An integrated genomics approach identifies drivers of proliferation in luminal-subtype human breast cancer

Journal name:
Nature Genetics
Volume:
46,
Pages:
1051–1059
Year published:
DOI:
doi:10.1038/ng.3073
Received
Accepted
Published online

Abstract

Elucidating the molecular drivers of human breast cancers requires a strategy that is capable of integrating multiple forms of data and an ability to interpret the functional consequences of a given genetic aberration. Here we present an integrated genomic strategy based on the use of gene expression signatures of oncogenic pathway activity (n = 52) as a framework to analyze DNA copy number alterations in combination with data from a genome-wide RNA-mediated interference screen. We identify specific DNA amplifications and essential genes within these amplicons representing key genetic drivers, including known and new regulators of oncogenesis. The genes identified include eight that are essential for cell proliferation (FGD5, METTL6, CPT1A, DTX3, MRPS23, EIF2S2, EIF6 and SLC2A10) and are uniquely amplified in patients with highly proliferative luminal breast tumors, a clinical subset of patients for which few therapeutic options are effective. This general strategy has the potential to identify therapeutic targets within amplicons through an integrated use of genomic data sets.

At a glance

Figures

  1. Patterns of genomic signature pathway activity in breast cancer.
    Figure 1: Patterns of genomic signature pathway activity in breast cancer.

    (a) Patterns of pathway activity (n = 52) were determined for each sample in the published TCGA breast cancer cohort (n = 476). Expression signature scores (y axis) are median centered and clustered by complete linkage hierarchical clustering. (b) ANOVA (P < 0.0001) for all signatures according to PAM50 subtype followed by Tukey's test for pairwise comparison demonstrates statistically significant differences in the levels of pathway expression between molecular subtypes. Box colors indicate the level of significance between subtypes, as indicated in the legend. NS, not significant.

  2. Identification of genomic pathway-specific CNAs.
    Figure 2: Identification of genomic pathway–specific CNAs.

    (a) Schematic outlining the strategy used to identify CNAs associated with pathway activity. Gain/loss indicates gains or losses; Pos/Neg indicates positive or negative. (b) For each signature, significant copy number gains and losses were calculated. The plot identifies those genes that had a positive Spearman rank correlation and increased amplification frequency (q < 0.01) (red) and those that had a negative Spearman rank correlation and an increased frequency of copy number losses in the top-scoring (top quartile) samples with pathway activity (q < 0.01) (blue). (ce) Spearman rank correlation was used to identify genes positively (black line) or negatively (dark blue) associated with pathway activity, and Fisher's exact test was used to compare the frequency of copy number gains (Amp, red) or losses (Del, light blue) for the HER2-AMP (c), HER1-C2 (d) and RB-LOH (e) signatures. Yellow arrowheads indicate known pathway drivers with q < 0.01 for each analysis; the black arrowhead indicates q < 0.01 for a single analysis. In each figure, chromosomal boundaries are indicated by vertical black lines.

  3. Identification of DNA CNAs in highly proliferative breast tumors.
    Figure 3: Identification of DNA CNAs in highly proliferative breast tumors.

    (a,b) Distribution of proliferation scores across all tumors (a) and by subtype (b). (b) Box and whisker plots indicate the median score (horizontal line), the interquartile range (IQR, box boundaries) and 1.5 times the IQR (whiskers). Basal-like (n = 88), HER2E (n = 55), LumA (n = 214) and LumB (n = 119). (c) Highly proliferative tumors (top quartile) are comprised of basal-like (49.6%), LumB (33.6%) and HER2E (16.8%) samples. (d) Highly proliferative luminal tumors are restricted to LumB (68.0%) and HER2E (32.0%) samples. (e) Frequency of CNAs in highly proliferative (black line) and all other (gray line) samples. (f) Statistical analyses of CNAs. Indicated are positive correlations (black) and negative correlations (dark blue) by Spearman rank and frequency of amplifications (red) and deletion (light blue) by Fisher's exact test. (g) Frequency of CNAs in highly proliferative luminal tumors; the color key used is the same as that in e. (h) Statistical analyses of CNAs in proliferative luminal tumors; the color key used is the same as that in f. Chromosomal boundaries in eh are defined by vertical black lines.

  4. Identification of genomic pathway-associated essential genes in cell lines.
    Figure 4: Identification of genomic pathway–associated essential genes in cell lines.

    (a) Schematic outlining the strategy used to identify pathway-specific genetic dependencies. (b) A panel of 27 breast cancer cell lines with both expression data and data from a genome-wide RNAi screen was used to identify pathway-specific genes that are required for cell viability using a negative Spearman rank correlation (with −log10 P values plotted); significant genes (P < 0.05) are shown according to chromosome location. Vertical black lines indicate chromosomal boundaries. (ce) ESR1 (c), ERBB2 (d) and STAT1 or JAK3 (e) shRNA levels are inversely associated with the ER, HER2 and STAT1 pathway scores, respectively.

  5. Identification of essential genes amplified in highly proliferative luminal tumors.
    Figure 5: Identification of essential genes amplified in highly proliferative luminal tumors.

    (a) Schematic outlining the integrated genomic strategy used to identify essential genes amplified in highly proliferative luminal breast tumors. (b) Identification of 21 genes in amplified loci that are unique to highly proliferative luminal tumors and are required specifically for luminal cell line proliferation in vitro. mRNA expression of the genes in red and blue was significantly associated with CNA status, with the subset highlighted in red being further validated in the METABRIC data set; genes in black did not show a significant mRNA-DNA correlation. Candidate genes demarcated by asterisks are located at the cusp of a CNA segment and were originally excluded but are mentioned here. Genes marked with # were not included on the mRNA expression microarrays, and the correlation between DNA and mRNA expression was not assessed.

  6. Candidate gene amplification correlates with a poor prognosis.
    Figure 6: Candidate gene amplification correlates with a poor prognosis.

    (ae) Amplification of FGD5 (nAmp = 51, nNo Amp = 337; a), METTL6 (nAmp = 51, nNo Amp = 337; b), DTX3 (nAmp = 71, nNo Amp = 317; c) and MRSP23 (nAmp = 127, nNo Amp = 261; d) correlated with poor disease-specific outcome in patients with luminal breast cancer in the TCGA data set (n = 388), whereas CPT1A (nAmp = 111, nNo Amp = 277; e) amplification had no effect on prognosis. (fj) Consistent results were observed in the METABRIC data set (n = 1,333) for FGD5 (nAmp = 42, nNo Amp = 1,218; f), METTL6 (nAmp = 44, nNo Amp = 1,278; g), DTX3 (nAmp = 67, nNo Amp = 1,266; h), MRPS23 (nAmp = 266, nNo Amp = 1,062; i) and CPT1A (nAmp = 241, nNo Amp = 1,029; j). Samples in the METABRIC data set missing CNA calls were excluded. For each analysis, the P value was determined by log-rank test, and hazard ratios (HR) are reported.

  7. Patterns of pathway activity correspond with molecular subtypes of breast cancer.
    Supplementary Fig. 1: Patterns of pathway activity correspond with molecular subtypes of breast cancer.

    Analysis of molecular subtypes of breast cancer based on 52 gene expression signature scores. Euclidean distance was used to calculate the relationship between samples based on scores of 52 gene express signatures. Samples are commonly ordered on the X and Y axis according to molecular subtype. These results demonstrate high concordance within a subtype (dark blue), and lower concordance across subtypes; each sample versus itself is the blue diagonal line.

  8. Correlation between calculated pathway activity.
    Supplementary Fig. 2: Correlation between calculated pathway activity.

    A Pearson correlation matrix of each signature versus all other signatures (including itself as the diagonal line) demonstrates a high degree of concordance amongst independently developed gene expression signatures measuring similar or associated pathways. Red indicates high positive correlation and blue a strong anti-correlation.

  9. Identification of pathway-specific copy number alterations by Spearman rank correlation.
    Supplementary Fig. 3: Identification of pathway-specific copy number alterations by Spearman rank correlation.

    A Spearman rank correlation, both positive (red) and negative (blue) were used to identify associations between predicted genomic signature pathway activity and gene-level DNA copy number content (n=476). The negative log10 Bonferroni adjust p-values are plotted according to chromosomal position. Chromosomal borders are delineated by vertical black lines.

  10. Identification of pathway-specific copy number alterations based on frequency of gains or losses calculated by Fisher/'s exact test.
    Supplementary Fig. 4: Identification of pathway-specific copy number alterations based on frequency of gains or losses calculated by Fisher’s exact test.

    A Fisher’s exact test was used to calculate the statistical significance of the frequency of copy number gains (red) or losses (blue) in samples with the highest (top quartile) pathway signature activity relative to all other samples (n=476). The negative log10 Bonferroni adjust p-values are plotted according to chromosomal position. Vertical black lines indicate chromosomal borders.

  11. Patterns of pathway activity in human breast cancer cell lines.
    Supplementary Fig. 5: Patterns of pathway activity in human breast cancer cell lines.

    The scored pathway activity for a panel of 51 breast cancer cell lines (GSE12777) was calculated for the 52 pathway signatures. Of these cell lines, 27 which are denoted by black squares in lower panel were subjected to a genome-wide RNAi screen.

  12. Identification of essential genes in proliferative breast cancer cell lines.
    Supplementary Fig. 6: Identification of essential genes in proliferative breast cancer cell lines.

    Identification of genes essential for cell viability in vitro in the context of the 11-gene PAM50 Proliferation signature in (A) all cell line samples and (B) in luminal and HER2+ breast cancer cell lines. The negative log10 Spearman rank correlation p values are plotted for each gene relative to chromosomal position.

  13. Correlation between candidate gene mRNA expression and DNA copy number status in TCGA samples.
    Supplementary Fig. 7: Correlation between candidate gene mRNA expression and DNA copy number status in TCGA samples.

    The mRNA expression levels of the 21 identified candidate genes that are required for cell viability and are uniquely amplified in highly proliferative luminal breast tumors. In each plot, the mRNA levels from the TCGA data are compared in those tumors with amplifications versus all others. Of the 21 genes, two (ANKRD56 and TMEM189) were not present on the mRNA expression array and are not included here. Of the remaining 19 genes, 12 had a significant relationship (p<0.05) between copy number status and mRNA expression levels.

  14. Correlation between candidate gene mRNA expression and DNA copy number status in METABRIC samples.
    Supplementary Fig. 8: Correlation between candidate gene mRNA expression and DNA copy number status in METABRIC samples.

    The mRNA expression levels of the 12 identified candidate genes that are required for cell viability and are uniquely amplified in highly proliferative luminal breast tumors were analyzed within the context of copy number level in the METABRIC dataset. Of these 12 genes, three (SNX21, ZBTB46 and DNAJC5) were not present on both of the METABRIC data platforms (mRNA expression and copy number). Of the remaining 9 genes, all had a significant relationship (p<0.05) between copy number status and mRNA expression levels.

  15. Validation of increased candidate gene copy number status in highly proliferative luminal breast tumors in METABRIC samples.
    Supplementary Fig. 9: Validation of increased candidate gene copy number status in highly proliferative luminal breast tumors in METABRIC samples.

    The relationship between amplification of each candidate gene within the context of highly proliferative luminal breast tumors was examined in the METABRIC dataset. Of the nine candidate genes, eight showed a significant enrichment in highly proliferative (top quartile) luminal breast tumors.

  16. Candidate gene expression correlation with PAM50 proliferation score independent of copy number status.
    Supplementary Fig. 10: Candidate gene expression correlation with PAM50 proliferation score independent of copy number status.

    The relationship between mRNA expression and the PAM50 proliferation signature was determined independent of copy number status (t-test) in the TCGA (n=388) and METABRIC (n=1,333) luminal/ ER+ subset of patients. Three classes of genes were identified (top rows) those that have a positive correlation with the signature score irrespective of CN status (EIF2S2, EIF6, MRPS23, CPT1A), those that have an inverse correlation (DTX3) and those that do not show a consistent pattern between datasets (FGD5, METTL6, SLC2A10).

  17. Amplification status of a subset of candidate genes has no reproducible effect on prognosis.
    Supplementary Fig. 11: Amplification status of a subset of candidate genes has no reproducible effect on prognosis.

    Kaplan-Meier survival analysis based upon the amplification status of highly proliferative luminal tumor genes. No consistent difference in disease specific survival was observed for EIF2S2 (A, D), EIF6 (B, E) or SLC2A10 (C, F) when comparing luminal tumors characterized by amplification of each candidate gene relative to luminal tumors without an amplification (log rank p>0.05) in the TCGA (A-C) and METABRIC (D-F) datasets.

Accession codes

Referenced accessions

Gene Expression Omnibus

References

  1. Perou, C.M. et al. Molecular portraits of human breast tumors. Nature 406, 747752 (2000).
  2. Cancer Genome Atlas Network. Comprehensive molecular portraits of human breast tumours. Nature 490, 6170 (2012).
  3. Curtis, C. et al. The genomic and transcriptomic architecture of 2,000 breast tumours reveals novel subgroups. Nature 486, 346352 (2012).
  4. Wood, L.D. et al. The genomic landscapes of human breast and colorectal cancers. Science 318, 11081113 (2007).
  5. Bild, A.H. et al. An integration of complementary strategies for gene-expression analysis to reveal novel therapeutic opportunities for breast cancer. Breast Cancer Res. 11, R55 (2009).
  6. Bild, A.H. et al. Oncogenic pathway signatures in human cancers as a guide to targeted therapies. Nature 439, 353357 (2006).
  7. Rhodes, D.R. et al. Molecular concepts analysis links tumors, pathways, mechanisms, and drugs. Neoplasia 9, 443454 (2007).
  8. Vogelstein, B. & Kinzler, K.W. Cancer genes and the pathways they control. Nat. Med. 10, 789799 (2004).
  9. Marcotte, R. et al. Essential gene profiles in breast, pancreatic, and ovarian cancer cells. Cancer Discov. 2, 172189 (2012).
  10. Gatza, M.L. et al. Analysis of tumor environmental response and oncogenic pathway activation identifies distinct basal and luminal features in HER2-related breast tumor subtypes. Breast Cancer Res. 13, R62 (2011).
  11. Gatza, M.L. et al. A pathway-based classification of human breast cancer. Proc. Natl. Acad. Sci. USA 107, 69946999 (2010).
  12. Fan, C. et al. Building prognostic models for breast cancer patients using clinical variables and hundreds of gene expression signatures. BMC Med. Genomics 4, 3 (2011).
  13. Hoadley, K.A. et al. EGFR associated expression profiles vary with breast tumor subtype. BMC Genomics 8, 258 (2007).
  14. Troester, M.A. et al. Gene expression patterns associated with p53 status in beast cancer. BMC Cancer 6, 276 (2006).
  15. Chandriani, S. et al. A core MYC gene expression signature is prominent in basal-like breast cancer but only partially overlaps the core serum response. PLoS ONE 4, e6693 (2009).
  16. Herschkowitz, J.I., He, X., Fan, C. & Perou, C.M. The functional loss of the retinoblastoma tumour suppressor is a common event in basal-like and luminal B breast carcinomas. Breast Cancer Res. 10, R75 (2008).
  17. Hu, Z. et al. A compact VEGF signature associated with distant metastases and poor outcomes. BMC Med. 7, 9 (2009).
  18. Hutti, J.E. et al. Oncogenic PI3K mutations lead to NF-κB–dependent cytokine expression following growth factor deprivation. Cancer Res. 72, 32603269 (2012).
  19. Oh, D.S. et al. Estrogen-regulated genes predict survival in hormone receptor–positive breast cancers. J. Clin. Oncol. 24, 16561664 (2006).
  20. Thorner, A.R. et al. In vitro and in vivo analysis of B-Myb in basal-like breast cancer. Oncogene 28, 742751 (2009).
  21. Thorner, A.R., Parker, J.S., Hoadley, K.A. & Perou, C.M. Potential tumor suppressor role for the c-Myb oncogene in luminal breast cancer. PLoS ONE 5, e13073 (2010).
  22. Troester, M.A. et al. Activation of host wound responses in breast cancer microenvironment. Clin. Cancer Res. 15, 70207028 (2009).
  23. Usary, J. et al. Mutation of GATA3 in human breast tumors. Oncogene 23, 76697678 (2004).
  24. Harrell, J.C. et al. Endothelial-like properties of claudin-low breast cancer cells promote tumor vascular permeability and metastasis. Clin. Exp. Metastasis 31, 3345 (2014).
  25. Wong, D.J. et al. Module map of stem cell genes guides creation of epithelial cancer stem cells. Cell Stem Cell 2, 333344 (2008).
  26. Ji, H. et al. LKB1 modulates lung cancer differentiation and metastasis. Nature 448, 807810 (2007).
  27. Saal, L.H. et al. Poor prognosis in carcinoma is associated with a gene expression signature of aberrant PTEN tumor suppressor pathway activity. Proc. Natl. Acad. Sci. USA 104, 75647569 (2007).
  28. Glinsky, G.V., Berezovska, O. & Glinskii, A.B. Microarray analysis identifies a death-from-cancer signature predicting therapy failure in patients with multiple types of cancer. J. Clin. Invest. 115, 15031521 (2005).
  29. Lim, E. et al. Aberrant luminal progenitors as the candidate target population for basal tumor development in BRCA1 mutation carriers. Nat. Med. 15, 907913 (2009).
  30. van 't Veer, L.J. et al. Gene expression profiling predicts clinical outcome of breast cancer. Nature 415, 530536 (2002).
  31. Parker, J.S. et al. Supervised risk predictor of breast cancer based on intrinsic subtypes. J. Clin. Oncol. 27, 11601167 (2009).
  32. Chang, J.T. et al. SIGNATURE: a workbench for gene expression signature analysis. BMC Bioinformatics 12, 443 (2011).
  33. Leone, G. et al. Myc requires distinct E2F activities to induce S phase and apoptosis. Mol. Cell 8, 105113 (2001).
  34. Grandis, J.R. et al. Requirement of Stat3 but not Stat1 activation for epidermal growth factor receptor–mediated cell growth in vitro. J. Clin. Invest. 102, 13851392 (1998).
  35. Weigman, V.J. et al. Basal-like breast cancer DNA copy number losses identify genes involved in genomic instability, response to therapy, and patient survival. Breast Cancer Res. Treat. 133, 865880 (2012).
  36. Park, K., Kwak, K., Kim, J., Lim, S. & Han, S. c-Myc amplification is associated with HER2 amplification and closely linked with cell proliferation in tissue microarray of nonselected breast cancers. Hum. Pathol. 36, 634639 (2005).
  37. Nevins, J.R. The Rb/E2F pathway and cancer. Hum. Mol. Genet. 10, 699703 (2001).
  38. Wirapati, P. et al. Meta-analysis of gene expression profiles in breast cancer: toward a unified understanding of breast cancer subtyping and prognosis signatures. Breast Cancer Res. 10, R65 (2008).
  39. Perreard, L. et al. Classification and risk stratification of invasive breast carcinomas using a real-time quantitative RT-PCR assay. Breast Cancer Res. 8, R23 (2006).
  40. Hoadley, K.A. et al. Multi-platform integration of 12 cancer types reveals cell-of-origin classes with distinct molecular signatures. Cell 158, 116 (2014).
  41. Hoeflich, K.P. et al. In vivo antitumor activity of MEK and phosphatidylinositol 3-kinase inhibitors in basal-like breast cancer models. Clin. Cancer Res. 15, 46494664 (2009).
  42. Rody, A. et al. T-cell metagene predicts a favorable prognosis in estrogen receptor–negative and HER2-positive breast cancers. Breast Cancer Res. 11, R15 (2009).
  43. Kurogane, Y. et al. FGD5 mediates proangiogenic action of vascular endothelial growth factor in human vascular endothelial cells. Arterioscler. Thromb. Vasc. Biol. 32, 988996 (2012).
  44. Kishi, N. et al. Murine homologs of deltex define a novel gene family involved in vertebrate Notch signaling and neurogenesis. Int. J. Dev. Neurosci. 19, 2135 (2001).
  45. Matsuno, K., Diederich, R.J., Go, M.J., Blaumueller, C.M. & Artavanis-Tsakonas, S. Deltex acts as a positive regulator of Notch signaling through interactions with the Notch ankyrin repeats. Development 121, 26332644 (1995).
  46. Benelli, D., Cialfi, S., Pinzaglia, M., Talora, C. & Londei, P. The translation factor eIF6 is a Notch-dependent regulator of cell migration and invasion. PLoS ONE 7, e32047 (2012).
  47. Miluzio, A. et al. Impairment of cytoplasmic eIF6 activity restricts lymphomagenesis and tumor progression without affecting normal growth. Cancer Cell 19, 765775 (2011).
  48. Lyng, H. et al. Gene expressions and copy numbers associated with metastatic phenotypes of uterine cervical cancer. BMC Genomics 7, 268 (2006).
  49. Tan, X.L. et al. Genetic variation predicting cisplatin cytotoxicity associated with overall survival in lung cancer patients receiving platinum-based chemotherapy. Clin. Cancer Res. 17, 58015811 (2011).
  50. Gandin, V. et al. Eukaryotic initiation factor 6 is rate-limiting in translation, growth and transformation. Nature 455, 684688 (2008).
  51. Biffo, S. et al. Isolation of a novel β4 integrin–binding protein (p27(BBP)) highly expressed in epithelial cells. J. Biol. Chem. 272, 3031430321 (1997).
  52. Shi, Z.Z. et al. Genomic alterations with impact on survival in esophageal squamous cell carcinoma identified by array comparative genomic hybridization. Genes Chromosom. Cancer 50, 518526 (2011).
  53. Liu, L., Wang, Y.D., Wu, J., Cui, J. & Chen, T. Carnitine palmitoyltransferase 1A (CPT1A): a transcriptional target of PAX3-FKHR and mediates PAX3-FKHR–dependent motility in alveolar rhabdomyosarcoma cells. BMC Cancer 12, 154 (2012).
  54. Samudio, I. et al. Pharmacologic inhibition of fatty acid oxidation sensitizes human leukemia cells to apoptosis induction. J. Clin. Invest. 120, 142156 (2010).
  55. Pacilli, A. et al. Carnitine-acyltransferase system inhibition, cancer cell death, and prevention of myc-induced lymphomagenesis. J. Natl. Cancer Inst. 105, 489498 (2013).
  56. Reich, M. et al. GenePattern 2.0. Nat. Genet. 38, 500501 (2006).

Download references

Author information

Affiliations

  1. Lineberger Comprehensive Cancer Center, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA.

    • Michael L Gatza,
    • Grace O Silva,
    • Joel S Parker,
    • Cheng Fan &
    • Charles M Perou
  2. Department of Genetics, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA.

    • Michael L Gatza,
    • Grace O Silva,
    • Joel S Parker &
    • Charles M Perou
  3. Curriculum in Bioinformatics and Computational Biology, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA.

    • Grace O Silva &
    • Charles M Perou
  4. Department of Pathology and Laboratory Medicine, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA.

    • Charles M Perou

Contributions

M.L.G., J.S.P. and C.M.P. conceived and designed the study. M.L.G., G.O.S. and C.F. performed analyses. M.L.G. and C.M.P. wrote the manuscript. All authors have reviewed and approved the final manuscript.

Competing financial interests

C.M.P. is an equity stock holder and board of director member of BioClassifier LLC and GeneCentric Diagnostics. C.M.P. is also listed as an inventor on a patent application for the PAM50 and VEGF-signature molecular assays. J.S.P. is also listed as an inventor on a patent application for the PAM50 and VEGF-signature molecular assays.

Corresponding author

Correspondence to:

Author details

Supplementary information

Supplementary Figures

  1. Supplementary Figure 1: Patterns of pathway activity correspond with molecular subtypes of breast cancer. (1,210 KB)

    Analysis of molecular subtypes of breast cancer based on 52 gene expression signature scores. Euclidean distance was used to calculate the relationship between samples based on scores of 52 gene express signatures. Samples are commonly ordered on the X and Y axis according to molecular subtype. These results demonstrate high concordance within a subtype (dark blue), and lower concordance across subtypes; each sample versus itself is the blue diagonal line.

  2. Supplementary Figure 2: Correlation between calculated pathway activity. (578 KB)

    A Pearson correlation matrix of each signature versus all other signatures (including itself as the diagonal line) demonstrates a high degree of concordance amongst independently developed gene expression signatures measuring similar or associated pathways. Red indicates high positive correlation and blue a strong anti-correlation.

  3. Supplementary Figure 3: Identification of pathway-specific copy number alterations by Spearman rank correlation. (443 KB)

    A Spearman rank correlation, both positive (red) and negative (blue) were used to identify associations between predicted genomic signature pathway activity and gene-level DNA copy number content (n=476). The negative log10 Bonferroni adjust p-values are plotted according to chromosomal position. Chromosomal borders are delineated by vertical black lines.

  4. Supplementary Figure 4: Identification of pathway-specific copy number alterations based on frequency of gains or losses calculated by Fisher’s exact test. (461 KB)

    A Fisher’s exact test was used to calculate the statistical significance of the frequency of copy number gains (red) or losses (blue) in samples with the highest (top quartile) pathway signature activity relative to all other samples (n=476). The negative log10 Bonferroni adjust p-values are plotted according to chromosomal position. Vertical black lines indicate chromosomal borders.

  5. Supplementary Figure 5: Patterns of pathway activity in human breast cancer cell lines. (615 KB)

    The scored pathway activity for a panel of 51 breast cancer cell lines (GSE12777) was calculated for the 52 pathway signatures. Of these cell lines, 27 which are denoted by black squares in lower panel were subjected to a genome-wide RNAi screen.

  6. Supplementary Figure 6: Identification of essential genes in proliferative breast cancer cell lines. (101 KB)

    Identification of genes essential for cell viability in vitro in the context of the 11-gene PAM50 Proliferation signature in (A) all cell line samples and (B) in luminal and HER2+ breast cancer cell lines. The negative log10 Spearman rank correlation p values are plotted for each gene relative to chromosomal position.

  7. Supplementary Figure 7: Correlation between candidate gene mRNA expression and DNA copy number status in TCGA samples. (234 KB)

    The mRNA expression levels of the 21 identified candidate genes that are required for cell viability and are uniquely amplified in highly proliferative luminal breast tumors. In each plot, the mRNA levels from the TCGA data are compared in those tumors with amplifications versus all others. Of the 21 genes, two (ANKRD56 and TMEM189) were not present on the mRNA expression array and are not included here. Of the remaining 19 genes, 12 had a significant relationship (p<0.05) between copy number status and mRNA expression levels.

  8. Supplementary Figure 8: Correlation between candidate gene mRNA expression and DNA copy number status in METABRIC samples. (152 KB)

    The mRNA expression levels of the 12 identified candidate genes that are required for cell viability and are uniquely amplified in highly proliferative luminal breast tumors were analyzed within the context of copy number level in the METABRIC dataset. Of these 12 genes, three (SNX21, ZBTB46 and DNAJC5) were not present on both of the METABRIC data platforms (mRNA expression and copy number). Of the remaining 9 genes, all had a significant relationship (p<0.05) between copy number status and mRNA expression levels.

  9. Supplementary Figure 9: Validation of increased candidate gene copy number status in highly proliferative luminal breast tumors in METABRIC samples. (125 KB)

    The relationship between amplification of each candidate gene within the context of highly proliferative luminal breast tumors was examined in the METABRIC dataset. Of the nine candidate genes, eight showed a significant enrichment in highly proliferative (top quartile) luminal breast tumors.

  10. Supplementary Figure 10: Candidate gene expression correlation with PAM50 proliferation score independent of copy number status. (393 KB)

    The relationship between mRNA expression and the PAM50 proliferation signature was determined independent of copy number status (t-test) in the TCGA (n=388) and METABRIC (n=1,333) luminal/ ER+ subset of patients. Three classes of genes were identified (top rows) those that have a positive correlation with the signature score irrespective of CN status (EIF2S2, EIF6, MRPS23, CPT1A), those that have an inverse correlation (DTX3) and those that do not show a consistent pattern between datasets (FGD5, METTL6, SLC2A10).

  11. Supplementary Figure 11: Amplification status of a subset of candidate genes has no reproducible effect on prognosis. (320 KB)

    Kaplan-Meier survival analysis based upon the amplification status of highly proliferative luminal tumor genes. No consistent difference in disease specific survival was observed for EIF2S2 (A, D), EIF6 (B, E) or SLC2A10 (C, F) when comparing luminal tumors characterized by amplification of each candidate gene relative to luminal tumors without an amplification (log rank p>0.05) in the TCGA (A-C) and METABRIC (D-F) datasets.

PDF files

  1. Supplementary Text and Figures (6,105 KB)

    Supplementary Figures 1–11

Excel files

  1. Supplementary Tables 1–5 (520 KB)

    Supplementary Tables 1–5

  2. Supplementary Tables 6–18 (17,165 KB)

    Supplementary Tables 6–18

  3. Supplementary Tables 19–31 (16,705 KB)

    Supplementary Tables 19–31

  4. Supplementary Tables 32–44 (17,417 KB)

    Supplementary Tables 32–44

  5. Supplementary Tables 45–57 (17,124 KB)

    Supplementary Tables 45–57

  6. Supplementary Tables 58–67 (9,444 KB)

    Supplementary Tables 58–67

Additional data