Multiple region whole-exome sequencing reveals dramatically evolving intratumor genomic heterogeneity in esophageal squamous cell carcinoma

• Published in: Oncogenesis (New York, NY) Vol. 4; no. 11; p. e175
• Main Authors: Cao, W, Wu, W, Yan, M, Tian, F, Ma, C, Zhang, Q, Li, X, Han, P, Liu, Z, Gu, J, Biddle, F G
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 30.11.2015; Nature Publishing Group
• Subjects: 38/23; 45; 631/208/191; 631/67/2329; Apoptosis; Cell Biology; Human Genetics; Internal Medicine; Medicine; Medicine & Public Health; Oncology; Original; original-article
• ISSN: 2157-9024; 2157-9024
• DOI: 10.1038/oncsis.2015.34

Cancer is a disease of genome instability and genomic alterations; now, genomic heterogeneity is rapidly emerging as a defining feature of cancer, both within and between tumors. Motivation for our pilot study of tumor heterogeneity in esophageal squamous cell carcinoma (ESCC) is that it is not well studied, but the highest incidences of esophageal cancers are found in China and ESCC is the most common type. We profiled the mutations and changes in copy number that were identified by whole-exome sequencing and array-based comparative genomic hybridization in multiple regions within an ESCC from two patients. The average mutational heterogeneity rate was 90% in all regions of the individual tumors in each patient; most somatic point mutations were nonsynonymous substitutions, small Indels occurred in untranslated regions of genes, and copy number alterations varied among multiple regions of a tumor. Independent Sanger sequencing technology confirmed selected gene mutations with more than 88% concordance. Phylogenetic analysis of the somatic mutation frequency demonstrated that multiple, genomically heterogeneous divergent clones evolve and co-exist within a primary ESCC and metastatic subclones result from the dispersal and adaptation of an initially non-metastatic parental clone. Therefore, a single-region sampling will not reflect the evolving architecture of a genomically heterogeneous landscape of mutations in ESCC tumors and the divergent complexity of this genomic heterogeneity among patients will complicate any promise of a simple genetic or epigenetic diagnostic signature in ESCC. We conclude that any potential for informative biomarker discovery in ESCC and targeted personalized therapies will require a deeper understanding of the functional biology of the ontogeny and phylogeny of the tumor heterogeneity.