High dimensional surrogacy: computational aspects of an upscaled analysis.

Identification of genomic biomarkers is an important area of research in the context of drug discovery experiments. These experiments typically consist of several high dimensional datasets that contain information about a set of drugs (compounds) under development. This type of data structure introduces the challenge of multi-source data integration. High-Performance Computing (HPC) has become an important tool for everyday research tasks. In the context of drug discovery, high dimensional multi-source data needs to be analyzed to identify the biological pathways related to the new set of drugs under development. In order to process all information contained in the datasets, HPC techniques are required. Even though R packages for parallel computing are available, they are not optimized for a specific setting and data structure. In this article, we propose a new framework, for data analysis, to use R in a computer cluster. The proposed data analysis workflow is applied to a multi-source high dimensional drug discovery dataset and compared with a few existing R packages for parallel computing.

Adjusting Overall Survival Estimates for Treatment Switching in Metastatic, Castration-Sensitive Prostate Cancer: Results from the LATITUDE Study.

BACKGROUND: LATITUDE was the first phase 3 trial examining the survival benefit of adding abiraterone acetate (AA) + prednisone (P) to androgen-deprivation therapy (ADT) in newly diagnosed metastatic, castration-sensitive prostate cancer (mCSPC). Due to significant improvement in overall survival after the first interim analysis, patients in the placebos + ADT arm could switch to AA + P + ADT during an open-label extension. As in other studies where switching is allowed, statistical adjustments are needed to assess the real benefit of new drugs. PATIENTS AND METHODS: This was a post hoc analysis to estimate the true survival benefit of AA + P + ADT in patients with newly diagnosed mCSPC by applying statistical adjustments commonly used to adjust for treatment switching. RESULTS: Of 112 patients still receiving placebos + ADT at the first interim analysis, 72 switched to AA + P + ADT during the open-label extension. Final analysis was conducted after median follow-up of 51.8 months. Compared to the placebos + ADT arm, the risk of death in the AA + P + ADT arm was 34% lower [hazard ratio (HR) = 0.663 (95% confidence interval 0.566-0.778)] by unadjusted intent-to-treat analysis, 37% lower [HR = 0.629 (95% confidence interval 0.526-0.753)] by rank preserving structure failure time modeling, and 38% lower [HR = 0.616 (95% confidence interval 0.524-0.724)] by inverse probability of censoring weights. CONCLUSIONS: Analyses adjusting for treatment switching using two different statistical approaches confirm the improved survival benefit of adding AA + P to ADT in patients with newly diagnosed mCSPC. TRIAL REGISTRATION: ClinicalTrials.gov identifier NCT01715285.