In 2011, a patient newly diagnosed with type II diabetes will likely receive the same drug, metformin, that was first approved in 1957. What’s more – in the absence of information on the underlying molecular mechanisms of her particular case – that patient will likely end up taking a one-size-fits-all pill rather than a drug tailored for her unique disease subtype. Nor are there tests available that would allow her family members to pinpoint their genetic-based risk for the disease.
A new report issued today by the National Academy of Sciences, aims to change all that. Requested last year by Francis Collins, director of the National Institutes of Health (NIH), the report was intended to advise Collins on the need for, and feasibility of, a whole new classification of human diseases based on molecular biology. Today, the committee responded by saying that not only should a new taxonomy of diseases be developed, but that making it happen will require a vast network of databases integrating emerging molecular research with clinical data from patients. In an era of electronic medical records, the report says, such a goal is a achievable and the results would be transformative.
In its 66 pages, the report (you can read a briefer version of it here), concludes that the new classification system emerging from such a network would define diseases by their underlying molecular causes and drastically improve scientists’ access to patients’ information even while they are being treated. It adds that legal safeguards, like the Genetic Information Nondiscrimination Act, should preserve patient privacy, but that the highest sensitivity to the privacy issue, which will “make or break” the initiative, must be used.
“It’s more than a tweak,” to existing, piecemeal efforts to integrate patient data with ballooning reams of genomic information, said Susan Desmond-Hellman, a committee co-chair who is chancellor of the University of California at San Francisco. “It’s a big undertaking.”
“The main theme here is integration,” of biomedical research data with patient data collected at the point-of-care, said committee co-chair Charles Sawyers, an oncologist at Memorial Sloan-Kettering Cancer Center in New York City.
Speaking to reporters at a press briefing in Washington DC, Sawyers and Desmond-Hellman added that the committee didn’t envision the network requiring new money, but redirection of existing resources to systematically collect information that is already being generated. “What was striking to us is how much information is really left unused at the point of clinical care,” said Desmond-Hellmann.
The effort the committee envisions is one of decades and not years, and of cooperating constituencies from pharmaceutical companies to insurance providers to patients and physicians and scientists. At its foundation would be an “Information Commons,” populated by data from huge populations of patients, ranging from genome sequences to medical records including signs and symptoms, test results and family history – and all of this not just at a single point in time, but over the history of a patient’s illness. A further “Knowledge Network of Disease” – essentially making sense of the data in the Information Commons by integrating them with evolving knowledge – would then allow for the new, molecular classification of diseases.
The committee recommends that NIH launch pilot projects that would begin to populate the”Information Commons” with data collected in the ordinary course of clinical care. The data, it writes “must be broadly accessible so that a wide diversity of researchers can mine them for specific purposes.” It concedes that this may require an “evolution in the public’s expectations” with regard to privacy and access to their health care data.
But, said Sawyers, “there was a consensus on the committee that current privacy protections are too extreme.” Helped by new health insurance and genetic information privacy laws, the committee writes, “there is little evidence that the public has the extreme sensitivity toward genetic data that many researchers anticipated 25 years ago.”
As for the next 25 years, if all goes according to plan, the newly-diagnosed diabetic patient will have a medication targeted to the molecular mechanism of her disease; a test for family members to assess their risk; and a physician who, with the click of a mouse, will be able to enter the specific information in that patient’s electronic medical record into an Information Commons that grows more robust, and more illuminating, with every passing day.