+1 Recommend
0 collections
      • Record: found
      • Abstract: found
      • Article: not found

      Weibull regression with Bayesian variable selection to identify prognostic tumour markers of breast cancer survival.

      Read this article at

          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.


          As data-rich medical datasets are becoming routinely collected, there is a growing demand for regression methodology that facilitates variable selection over a large number of predictors. Bayesian variable selection algorithms offer an attractive solution, whereby a sparsity inducing prior allows inclusion of sets of predictors simultaneously, leading to adjusted effect estimates and inference of which covariates are most important. We present a new implementation of Bayesian variable selection, based on a Reversible Jump MCMC algorithm, for survival analysis under the Weibull regression model. A realistic simulation study is presented comparing against an alternative LASSO-based variable selection strategy in datasets of up to 20,000 covariates. Across half the scenarios, our new method achieved identical sensitivity and specificity to the LASSO strategy, and a marginal improvement otherwise. Runtimes were comparable for both approaches, taking approximately a day for 20,000 covariates. Subsequently, we present a real data application in which 119 protein-based markers are explored for association with breast cancer survival in a case cohort of 2287 patients with oestrogen receptor-positive disease. Evidence was found for three independent prognostic tumour markers of survival, one of which is novel. Our new approach demonstrated the best specificity.

          Related collections

          Most cited references49

          • Record: found
          • Abstract: not found
          • Article: not found

          Variable Selection via Nonconcave Penalized Likelihood and its Oracle Properties

            • Record: found
            • Abstract: not found
            • Article: not found

            The Adaptive Lasso and Its Oracle Properties

            Hui Zou (2006)
              • Record: found
              • Abstract: found
              • Article: not found

              Tissue microarrays for high-throughput molecular profiling of tumor specimens.

              Many genes and signalling pathways controlling cell proliferation, death and differentiation, as well as genomic integrity, are involved in cancer development. New techniques, such as serial analysis of gene expression and cDNA microarrays, have enabled measurement of the expression of thousands of genes in a single experiment, revealing many new, potentially important cancer genes. These genome screening tools can comprehensively survey one tumor at a time; however, analysis of hundreds of specimens from patients in different stages of disease is needed to establish the diagnostic, prognostic and therapeutic importance of each of the emerging cancer gene candidates. Here we have developed an array-based high-throughput technique that facilitates gene expression and copy number surveys of very large numbers of tumors. As many as 1000 cylindrical tissue biopsies from individual tumors can be distributed in a single tumor tissue microarray. Sections of the microarray provide targets for parallel in situ detection of DNA, RNA and protein targets in each specimen on the array, and consecutive sections allow the rapid analysis of hundreds of molecular markers in the same set of specimens. Our detection of six gene amplifications as well as p53 and estrogen receptor expression in breast cancer demonstrates the power of this technique for defining new subgroups of tumors.

                Author and article information

                Stat Methods Med Res
                Statistical methods in medical research
                SAGE Publications
                Feb 2017
                : 26
                : 1
                [1 ] 1 MRC Biostatistics Unit, Cambridge, UK.
                [2 ] 2 Cancer Research UK Cambridge Institute, Cambridge, UK.
                [3 ] 3 Department of Pathology, University of Cambridge, Cambridge, UK.
                [4 ] 4 Cambridge Experimental Cancer Medicine Centre and NIHR Cambridge Biomedical Research Centre, Cambridge, UK.
                [5 ] 5 Department of Oncology, University of Cambridge, Cambridge, UK.
                [6 ] 6 NIH Cambridge Biomedical Research Centre, Cambridge, UK.
                [7 ] 7 Strangeways Research Laboratory, Cambridge, UK.
                0962280214548748 EMS78373

                reversible jump,Bayesian variable selection,MCMC,survival analysis,breast cancer,gene expression,stability selection,penalised regression


                Comment on this article