The clonal and mutational evolution spectrum of primary triple negative breast cancers

Authors

Shah SP

Roth A

Goya R

Oloumi A

Ha G

Zhao Y

Turashvili G

Ding J

Tse K

Haffari G

Bashashati A

Prentice LM

Khattra J

Burleigh A

Yap D

Bernard V

McPherson A

Shumansky K

Crisan A

Giuliany R

Heravi-Moussavi A

Rosner J

Lai D

Curtis C

Caldas C

Marra MA

Aparicio S

Doi

10.1038/nature10933

PMID: 22495314 · DOI: 10.1038/nature10933 · Journal: Nature (2012)

TL;DR

This study characterized the mutational and clonal evolution landscape of 104 primary triple-negative breast cancers (TNBC) using integrated whole-genome/exome sequencing, RNA-seq, and SNP array profiling. The authors found that TNBC exhibit a wide, continuous spectrum of genomic evolution at diagnosis – ranging from tumors with few somatic aberrations to those with hundreds of mutations across multiple pathways. Deep re-sequencing of 2,414 somatic SNVs revealed that clonal heterogeneity varies continuously across cases, with basal-subtype TNBC showing greater clonal diversity than non-basal TNBC. TP53 and PIK3CA/PTEN mutations tend to be clonally dominant (early events), while cytoskeletal/cell motility gene mutations appear at lower clonal frequencies (later events).

Cohort & data

  • 104 primary TNBC cases, treatment-naive at time of diagnosis
  • Affymetrix SNP6.0 copy number profiling: 104 cases
  • RNA-seq: 80 cases
  • Whole-genome or whole-exome sequencing: 65 cases
  • Deep re-sequencing validation (median >20,000x) of 2,414 somatic SNVs
  • Validation cohort: 159 additional breast cancers (82 ER+, 77 ER-) with targeted exon resequencing of 29 genes
  • Dataset: brca_bccrc
  • Data deposited at European Genome-phenome Archive (EGAS00001000132)

Key findings

  • Somatic mutation abundance varies continuously across TNBC; mutation count is unrelated to the proportion of genome altered by copy number changes or tumor cellularity.
  • Only ~36% of validated somatic SNVs are expressed in matched RNA-seq data, consistent with low-abundance alleles representing rarer subclones.
  • TP53 is the most frequently mutated gene: 62% of basal TNBC and 43% of non-basal TNBC cases harbor validated somatic mutations (PMID:22495314).
  • PIK3CA mutated in 10.2% (7/65), USH2A in 9.2% (6/65), MYO3A in 9.2% (6/65), PTEN and RB1 each in 7.7% (5/65) of cases.
  • Considering background mutation rates, TP53, PIK3CA, RB1, PTEN, MYO3A, and GH1 showed evidence of positive selection (q < 0.1).
  • SYNE1/SYNE2 mutations found in 9.2% (6/65) of cases.
  • ~20% of cases harbor potentially clinically actionable aberrations including BRAF V600E, high-level EGFR amplifications, and ERBB2/ERBB3 mutations.
  • Significantly overrepresented pathways (FDR < 0.001): TP53, chromatin remodeling, PI3K signaling, ERBB2 signaling, integrin/focal adhesion, WNT/cadherin signaling, ATM/Rb pathways.
  • Intragenic deletions in PRKN (PARK2) tumor suppressor found in 6% of cases, specifically linking PRKN with TNBC for the first time.
  • Retinoblastoma (Rb) binding sites are significantly overrepresented among non-coding regulatory mutations (31.9% vs expected 2.5%, Fisher exact test p = 2 x 10^-19).
  • Dirichlet process clustering (PyClone) of deeply sequenced SNVs revealed a wide spectrum of clonal frequencies: some tumors have 1-2 clonal frequency modes while others have many, indicating extensive clonal evolution.
  • Basal-subtype TNBC exhibit more clonal frequency modes than non-basal TNBC.
  • TP53 and PIK3CA pathway mutations skew toward higher clonal frequencies (Wilcoxon q < 0.01), consistent with early driver status; median clonal frequency for “p53 pathway feedback loops” was 73% (q = 0.0007).
  • Cytoskeletal gene mutations (integrins, laminins, collagens, myosins) occur at lower clonal frequencies; “Integrin cell surface interactions” pathway had median clonal frequency of 42% (q = 0.9569).
  • driverNet analysis identified novel candidate driver genes including PRPS2 (homozygous deletions in 3 cases), NR3C1 (SNVs in 3 cases), and multiple PKC family genes (PRKCZ, PRKCQ, PRKG1, PRKCE).

Genes & alterations

Clinical implications

  • ~20% of TNBC cases harbor potentially clinically actionable somatic aberrations, including BRAF V600E mutations, high-level EGFR amplifications, and ERBB2/ERBB3 mutations, suggesting therapeutic opportunities beyond standard chemotherapy.
  • The wide variation in clonal composition at diagnosis implies that future therapeutic strategies for TNBC will need to account for individual tumor clonal genotypes rather than treating TNBC as a single disease entity.
  • The continuous spectrum of clonal evolution suggests that “low clonality” versus “high clonality” tumors may require fundamentally different biological and therapeutic approaches.

Limitations & open questions

  • Clonal frequency estimation relies on computational modeling (PyClone Dirichlet process) applied to bulk sequencing data; single-cell validation was not performed.
  • The validation cohort (159 additional breast cancers) used targeted exon resequencing of only 29 genes, limiting the scope of confirmation.
  • Whether the cytoskeletal/cell motility mutation pattern represents true disease-modifying driver mutations or transcription-related hypermutation is unknown.
  • ~12% of cases contained no somatic aberrations in any frequent driver or cytoskeletal gene, suggesting additional undiscovered drivers.
  • The functional significance of non-coding regulatory mutations (particularly in Rb binding sites) requires experimental validation.
  • The study is limited to treatment-naive primary tumors; how clonal architecture evolves under therapy was not addressed.

Citations from this paper used in the wiki

  • “at the time of diagnosis these cancers exhibit a wide and continuous spectrum of genomic evolution, with some exhibiting only a handful of somatic aberrations in a few pathways, whereas others contain hundreds of somatic events and multiple pathways implicated” (Abstract)
  • “p53 is the most frequently mutated gene with 62% of basal TNBC and 43% of non-basal TNBC cases harbouring a validated somatic mutation” (p. 3)
  • “Approximately 20% of cases contained examples of potentially ‘clinically actionable’ somatic aberrations, including BRAF V600E, high level EGFR amplifications and ERBB2/ERBB3 mutations” (p. 4)
  • “the median clonal frequency for Reactome pathway ‘p53 pathway feedback loops’ including 46 mutations in ATM, ATR, NRAS, PIK3CA, PTEN, SIAH1, and TP53 was 73% (Wilcoxon, q=0.0007) whereas ‘Integrin cell surface interactions’ including 23 mutations in integrin, laminin and collagen genes had a median clonal frequency of 42% (Wilcoxon, q=0.9569)” (p. 5)
  • “Here we report intragenic deletions in the PARK2 tumour suppressor, specifically linking PARK2 with TNBC for the first time” (p. 3)
  • “basal TNBC have more clonal frequency modes than non-basal TNBC” (p. 5)

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