Comparison of gene expression patterns across 12 tumor types identifies a cancer supercluster characterized by TP53 mutations and cell cycle defects

E. Martínez, K. Yoshihara, H. Kim, Gordon Mills, V. Treviño, R. G.W. Verhaak

Research output: Contribution to journalArticle

25 Citations (Scopus)

Abstract

Transcriptional profile-based subtypes of cancer are often viewed as identifying different diseases from the same tissue origin. Understanding the mechanisms driving the subtypes may be key in development of novel therapeutics but is challenged by lineage-specific expression signals. Using a t-test statistics approach, we compared gene expression subtypes across 12 tumor types, which identified eight transcriptional superclusters characterized by commonly activated disease pathways and similarities in gene expression. One of the largest superclusters was determined by the upregulation of a proliferation signature, significant enrichment in TP53 mutations, genomic loss of CDKN2A (p16 ARF), evidence of increased numbers of DNA double strand breaks and high expression of cyclin B1 protein. These correlations suggested that abrogation of the P53-mediated apoptosis response to DNA damage results in activation of cell cycle pathways and represents a common theme in cancer. A second consistent pattern, observed in 9 of 11 solid tumor types, was a subtype related to an activated tumor-associated stroma. The similarity in transcriptional footprints across cancers suggested that tumor subtypes are commonly unified by a limited number of molecular themes.

Original languageEnglish (US)
Pages (from-to)2732-2740
Number of pages9
JournalOncogene
Volume34
Issue number21
DOIs
StatePublished - May 21 2015
Externally publishedYes

Fingerprint

Cell Cycle
Gene Expression
Mutation
Neoplasms
Cyclin B1
Double-Stranded DNA Breaks
DNA Damage
Up-Regulation
Apoptosis
Proteins

ASJC Scopus subject areas

  • Molecular Biology
  • Genetics
  • Cancer Research

Cite this

Comparison of gene expression patterns across 12 tumor types identifies a cancer supercluster characterized by TP53 mutations and cell cycle defects. / Martínez, E.; Yoshihara, K.; Kim, H.; Mills, Gordon; Treviño, V.; Verhaak, R. G.W.

In: Oncogene, Vol. 34, No. 21, 21.05.2015, p. 2732-2740.

Research output: Contribution to journalArticle

Martínez, E. ; Yoshihara, K. ; Kim, H. ; Mills, Gordon ; Treviño, V. ; Verhaak, R. G.W. / Comparison of gene expression patterns across 12 tumor types identifies a cancer supercluster characterized by TP53 mutations and cell cycle defects. In: Oncogene. 2015 ; Vol. 34, No. 21. pp. 2732-2740.
@article{27bf62c193d64428b01cbf3320ae7064,
title = "Comparison of gene expression patterns across 12 tumor types identifies a cancer supercluster characterized by TP53 mutations and cell cycle defects",
abstract = "Transcriptional profile-based subtypes of cancer are often viewed as identifying different diseases from the same tissue origin. Understanding the mechanisms driving the subtypes may be key in development of novel therapeutics but is challenged by lineage-specific expression signals. Using a t-test statistics approach, we compared gene expression subtypes across 12 tumor types, which identified eight transcriptional superclusters characterized by commonly activated disease pathways and similarities in gene expression. One of the largest superclusters was determined by the upregulation of a proliferation signature, significant enrichment in TP53 mutations, genomic loss of CDKN2A (p16 ARF), evidence of increased numbers of DNA double strand breaks and high expression of cyclin B1 protein. These correlations suggested that abrogation of the P53-mediated apoptosis response to DNA damage results in activation of cell cycle pathways and represents a common theme in cancer. A second consistent pattern, observed in 9 of 11 solid tumor types, was a subtype related to an activated tumor-associated stroma. The similarity in transcriptional footprints across cancers suggested that tumor subtypes are commonly unified by a limited number of molecular themes.",
author = "E. Mart{\'i}nez and K. Yoshihara and H. Kim and Gordon Mills and V. Trevi{\~n}o and Verhaak, {R. G.W.}",
year = "2015",
month = "5",
day = "21",
doi = "10.1038/onc.2014.216",
language = "English (US)",
volume = "34",
pages = "2732--2740",
journal = "Oncogene",
issn = "0950-9232",
publisher = "Nature Publishing Group",
number = "21",

}

TY - JOUR

T1 - Comparison of gene expression patterns across 12 tumor types identifies a cancer supercluster characterized by TP53 mutations and cell cycle defects

AU - Martínez, E.

AU - Yoshihara, K.

AU - Kim, H.

AU - Mills, Gordon

AU - Treviño, V.

AU - Verhaak, R. G.W.

PY - 2015/5/21

Y1 - 2015/5/21

N2 - Transcriptional profile-based subtypes of cancer are often viewed as identifying different diseases from the same tissue origin. Understanding the mechanisms driving the subtypes may be key in development of novel therapeutics but is challenged by lineage-specific expression signals. Using a t-test statistics approach, we compared gene expression subtypes across 12 tumor types, which identified eight transcriptional superclusters characterized by commonly activated disease pathways and similarities in gene expression. One of the largest superclusters was determined by the upregulation of a proliferation signature, significant enrichment in TP53 mutations, genomic loss of CDKN2A (p16 ARF), evidence of increased numbers of DNA double strand breaks and high expression of cyclin B1 protein. These correlations suggested that abrogation of the P53-mediated apoptosis response to DNA damage results in activation of cell cycle pathways and represents a common theme in cancer. A second consistent pattern, observed in 9 of 11 solid tumor types, was a subtype related to an activated tumor-associated stroma. The similarity in transcriptional footprints across cancers suggested that tumor subtypes are commonly unified by a limited number of molecular themes.

AB - Transcriptional profile-based subtypes of cancer are often viewed as identifying different diseases from the same tissue origin. Understanding the mechanisms driving the subtypes may be key in development of novel therapeutics but is challenged by lineage-specific expression signals. Using a t-test statistics approach, we compared gene expression subtypes across 12 tumor types, which identified eight transcriptional superclusters characterized by commonly activated disease pathways and similarities in gene expression. One of the largest superclusters was determined by the upregulation of a proliferation signature, significant enrichment in TP53 mutations, genomic loss of CDKN2A (p16 ARF), evidence of increased numbers of DNA double strand breaks and high expression of cyclin B1 protein. These correlations suggested that abrogation of the P53-mediated apoptosis response to DNA damage results in activation of cell cycle pathways and represents a common theme in cancer. A second consistent pattern, observed in 9 of 11 solid tumor types, was a subtype related to an activated tumor-associated stroma. The similarity in transcriptional footprints across cancers suggested that tumor subtypes are commonly unified by a limited number of molecular themes.

UR - http://www.scopus.com/inward/record.url?scp=84930275553&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84930275553&partnerID=8YFLogxK

U2 - 10.1038/onc.2014.216

DO - 10.1038/onc.2014.216

M3 - Article

VL - 34

SP - 2732

EP - 2740

JO - Oncogene

JF - Oncogene

SN - 0950-9232

IS - 21

ER -