Clinical Genomics Data Use: Protecting Patients Privacy Rights

Abstract

The rate of any biomarker discovery is highly influenced by the costs associated with obtaining the relevant genetic, genomic, and clinical data. Currently, to identify a human genetic or genomic biomarker, a clinical researcher must identify a set of transplant recipients that either has been or can be clinically phenotyped, and then obtain samples from patients that can be subjected to gene expression and proteomic analysis. This process of obtaining a clinical cohort by matching donors to recipients is usually time-consuming and expensive. It is further time constrained by patient's signed written consent which has a use-by date.Biomarker database systems can play a pivotal role in establishing mechanisms that enable clinical researchers to access clinically derived genetic and co-variate data in a secure, privacy-aware, de-identified and auditable manner. They can also establish mechanisms and audit trails that monitor the movement of research data across institutions due to their custodial responsibilities to protect patient privacy.In this paper, we prototype the concept of a de-identified centralized biomarker repository that empowers principal investigators and researchers to access multiple data sources needed for the conduct of genomic, proteomic interpretation and computational experiments. The repository takes the form of a research data mart such that the enrolled patients' clinical and treatment history can be tracked during pre and post transplant phases. We demonstrate how technology and information engineering design principles help to implement a de-identification schema for patients to protect patient privacy and prevent unauthorized data access.