The road toward personalized medicine still has a few gaps that need to be bridged before DNA-targeted therapies and diagnostics become the new standard of care.
Perhaps the largest gap to overcome is a uniform system of biobanks.
Biobanking is "one of the biggest levers to accelerate" pharmacogenomics, said Michael Svinte, vice president of Information-Based Medicine, of IBM Healthcare and Life Sciences.
Svinte said biobanks, also called biorepositories, are "the fuel for personalized medicine, bringing together the biological samples, as well as the data associated with those samples."
Researchers can access from biobanks actual biological samples and data, and obtain molecular and corresponding phenotypic information. Integration and analysis of those data could lead to focused drug development efforts, such as using biomarkers to figure out why one person's response to a certain drug or dosage level is different from another's, or determining genetic or environmental disease factors that could lead to preventive medicine, or enhancing the ability to discover cures.
While there are a number of biobanks in existence - such as government-sponsored biological respositories, consortiums specializing in particular diseases and large biobanking efforts, such as UK Biobank and the Karolinsa Institute Biobank - there lacks a consistency regarding how and what data and samples are collected.
Svinte said that as IBM was talking to different leaders about biobanking, it "became very apparent to us that there was a core set of issues and challenges a lot of people were wrestling with."
In order to coordinate biobanking efforts, the company hosted two World Wide Biobank summits in 2004 and had its third event last month in Stockholm, Sweden, to discuss the challenges in establishing biobanks and standardizing biobanking procedures.
Summit Probes Biobanking Challenges
The biobank summits are part of the initiative started by Information-Based Medicine, a start-up unit created from IBM's Emerging Business Opportunity (EBO) program. EBOs were established in response to shifts in technology or business models, and Information-Based Medicine is designed to aid personalized and predictive medicine, Svinte said.
The meeting looked at a host of issues, including the initial problem of money. Establishing biobanks requires seed funding, and it is not clear where that will come from. But one idea that came out of the summits eventually is transitioning to an access-fee system, in which companies pay for the information they need.
Other issues center around management organizations and whether ethics boards are needed. Other challenges are establishing a pecking order for companies and researchers to obtain biological samples, plus there is a lack of biobanking legislation and no public education to gain the support of doctors and patients. With a handful of exceptions - Sweden, for instance, has well-established biobanking laws - there are few policies in place, and those vary from country to country.
"There's a lot of work out there still," Svinte said.
Developing standards also will have to include quality assurance from a scientific perspective, so that the data and the quality of the data are useful to researchers gaining access.
Although information technology is an essential factor for accruing biobank data, "personally, I don't think information technology is going to be one of the key inhibitors here," Svinte told BioWorld Today. "To an extent, we've already solved many of the needs," including computer capability, molecular modeling, storage and grid capability.
There are obstacles, to be sure, but the benefits of establishing national or global biobanks are obvious, at least from a research perspective. Biobanks represent a resource that could lead to a better understanding of gene interaction, the effect of environment and lifestyle on disease, and a method for translating that information into clinical practice.
Plumbing The Salt Lake Gene Pool
The unit of IBM also works with other areas of genomics. It collaborates with pharma companies on pharmacogenomics efforts, including infrastructure for clinical trials and patient stratification.
On the patient side, efforts focus on the translation of clinical information to assist research and patient care. Last month, IBM signed a collaboration with the University of Utah Health Sciences Center and Salt Lake City-based LineaGen Research Corp. to create a clinical genomics infrastructure designed to accelerate the molecular understanding of complex diseases for developing molecular diagnostics and targeted treatments.
Researchers will use the Utah Population Database and Utah Genetic Reference Project Database, which includes more than 7 million records comprising pathology, pharmaceutical history, imaging, diagnosis, demographic information and other data from regional population groups. The collaboration will begin by focusing on rheumatoid arthritis and other autoimmune diseases.
"There's a lot happening right now in the space," Svinte said. Recently the FDA published its guidelines for the industry on pharmacogenomics, providing a framework for submitting data to supplement marketing applications and regulatory review. (See BioWorld Today, March 24, 2005.)
"The point is that we have so much data right now," he said, "that we need to integrate it and mine it, so that it will provide actual insight that will help advance the personalized medicine area."