Skip to main content
. 2020 May 13;9:e51480. doi: 10.7554/eLife.51480

Figure 2. Copy number heterogeneity impacts a significant proportion of breast cancer genomes and occurs at regions in the genome that are biologically and clinically important.

(A) Schematic illustration of CNA heterogeneity in tumor P5. Single-cell genomes (n = 131) are plotted on the same copy number diagram. Heterogeneous/sub-clonal regions can be identified by the presence of multiple copy number states (ex: 1q and chromosome 13). Red arrows point to representative clonal alterations. Blue arrows point to representative sub-clonal alterations. (B) Quantification of copy number heterogeneity as fraction (%) of the genome found to be sub-clonal across all sequenced biopsies. (C–H) Representative chromosome wide views of identified sub-clonal CNAs affecting: regions recurrently altered in breast cancer genomes by CNAs (C), regions containing genes of therapeutic relevance (D), regions containing genes known to be affected by somatic SNVs (E), regions with experimental evidence of involvement in breast cancer metastasis (F), regions found at three or more different copy number states (G). Gray vertical bars denote the location of genes or CNAs. (H) Distant relapse-free survival curves for cases with 1q and 8q gains that are ER+ and HER2- (n = 428). Cases are stratified based on their level of 1q or 8q gain (low vs. high).

Figure 2—source data 1. Inferred integer copy number values of all single cancer nuclei sequenced for all patients.

Figure 2.

Figure 2—figure supplement 1. Copy number heterogeneity variably affects the genomes of breast cancers.

Figure 2—figure supplement 1.

Schematic illustration of representative tumors with varying levels of copy number heterogeneity found in the analyzed tumor cohort: significant (A), moderate (B), and low (C).
Figure 2—figure supplement 2. Copy number heterogeneity occurs in regions of the genome that are important and harbor biologically relevant breast cancer genes.

Figure 2—figure supplement 2.

(A) Chromosome zoom-in-views of regions affected by copy number heterogeneity that are important in breast cancer genomes. Categories are: regions recurrently altered by CNA, regions harboring therapeutically relevant genes, regions harboring genes known to be recurrently altered by SNV, and regions harboring genes for which experimental evidence exists for importance in the metastatic process. (B) DNA-FISH validation of a selected subset of heterogeneous CNAs. The selected alterations are: Chromosome 16q and 16 p sub-clonality seen in HER2+ tumor P14 (left panel) and PIK3CA sub-clonality in a LumA tumor, P5 (right panel). (C) Distant relapse-free survival curves for all METABRIC cases with 1q/8q gains(n = 446). Within the cohort that was ER+/HER2-, survival curves are also shown for patients that are lymph node positive (LN+, n = 192) and lymph node negative (LN-, n = 236). (D) Contingency table summarizing the relationship between high- vs. low-level 1q/8q amplification and IntClust (IC) subtype1 for all METABRIC cases with 1q/8q amplification. Among high risk ER+/HER2- groups (IC 1,2,6,9)2, IC9 has the highest rate of high level 1q/8q amplifications. A t-test for proportions between IC 9 (high risk) vs IC 3 (lower risk) shows that high-level 1q/8q amplifications are statistically significantly more likely to occur in IC9 (p-value=3.021e-07).