Creating long term value and competitive advantage in the life sciences industry requires companies to maintain a constant flow of innovative drugs into the market. Discovering and bringing a new drug to market can cost hundreds of millions of dollars and take a decade or more of research and development. Although the industry has been able to increase the number of new chemical entities (NCEs) that may have therapeutic value the vast majority (70-90%) fail Food and Drug Administration (FDA) test requirements for safety and efficacy. And many of these failures are not identified until the later, more costly phase of clinical human trials making drug discovery a very expensive trial and error process. I believe life sciences companies can use business intelligence and information management technologies to reduce the time and cost needed to identify and select chemical compounds with the greatest therapeutic potential and commercial value.
Drug discovery begins with scientists identifying cellular and genetic factors called biomarkers that play a role in specific diseases. Unfortunately at many life sciences companies well funded divisions and functions make their own IT decisions, resulting in inconsistent processes, redundant systems and fragmented data. Data warehousing and business intelligence technology can help life sciences companies increase productivity in the drug discovery process through improved accessibility and collaboration across divisions and functions.
Furthermore, information about biomarkers resides both inside the corporate boundaries and in external sources such as academic and government research. Although tools like the Basic Local Alignment Search Tool (BLAST) are freely available in the public domain to facilitate analysis of gene bank databases, I believe life science companies can increase productivity in finding relevant information from unstructured sources such as documents, email, medical literature, and patent registrations through the use of search, text mining, semantic technologies and content management systems.
The next step in drug discovery involves identification of NCEs that target biomarkers and are likely to have a therapeutic effect and selecting leading candidates for further development. In this phase large amounts of data are generated about how biomarkers react to NCEs, as well as the molecular pathways involved and mechanisms used by the NCEs to produce a therapeutic effect. I believe data mining and visualization technologies can help companies to better model the effectiveness and safety of drugs, and predict how different subsets of patients will respond to a particular treatment. This analysis helps companies decrease costly late stage failures, speed time to market and improve product quality by focusing limited development resources on NCEs with the greatest potential for success.