A validated gene expression profile for detecting clinical outcome in breast cancer using artificial neural networks

Gene expression microarrays allow for the high throughput analysis of huge numbers of gene transcripts and this technology has been widely applied to the molecular and biological classification of cancer patients and in predicting clinical outcome. A potential handicap of such data intensive molecular technologies is the translation to clinical application in routine practice. In using an artificial neural network bioinformatic approach, we have reduced a 70 gene signature to just 9 genes capable of accurately predicting distant metastases in the original dataset. Upon validation in a follow-up cohort, this signature was an independent predictor of metastases free and overall survival in the presence of the 70 gene signature and other factors. Interestingly, the ANN signature and CA9 expression also split the groups defined by the 70 gene signature into prognostically distinct groups. Subsequently, the presence of protein for the principal prognosticator gene was categorically assessed in breast cancer tissue of an experimental and independent validation patient cohort, using immunohistochemistry. Importantly our principal prognosticator, CA9, showed that it is capable of selecting an aggressive subgroup of patients who are known to have poor prognosis.

[1]  H. Bloom,et al.  Histological Grading and Prognosis in Breast Cancer , 1957, British Journal of Cancer.

[2]  R. Dennis Cook,et al.  Cross-Validation of Regression Models , 1984 .

[3]  Geoffrey E. Hinton,et al.  Learning representations by back-propagating errors , 1986, Nature.

[4]  B. Efron.  How Biased is the Apparent Error Rate of a Prediction Rule , 1986 .

[5]  J. Shao.  Linear Model Selection by Cross-validation , 1993 .

[6]  P. Disaia,et al.  Identification of the MN antigen as a diagnostic biomarker of cervical intraepithelial squamous and glandular neoplasia and cervical carcinomas. , 1994, The American journal of pathology.

[7]  M. Dewhirst,et al.  Tumor oxygenation predicts for the likelihood of distant metastases in human soft tissue sarcoma. , 1996, Cancer research.

[8]  P Vaupel,et al.  Association between tumor hypoxia and malignant progression in advanced cancer of the uterine cervix. , 1996, Cancer research.

[9]  J. Pastorek,et al.  Immunohistochemical study of colorectal tumors for expression of a novel transmembrane carbonic anhydrase, MN/CA IX, with potential value as a marker of cell proliferation. , 1998, The American journal of pathology.

[10]  D. Botstein,et al.  Exploring the new world of the genome with DNA microarrays , 1999, Nature Genetics.

[11]  R. Gillies,et al.  ©1999 Cancer Research Campaign Article no. bjoc.1998.0455 Enhancement of chemotherapy by manipulation of tumour pH , 2022 .

[12]  John R. W. Masters,et al.  How diagnosis with microarrays can help cancer patients , 2000, Nature.

[13]  Christian A. Rees,et al.  Molecular portraits of human breast tumours , 2000, Nature.

[14]  R. Tibshirani,et al.  Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications , 2001, Proceedings of the National Academy of Sciences of the United States of America.

[15]  A. Harris,et al.  Prognostic significance of a novel hypoxia-regulated marker, carbonic anhydrase IX, in invasive breast carcinoma. , 2001, Journal of clinical oncology : official journal of the American Society of Clinical Oncology.

[16]  M. Ringnér,et al.  Classification and diagnostic prediction of cancers using gene expression profiling and artificial neural networks , 2001, Nature Medicine.

[17]  E. Lander,et al.  Classification of human lung carcinomas by mRNA expression profiling reveals distinct adenocarcinoma subclasses , 2001, Proceedings of the National Academy of Sciences of the United States of America.

[18]  R. Spang,et al.  Predicting the clinical status of human breast cancer by using gene expression profiles , 2001, Proceedings of the National Academy of Sciences of the United States of America.

[19]  G. Li,et al.  An integrated approach utilizing artificial neural networks and SELDI mass spectrometry for the classification of human tumours and rapid identification of potential biomarkers , 2002, Bioinform..

[20]  Yudong D. He,et al.  Gene expression profiling predicts clinical outcome of breast cancer , 2002, Nature.

[21]  Yudong D. He,et al.  A Gene-Expression Signature as a Predictor of Survival in Breast Cancer , 2002 .

[22]  C. Sawyers,et al.  TMEFF2 is an androgen-regulated gene exhibiting antiproliferative effects in prostate cancer cells , 2002, Oncogene.

[23]  Meland,et al.  THE USE OF MOLECULAR PROFILING TO PREDICT SURVIVAL AFTER CHEMOTHERAPY FOR DIFFUSE LARGE-B-CELL LYMPHOMA , 2002 .

[24]  Carlos Caldas,et al.  Molecular Classification of Breast Carcinomas Using Tissue Microarrays , 2003, Diagnostic molecular pathology : the American journal of surgical pathology, part B.

[25]  S. Gery,et al.  Repression of the TMEFF2 promoter by c-Myc. , 2003, Journal of molecular biology.

[26]  A. Gown,et al.  Immunohistochemical and Clinical Characterization of the Basal-Like Subtype of Invasive Breast Carcinoma , 2004, Clinical Cancer Research.

[27]  M. Cronin,et al.  A multigene assay to predict recurrence of tamoxifen-treated, node-negative breast cancer. , 2004, The New England journal of medicine.

[28]  Yi-Zeng Liang,et al.  Monte Carlo cross‐validation for selecting a model and estimating the prediction error in multivariate calibration , 2004 .

[29]  E. van Marck,et al.  The Presence of a Fibrotic Focus in Invasive Breast Carcinoma Correlates with the Expression of Carbonic Anhydrase IX and is a Marker of Hypoxia and Poor Prognosis , 2003, Breast Cancer Research and Treatment.

[30]  A. Harris,et al.  Necrosis and Hypoxia in Invasive Breast Carcinoma , 2003, Breast Cancer Research and Treatment.

[31]  I. Hedenfalk,et al.  ERK1/2 inhibition increases antiestrogen treatment efficacy by interfering with hypoxia-induced downregulation of ERα: a combination therapy potentially targeting hypoxic and dormant tumor cells , 2005, Oncogene.

[32]  G. Glinsky,et al.  Microarray analysis identifies a death-from-cancer signature predicting therapy failure in patients with multiple types of cancer. , 2005, The Journal of clinical investigation.

[33]  A. Harris,et al.  Differential prognostic impact of hypoxia induced and diffuse HIF-1α expression in invasive breast cancer , 2005, Journal of Clinical Pathology.

[34]  Graham R. Ball,et al.  Classification of bacterial species from proteomic data using combinatorial approaches incorporating artificial neural networks, cluster analysis and principal components analysis , 2005, Bioinform..

[35]  G. Ball,et al.  High‐throughput protein expression analysis using tissue microarray technology of a large well‐characterised series identifies biologically distinct classes of breast cancer confirming recent cDNA expression analyses , 2005, International journal of cancer.

[36]  J. Reis-Filho,et al.  Basal-like breast cancer and the BRCA1 phenotype , 2006, Oncogene.

[37]  R. Millikan,et al.  CA IX is an Independent Prognostic Marker in Premenopausal Breast Cancer Patients with One to Three Positive Lymph Nodes and a Putative Marker of Radiation Resistance , 2006, Clinical Cancer Research.

[38]  A. Nobel,et al.  Concordance among Gene-Expression – Based Predictors for Breast Cancer , 2011 .

[39]  M. Izquierdo,et al.  RNA interference against Hec1 inhibits tumor growth in vivo , 2006, Gene Therapy.

[40]  J. Foekens,et al.  Gene expression profiles associated with the growth pattern and presence of a fibrotic focus in breast cancer , 2006 .

[41]  I. Ellis,et al.  A gene-expression signature to predict survival in breast cancer across independent data sets , 2007, Oncogene.

[42]  G. Ball,et al.  Clinical proteomics: discovery of cancer biomarkers using mass spectrometry and bioinformatics approaches--a prostate cancer perspective. , 2007, Vaccine.

[43]  Emmanuel Chamorey,et al.  HIF-1alpha and CA IX staining in invasive breast carcinomas: prognosis and treatment outcome. , 2007, International journal of cancer.

[44]  O. Fadare,et al.  The Phenotypic Spectrum of Basal-like Breast Cancers: A Critical Appraisal , 2007, Advances in anatomic pathology.

[45]  Emmanuel Chamorey,et al.  HIF‐1α and CA IX staining in invasive breast carcinomas: Prognosis and treatment outcome , 2007 .

[46]  早馬 聡.  Activation of CDCA1-KNTC2, members of centromere protein complex, involved in pulmonary carcinogenesis , 2007 .

[47]  C. Sotiriou,et al.  Meta-analysis of gene expression profiles in breast cancer: toward a unified understanding of breast cancer subtyping and prognosis signatures , 2007, Breast Cancer Research.

[48]  L. Montanaro,et al.  The basal‐like breast carcinoma phenotype is regulated by SLUG gene expression , 2008, The Journal of pathology.

[49]  Graham R. Ball,et al.  Identification of gene transcript signatures predictive for estrogen receptor and lymph node status using a stepwise forward selection artificial neural network modelling approach , 2008, Artif. Intell. Medicine.

[50]  Samuel Leung,et al.  Basal-Like Breast Cancer Defined by Five Biomarkers Has Superior Prognostic Value than Triple-Negative Phenotype , 2008, Clinical Cancer Research.