Predicting Platinum Resistance in Ovarian Cancer Treatment from Integrated Genomic Data

• Activity Number: 36 - Diagnostic, Prognostic, and Predictive Genomic Biomarkers for Cancer
• Type: Contributed
• Date/Time: Sunday, July 30, 2017 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics in Genomics and Genetics
• Author(s): Esther Drill* and Ronglai Shen and Yuanjia Wang
• Companies: Memorial Sloan Kettering Cancer Center and Memorial Sloan Kettering Cancer Center and Columbia University
• Keywords: Integrative data analysis ; Cancer genomics ; Latent variable models ; Variable selection

Abstract

Motivated by the challenges of improving the prediction of platinum resistance in high grade serous ovarian cancer (HGSOC) treatment, we propose novel integrative methods to leverage data across multiple genomic platforms to perform classification using a joint latent variable model. This approach provides effective dimension reduction while handling heterogeneous data types. It also provides a natural framework to incorporate covariates. Effective feature selection is performed through a thresholding parameter that combines both latent variable and feature coefficients. We demonstrate increased accuracy in prediction over methods that assume homogeneous data type, such as linear discriminant analysis and lasso regression, and improved feature selection. In simulations, our method minimizes false positive and notably false negative rates compared to lasso regression. We apply it to a TCGA cohort of HGSOC patients with 3 types of genomic data and platinum response data. This methodology has broad applications beyond predicting treatment outcomes and disease progression in cancer, including predicting prognosis and diagnosis in other diseases with major public health implications.