The pathway informatics sector got a boost last week at IBC’s Drug Discovery Technology 2004 congress, where two high-profile keynotes emphasized the plethora of benefits that a better understanding of molecular pathways will bring to the pharmaceutical industry, and numerous companies highlighted their products in this area.
In his keynote address, National Institutes of Health Director Elias Zerhouni said that the “Pathways to Discovery” component of the NIH Roadmap initiative will be an important step toward “understanding the pre-clinical molecular events associated with disease” that will eventually help detect at-risk patients before symptoms appear. One key to this initiative, Zerhouni said, will be a new way of looking at biological data. This “biological data of the future,” he said, will need to be quantitative, multidimensional, and cumulative, with high temporal resolution and high data density in order to accurately represent the complexity of biology and of biological networks.
In a second keynote, Mark Fishman, president of the Novartis Institutes for Biomedical Research, recommended that all biomedical knowledge be organized around the framework of molecular pathways — a method that he said corresponds well with the goals of the pharmaceutical industry because pathways are essentially “collections of druggable targets.” Fishman stressed the need to “functionalize the genome for pharmaceutical discovery,” and noted that pathways represent a logical means to do this because “you can reduce the genome from 22,000 genes to around 100 key pathways.” Eventually, he said, the field will be able to “organize disease by pathway, not pathology.”
Such endorsements were music to the ears of pathway informatics tool shops at the conference. Frank Mara, senior vice president of marketing at Ingenuity, told BioInform that he was pleased to see the industry enthusiastically embrace an approach that was not well understood even a year ago. “This is really good timing for us,” he said.
Indeed, Ingenuity arrived at DDT fully outfitted with several product upgrades, as well as the results of a return-on-investment study for its platform conducted by IDC’s Life Science Insights, and the news that it had extended its collaboration with Millennium Pharmaceuticals for another three years. Other pathway informatics firms also came prepared with new products and client wins. GeneGo, for example, launched the latest version of its MetaCore platform at the conference and announced that Bristol-Myers Squibb had taken a license for the technology. Julie Bryant, vice president of business development and sales for GeneGo, told BioInform that “things are really taking off” for the small firm, particularly within the pharmaceutical sector. A number of companies, she said, have requested that GeneGo provide just its data without the MetaCore analysis front end, a possibility that she said the company is considering.
Genstruct, meanwhile, had a poster demonstrating a curated, computable model for the LNCaP human prostate cancer cell line that it built with more than 55,000 nodes, 115,000 interactions, and 10,500 cause-and-effect relationships. The company is using the model to interpret microarray experiments in a partnership with the Dana-Farber Cancer Institute. Genstruct CEO Keith Elliston said that the LNCaP model represents only a fraction of the company’s accumulated cause-and-effect relationships, which number around 30,000. Elliston said that Genstruct has several pilot projects underway with pharmaceutical firms.
This activity follows on a similar flurry of pathway-related projects highlighted at ISMB 2004 earlier in the month [BioInform 08-09-04], and served as further evidence that the pathway informatics sector is heating up. Fishman cautioned, however, that a pathway-based approach is not a cure-all for the pharmaceutical industry — or the ailing informatics sector. Some of the perhaps-insurmountable “complexities” in the field that Fishman highlighted include the “variable roles” that pathways can play in the same tissue, feedback loops that result in unforeseen changes after stimulation, and the intersection of multiple pathways in a network. Fishman called for “new molecular imaging probes” and other experimental tools to help resolve some of these complexities.
Nevertheless, it appears that the pharmaceutical industry is open to using the pathway tools that are currently on the market, while vendors are reciprocating with improved methods. Ramon Felciano, CTO of Ingenuity, told BioInform that the company was actually a bit surprised at the overwhelmingly positive results of its ROI study, in which respondents reported that they were able to increase their productivity by an average of 10 percent while reducing failures by an average of 33 percent. Felciano said that the study serves as proof that a pathway-based approach can resonate “in every phase of the drug discovery and development pipeline,” including preclinical and clinical testing, FDA review, and even aftermarket studies exploring new indications for approved drugs.
“We’re seeing use of these tools in all stages of the pipeline, not just target identification, which is where we originally set our sights,” Felciano said. “This is not a silo technology.”