Integrating clinical information and R&D data will ensure success.
By Neil de Crescenzo
March 24, 2009 | Personalized medicine promises to improve understanding of the mechanisms of disease and permit more effective patient care. It also stands to transform the focus of the pharmaceutical industry as the cost of drug discovery exceeds $1 billion, and established blockbuster drugs lose patent protection.
Personalized treatments based on an individual’s genetic profile would usually be targeted toward sub-segments of the population, creating new opportunities for drugs that might have failed traditional broad-scale clinical testing to potentially treat target populations. Although it holds great promise, realizing the benefits of personalized medicine presents significant challenges for life sciences companies, particularly surrounding the integration of research and clinical data—both within and between life sciences companies and health care organizations.
The breadth and depth of data available electronically from health care providers, pathology labs, genetic diagnostic labs, and other research institutions is exploding. The challenge is to organize and integrate this clinical information with R&D data to guide research—as valuable insight can emerge when phenotypic data are combined with pathology and genetic data in the context of a specific disease. To ensure an economical and scalable personalized medicine model, pharma R&D will have to link more closely to clinical care delivery. Informatics—from the perspective of the health care organization and pharma—is at the forefront of surmounting this challenge.
Clinical R&D data are fragmented across many silos. As they engage with academic medical centers (AMCs) to develop new personalized treatments, life sciences companies are looking toward standards-based platforms that facilitate integration of internal and external data sources. Such a platform can provide “vertical” informatics capabilities around specific drugs (from discovery, through clinical trials, to post-market surveillance) as well as “horizontal” informatics capabilities required for portfolio management across the pipeline. Much of the data required to support informatics around costs and outcomes lie within provider organizations. Hence, the architecture for data integration for life sciences companies must support internal R&D data models, as well as standards-based data models within provider organizations.
The most ambitious frontier in R&D informatics focuses on the capture and use of scientific knowledge within an enterprise. While companies routinely consolidate drug pipelines through mergers and acquisitions, the knowledge and insight within the merging companies have been difficult to consolidate and leverage. Web 2.0 tools promise to simplify semantic information sharing in an enterprise. Ontology-based search engines combined with natural language processing capabilities can help researchers find and correlate relevant scientific insights buried across multiple data sources, taking content management and data mining to a whole new level.
Health Care Informatics
As diagnostics and therapeutics become more closely tied in the personalized medicine world, the resulting informatics needs are becoming more pervasive within health care organizations. There are three levels of informatics capabilities that health care organizations require to support the transition to personalized medicine:
In-silo Informatics, which focuses on clinical specialties, such as oncology. These capabilities integrate specialized clinical data with relevant content to improve the evidence base for the specialty and guide specific care protocols.
Cross-silo Informatics, which supports disease management and enables health care organizations to track outcomes. These capabilities are most valuable for primary care, emergency care, and chronic disease management, as well as infection control and public health.
Performance and Cost Informatics, which enable cost-benefit analysis of personalized treatments. With targeted treatments serving smaller patient populations, current pay-for-service reimbursement models may prove to be ineffective.
Integrating the applications and data required to support the informatics required for personalized medicine is a challenge for many life sciences companies. Mergers and acquisitions further complicate the IT landscape by introducing more heterogeneity. Over the next decade, the drive to implement informatics capabilities will surely lead to a major overhaul in the way life sciences companies manage and utilize R&D information. Organizations leading the charge must be careful to ensure that incremental investments in applications and infrastructure also provide direct business value along the way, while adhering to a long-term road map. Those that succeed stand to transform the treatment of disease while gaining new levels of productivity within their own R&D programs.
Neil de Crescenzo is senior VP and General Manager, Oracle Health Sciences. He can be reached at firstname.lastname@example.org.
This article also appeared in the March-April 2009 issue of Bio-IT World Magazine.
Subscriptions are free for qualifying individuals. Apply today.