The application of artificial intelligence to microarray data: Identification of a novel gene signature to identify bladder cancer progression

BACKGROUND New methods for identifying bladder cancer (BCa) progression are required. Gene expression microarrays can reveal insights into disease biology and identify novel biomarkers. However, these experiments produce large datasets that are difficult to interpret. OBJECTIVE To develop a novel method of microarray analysis combining two forms of artificial intelligence (AI): neurofuzzy modelling (NFM) and artificial neural networks (ANN) and validate it in a BCa cohort. DESIGN, SETTING, AND PARTICIPANTS We used AI and statistical analyses to identify progression-related genes in a microarray dataset (n=66 tumours, n=2800 genes). The AI-selected genes were then investigated in a second cohort (n=262 tumours) using immunohistochemistry. MEASUREMENTS We compared the accuracy of AI and statistical approaches to identify tumour progression. RESULTS AND LIMITATIONS AI identified 11 progression-associated genes (odds ratio [OR]: 0.70; 95% confidence interval [CI], 0.56-0.87; p=0.0004), and these were more discriminate than genes chosen using statistical analyses (OR: 1.24; 95% CI, 0.96-1.60; p=0.09). The expression of six AI-selected genes (LIG3, FAS, KRT18, ICAM1, DSG2, and BRCA2) was determined using commercial antibodies and successfully identified tumour progression (concordance index: 0.66; log-rank test: p=0.01). AI-selected genes were more discriminate than pathologic criteria at determining progression (Cox multivariate analysis: p=0.01). Limitations include the use of statistical correlation to identify 200 genes for AI analysis and that we did not compare regression identified genes with immunohistochemistry. CONCLUSIONS AI and statistical analyses use different techniques of inference to determine gene-phenotype associations and identify distinct prognostic gene signatures that are equally valid. We have identified a prognostic gene signature whose members reflect a variety of carcinogenic pathways that could identify progression in non-muscle-invasive BCa.

[1]  Peter A. Jones,et al.  Prognostic relevance of methylation markers in patients with non-muscle invasive bladder carcinoma. , 2005, European journal of cancer.

[2]  C. Begg,et al.  Comparing tumour staging and grading systems: a case study and a review of the issues, using thymoma as a model. , 2000, Statistics in medicine.

[3]  Alexandre Pelzer,et al.  Small G‐protein RhoE is underexpressed in prostate cancer and induces cell cycle arrest and apoptosis , 2005, The Prostate.

[4]  D Haussler,et al.  Knowledge-based analysis of microarray gene expression data by using support vector machines. , 2000, Proceedings of the National Academy of Sciences of the United States of America.

[5]  Annette M. Molinaro,et al.  Prediction error estimation: a comparison of resampling methods , 2005, Bioinform..

[6]  D. Linkens,et al.  Application of artificial intelligence to the management of urological cancer. , 2007, The Journal of urology.

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

[8]  K. Rieger-Christ,et al.  Restoration of plakoglobin expression in bladder carcinoma cell lines suppresses cell migration and tumorigenic potential , 2005, British Journal of Cancer.

[9]  G. Ball,et al.  A validated gene expression profile for detecting clinical outcome in breast cancer using artificial neural networks , 2010, Breast Cancer Research and Treatment.

[10]  T. H. van der Kwast,et al.  FGFR3 mutations and a normal CK20 staining pattern define low-grade noninvasive urothelial bladder tumours. , 2007, European urology.

[11]  M. Bogliolo,et al.  Reduced ligation during DNA base excision repair supported by BRCA2 mutant cells , 2000, Oncogene.

[12]  Maysam F. Abbod,et al.  Artificial intelligence for the prediction bladder cancer , 2004 .

[13]  K Takahashi,et al.  Prognostic impact of FAS/CD95/APO-1 in urothelial cancers: decreased expression of Fas is associated with disease progression , 2005, British Journal of Cancer.

[14]  Christian Pilarsky,et al.  Gene Expression Profiling of Progressive Papillary Noninvasive Carcinomas of the Urinary Bladder , 2005, Clinical Cancer Research.

[15]  Maysam F Abbod,et al.  Neuro-fuzzy modeling: an accurate and interpretable method for predicting bladder cancer progression. , 2006, The Journal of urology.

[16]  D. Ruppert.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction , 2004 .

[17]  W. Vach,et al.  On the misuses of artificial neural networks for prognostic and diagnostic classification in oncology. , 2000, Statistics in medicine.

[18]  Freddie C Hamdy,et al.  Differential expression of hMLH1 and hMSH2 is related to bladder cancer grade, stage and prognosis but not microsatellite instability , 2003, International journal of cancer.

[19]  Yair Lotan,et al.  Recurrence and progression of disease in non-muscle-invasive bladder cancer: from epidemiology to treatment strategy. , 2009, European urology.

[20]  Minyou Chen,et al.  Artificial intelligence in predicting bladder cancer outcome: a comparison of neuro-fuzzy modeling and artificial neural networks. , 2003, Clinical cancer research : an official journal of the American Association for Cancer Research.

[21]  F. Hamdy,et al.  Less is more: artificial intelligence and gene-expression arrays , 2004, The Lancet.

[22]  J Alfred Witjes,et al.  Predicting recurrence and progression in individual patients with stage Ta T1 bladder cancer using EORTC risk tables: a combined analysis of 2596 patients from seven EORTC trials. , 2006, European urology.

[23]  Torben F. Ørntoft,et al.  Identifying distinct classes of bladder carcinoma using microarrays , 2003, Nature Genetics.

[24]  D. Ransohoff.  Rules of evidence for cancer molecular-marker discovery and validation , 2004, Nature Reviews Cancer.

[25]  Vojislav Kecman,et al.  Gene extraction for cancer diagnosis by support vector machines - An improvement , 2005, Artif. Intell. Medicine.

[26]  Anne-Mette K. Hein,et al.  Gene Expression Signatures Predict Outcome in Non–Muscle-Invasive Bladder Carcinoma: A Multicenter Validation Study , 2007, Clinical Cancer Research.

[27]  A. Dupuy,et al.  Critical review of published microarray studies for cancer outcome and guidelines on statistical analysis and reporting. , 2007, Journal of the National Cancer Institute.

[28]  D. Linkens,et al.  Neurofuzzy Modeling to Determine Recurrence Risk Following Radical Cystectomy for Nonmetastatic Urothelial Carcinoma of the Bladder , 2009, Clinical Cancer Research.

[29]  Derek A. Linkens,et al.  A systematic neuro-fuzzy modeling framework with application to material property prediction , 2001, IEEE Trans. Syst. Man Cybern. Part B.