Integrative genomic profiling of human prostate cancer

PMID: 20579941 · DOI: 10.1016/j.ccr.2010.05.026 · Journal: Cancer Cell (2010)

TL;DR

Integrative genomic profiling (DNA copy number, mRNA/miRNA expression, and exon resequencing) of 218 prostate cancers (181 primary, 37 metastases) identified NCOA2 as an oncogene amplified in ~11% of tumors that enhances AR transcriptional output, discovered a prostate-specific 3p14 deletion associated with TMPRSS2-ERG fusion implicating FOXP1, RYBP, and SHQ1 as cooperative tumor suppressors, and demonstrated that copy-number alteration patterns robustly stratify patients into prognostic subgroups beyond Gleason score. The dataset is publicly available as the prad_mskcc cohort on cBioPortal.

Cohort & data

• 218 prostate tumors: 181 primaries + 37 metastases, plus 12 cell lines/xenografts (PRAD)
• All samples had at least 70% tumor content; median follow-up ~5 years
• Platforms: Agilent 244K aCGH (agilent-244k), Affymetrix Human Exon 1.0 ST (mRNA), Agilent miRNA V2, Sanger exon resequencing of 138 genes in 80 tumors, iPLEX Sequenom for 22 oncogene hotspots in 156 tumors
• Dataset: prad_mskcc (GEO: GSE21032)

Key findings

• NCOA2 amplification at 8q13.3 in ~11% of tumors (1.9% primary, 24.3% metastases with focal/high-level amplification); NCOA2 overexpression correlated with amplification (p < 10^-16) and with increased AR transcriptional output (p < 0.0001)
• TMPRSS2-ERG fusion present in 52% of tumors; associated with a novel prostate-specific deletion at 3p14 implicating FOXP1, RYBP, and SHQ1 as cooperative tumor suppressors
• PI3K pathway altered in 42% of primaries and 100% of metastases (including PTEN, INPP4B, PIK3CA, PIK3R1)
• RAS/RAF pathway altered in 43% of primaries and 90% of metastases
• RB1 pathway altered in 34% of primaries and 74% of metastases
• AR pathway altered in 56% of primaries and 100% of metastases
• Unsupervised hierarchical clustering of CNA data identified 6 subgroups with significantly different biochemical recurrence rates (log-rank p-value significant); CNA clusters provided prognostic information independent of Gleason score
• Low somatic mutation rate (~0.31 mutations/Mb); TP53 and PTEN primarily lost through copy-number deletion rather than point mutation

Genes & alterations

Gene	Alteration	Finding
NCOA2	Amplification (8q13.3), somatic mutations	Oncogene in ~11% of tumors; enhances AR signaling
ERG	TMPRSS2-ERG fusion (21q22 deletion)	Present in 52% of tumors
TMPRSS2	Fusion with ERG	Androgen-driven gene fusion
PTEN	Deletion (10q23.31), rare mutation	Lost in ~21% of tumors; co-occurs with TMPRSS2-ERG
TP53	Deletion (17p13.1), rare mutation	Lost in ~24% of tumors
RB1	Deletion (13q14.2)	Common deletion in RB pathway
AR	Amplification, mutation, overexpression	Altered in 58% of metastases; exclusively in castration-resistant disease
MYC	Amplification (8q24.21)	Common amplified locus
FOXP1	Deletion (3p14)	Candidate tumor suppressor associated with TMPRSS2-ERG
RYBP	Deletion (3p14)	Candidate tumor suppressor
SHQ1	Deletion (3p14), somatic P22S mutation	Candidate tumor suppressor
NKX3-1	8p loss (but mRNA uncorrelated)	Common 8p deletion harbors NKX3-1
PIK3CA	Rare mutations (H1047R, E545K)	2 tumors with activating mutations
KRAS	Rare mutations (G12V, Q61H/L)	2 tumors
BRAF	Rare V600E mutation	1 tumor
IDH2	R172K mutation	1 tumor
MSH6	V250A mutation	Associated with mutator phenotype (11 somatic mutations)
NCOR2	Somatic mutations	3 tumors; AR pathway corepressor

Clinical implications

• CNA-based clustering stratifies prostate cancer patients into prognostic groups independent of Gleason score, suggesting potential utility for a CNA-based diagnostic test to guide treatment decisions (surgery vs. watchful waiting) in newly diagnosed patients
• The high frequency of PI3K pathway alterations (42% primary, 100% metastatic) provides strong rationale for clinical development of PI3K pathway inhibitors in prostate cancer
• NCOA2 amplification/overexpression in non-castrate primary tumors is associated with significantly higher rates of biochemical recurrence, implicating it as a prognostic biomarker
• AR pathway alterations extend beyond AR itself to coactivators like NCOA2 in primary tumors, suggesting AR signaling perturbation contributes to disease initiation, not only castration resistance

Limitations & open questions

• Exon resequencing was limited to 138 pre-selected genes; comprehensive whole-exome/whole-genome sequencing would better define the mutational landscape
• Stringent tumor selection criteria (70% tumor content) biased cohort toward larger, more aggressive tumors; genomic findings may not fully represent indolent prostate cancers
• The CNA-based prognostic clusters require validation in a larger independent cohort and from biopsy (rather than prostatectomy) material before clinical translation
• The functional roles of FOXP1, RYBP, and SHQ1 as tumor suppressors in prostate cancer remain to be validated experimentally
• Combined loss of 13q/18q and focal 5p amplification were associated with poor outcome, but specific driver genes in these regions were not identified

Citations from this paper used in the wiki

• “the nuclear receptor coactivator NCOA2 as an oncogene in ~11 percent of tumors” (Summary)
• “the androgen-driven TMPRSS2-ERG fusion was associated with a previously unrecognized, prostate-specific deletion at chromosome 3p14 that implicates FOXP1, RYBP and SHQ1 as potential cooperative tumor suppressors” (Summary)
• “DNA copy-number data from primary tumors revealed that copy-number alterations robustly define clusters of low- and high-risk disease beyond that achieved by Gleason score” (Summary)
• “PI3K pathway… altered in nearly half of primaries and all metastases examined” (Results, page 4)
• “AR pathway analysis… revealed alteration in 56 percent of primaries and 100 percent in metastases” (Results, page 4)
• “Increasing levels of NCOA2 shifted the DHT dose-response curve leftward and upward, indicating that NCOA2 can not only prime AR to respond to lower androgen concentrations but can also boost the total magnitude of AR transcriptional output” (Results, page 5)

This page was processed by paper-compiler on 2026-05-06.