By Vijay Pillai
March 1, 2008 | To improve the probability of success in biomarker discovery, accurately classifying the patient population for a given disease or disease subtype is essential. Translational medicine — which attempts to more directly connect basic research to patient care — represents a fundamental shift in the pharmaceutical industry’s approach to drug development. Historically, drug discovery and development had little connection to clinical medicine outside of leveraging the clinical setting for trials in the final stages of the drug development process. Together, translational medicine and adaptive trials will create more of a dynamic environment of constant study, experimentation, and adaptation. This approach promises to deliver new insights into disease processes, guide and accelerate more successful drug discovery and development processes, improve drug efficacy, and enable personalized medicine that ultimately improves patient care.
These transformations within the industry are starting to bring more focus on information technology as an enabler of innovation in addition to the rather typical role IT played in process streamlining. Participants in the research and development (R&D) process will need to triangulate clinical information (for phenotypic data), laboratory data (pathology and genetic), and related medical images through integrated platforms in order to understand kinetics, safety, and drug efficacy. Any veteran of the pharmaceutical or health care industry, however, can tell you that such connectivity has historically been elusive, complicated by silos of systems, data formats, and standards.
As pharmaceuticals companies continue to outsource a growing share of development activities to Contract Research Organizations (CROs), they receive study data from multiple sources using multiple software platforms — creating an environment with massive integration and interoperable challenges. Simply using IT to eliminate paper or streamlining study data capture, alone, will not improve the probability of success in drug development. Moving from using IT to reduce data errors and improve process efficiencies, companies are now beginning to focus on leveraging IT to assemble data from current and past studies to build sophisticated knowledge bases around therapeutic areas.
Study data sent by CROs are collected using multiple electronic data capture (EDC) solutions and present an integration challenge for sponsor companies. Similarly, laboratory data and results come into the sponsor institutions from multiple sources and need to be tied to the study data. Medical image data from radiology, cardiology, and other specialties also come in from various clinical sites, each having its own picture archiving and communication system (PACS). Traditionally, a medical image generated and stored in one vendor’s PACS platform cannot be viewed by another vendor’s viewing station, or stored in a PACS due to the proprietary nature in which images are stored. In spite of the Digital Imaging and Communications in Medicine (DICOM) standards for medical images, subtle variations from one vendor to another persist.
As data integration challenges continue to plague study sponsors, technology providers are beginning to create platforms to assemble and pull all relevant data together to provide an accurate and complete picture of the study — ensuring safety and efficacy throughout the drug development process.
In an adaptive trial environment, early access to clinical data is critical not only in order to stratify and identify cohorts for patient selection, but also to correlate study outcomes to phenotypes. Sponsors also need near real-time access to patient study data to analyze and better understand toxicology (TOX) and absorption, distribution, metabolism, and excretion (ADME) results, and to fine tune study protocols on an as-needed basis. Further, pharmas require increased visibility into post-market data as well as past study data to better understand adverse events and drug failures or to develop a new generation of an existing drug.
Mounting data complexities have a direct impact on the increased need for sophisticated analytics. Apart from the usual operational analytics, study sponsors and clinical research centers look toward securing knowledge-based analytics using clinical ontologies to improve our understanding around therapeutic areas and retain institutional knowledge.
While translational medicine holds potential to provide significant benefits to the development of new drugs and delivery of health care, IT will play a pivotal role in harmonizing information and extracting critical data points for biomarker discovery. Pharmaceutical organizations today need IT solutions that enable the two-way flow of understandable and actionable data between the R&D lab and the clinical environment to successfully develop safe and better drugs, and help adopt a personalized approach to patient care.
Vijay Pillai is the director of life sciences and translational medicine, global industry business unit, Oracle. He can be reached at firstname.lastname@example.org.
This article appeared in Bio-IT World Magazine.
Subscriptions are free for qualifying individuals. Apply Today.