Farzana Kabir Ahmad, Safaai Deris & Nor Hayati Othman International Journal of Biometrics and Bioinformatics , (IJBB), Volume (3) : Issue (4) 31 Toward Integrated Clinical and Gene- Expression Profiles For Breast Cancer Prognosis: A Review Paper Farzana Kabir Ahmad farzana58@uum.edu.my Graduate Department of Computer Science, College of Arts and Sciences, Universiti Utara Malaysia, 06010 Sintok, Kedah, Malaysia Safaai Deris safaai@utm.my School of Postgraduate Studies, Universiti Teknologi Malaysia, 81310 Skudai, Johor, Malaysia Nor Hayati Othman hayati@kb.usm.my Clinical Research Platform & Pathologist, Health Campus Universiti Sains Malaysia, 16150 Kubang Kerian, Kelantan,Malaysia Abstract Breast cancer patients with the same diagnostic and clinical prognostics profile can have markedly different clinical outcomes. This difference is possibly caused by the limitation of current breast cancer prognostic indices, which group molecularly distinct patients into similar clinical classes based mainly on the morphology of diseases. Traditional clinical-based prognosis models were discovered to contain some restrictions to address the heterogeneity of breast cancer. The invention of microarray technology and its ability to simultaneously interrogate thousands of genes has changed the paradigm of molecular classification of human cancers as well as shifting clinical prognosis models to a broader prospect. Numerous studies have revealed the potential value of gene- expression signatures in examining the risk of disease recurrence. However, most of these studies attempted to implement genetic-marker based prognostic models to replace the traditional clinical markers, yet neglecting the rich information contained in clinical information. Therefore, this research took the effort to integrate both clinical and microarray data in order to obtain accurate breast cancer prognosis, by taking into account that these data complement each other. This article presents a review of the development of breast cancer prognosis models, concentrating precisely on clinical and gene-expression profiles. The literature is reviewed in an explicit machine-learning framework, which includes the elements of feature selection and classification techniques. Keywords: Breast cancer, Prognosis, Gene-Expression Profiles, Feature selection, Classification.