracil
Section snippets
Overall Design of This Study
Figure 1 presents the overall design of this study.
Patients and Tumors
Singaporean patients were recruited from the National Cancer Centre and hospitals of the National Healthcare Group. Australian patients were recruited from the Peter MacCallum Cancer Centre in Melbourne. All patients provided written informed consent, and all tissue samples were collected with approval from the respective ethics committees. Tumors were macrodissected. Supplementary Table 1 summarizes the information associated with the patients
Three Subtypes of Gastric Cancer
We assembled a collection of 248 gene-expression profiles by combining our previously reported data12 with new data from 56 gastric adenocarcinomas. We dealt with the issue of batch effects using the ComBat algorithm13 as described in the Supplementary Materials and Methods, and we then used CHC_IFS to discover intrinsic subtypes among these tumors (Materials and Methods). This CHC_IFS approach addressed 2 important issues in unsupervised clustering. One issue is avoidance of the creation of
Discussion
By using CHC_IFS, we found 3 well-defined subtypes of gastric adenocarcinoma: mesenchymal, proliferative, and metabolic. We developed a predictor, GC-class, for these subtypes and showed that the 3-subtype classification could be applied to an additional set of 70 gastric tumors. Table 3 summarizes the biological and clinical characteristics of each of the 3 subtypes. Notably, we found that patients with metabolic-subtype tumors benefited preferentially from 5-FU treatment (Figure 4,
Acknowledgements
All microarray data are available at the NCBI Gene Expression Omnibus repository (http://www.ncbi.nlm.nih.gov/projects/geo/): (1) gene expression, Singapore batch A: GSE15459; (2) gene expression, Singapore batch B: GSE34942; (3) gene expression, Australian cohort: GSE35809; (4) gene expression, gastric cancer cell lines: GSE22183; (5) DNA copy number, Singapore cohort: GSE31168; (6) DNA methylation, Singapore cohort: GSE30601; and (7) gene expression reported by Cho et al44: GSE13861.
References (52)
- et al.
Genetic and epigenetic changes in stomach cancer
Int Rev Cytol
(2001) - et al.
Intrinsic subtypes of gastric cancer, based on gene expression pattern, predict survival and respond differently to chemotherapy
Gastroenterology
(2011) - et al.
REporting recommendations for tumour MARKer prognostic studies (REMARK)
Eur J Cancer
(2005) - et al.
Integrated genomic analysis identifies clinically relevant subtypes of glioblastoma characterized by abnormalities in PDGFRA, IDH1, EGFR, and NF1
Cancer Cell
(2010) - et al.
Molecular definition of breast tumor heterogeneity
Cancer Cell
(2007) - et al.
Gene expression profiling of metaplastic lineages identifies CDH17 as a prognostic marker in early stage gastric cancer
Gastroenterology
(2010) - et al.
Global DNA demethylation in gastrointestinal cancer is age dependent and precedes genomic damage
Cancer Cell
(2006) - et al.
DNA hypomethylation and human diseases
Biochim Biophys Acta
(2007) - et al.
Let-7 regulates self renewal and tumorigenicity of breast cancer cells
Cell
(2007) - et al.
Phosphoinositide 3-kinase signalling–which way to target?
Trends Pharmacol Sci
(2003)
Gastric adenocarcinoma: review and considerations for future directions
Ann Surg
Cancer statistics
CA Cancer J Clin
A combined comparative genomic hybridization and expression microarray analysis of gastric cancer reveals novel molecular subtypes
Cancer Res
Expression profiling and subtype-specific expression of stomach cancer
Cancer Res
Gastric cancer
Consensus clustering: a resampling-based method for class discovery and visualization of gene expression microarray data
Machine Learning
Linear models and empirical bayes methods for assessing differential expression in microarray experiments
Stat Appl Genet Mol Biol
Nearest template prediction: a single-sample-based flexible class prediction with confidence assessment
PLoS One
Oncogenic pathway combinations predict clinical prognosis in gastric cancer
PLoS Genet
Adjusting batch effects in microarray expression data using empirical Bayes methods
Biostatistics
ConsensusClusterPlus: a class discovery tool with confidence assessments and item tracking
Bioinformatics
Statistical significance of clustering for high-dimension, low-sample size data
J Am Stat Assoc
DAVID: Database for Annotation, Visualization, and Integrated Discovery
Genome Biol
Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources
Nat Protoc
High-dimensional sparse factor modeling: applications in gene expression genomics
J Am Stat Assoc
Gene expression profiling of human prostate cancer stem cells reveals a pro-inflammatory phenotype and the importance of extracellular matrix interactions
Genome Biol
Cited by (0)
Author names in bold designate shared co-first authorship.
Conflicts of interest The authors disclose the following: As mandated by our funders, a patent application covering the gene-expression–based classification of gastric adenocarcinoma has been filed by Exploit Technologies Pte, Ltd, a technology transfer arm of the Agency for Science, Technology and Research, Singapore.
Funding Supported by the Duke-NUS Signature Research Programs, funded by the Singapore Agency for Science, Technology, and Research, the Singapore Ministry of Health, the Singapore National Medical Research Council, a Translational Clinical Research grant (NMRC/TCR/001/2007), by the Singapore National Research Foundation and the Ministry of Education (grant CL2008-07), and by the Singapore Biomedical Research Council (grant 10/1/24/19/665).