Rewriting the Textbook on Disease

BY Jason Kane  November 4, 2011 at 11:13 AM EDT

Imagine if diabetes could be peeled apart — if a doctor could look at test results and instantly see the genetic and environmental factors that triggered and fueled the disease.

Looking at a patient’s medical history and genomic information, it may soon be possible to prescribe a tailored treatment plan, recommend ways to avoid kidney failure, blindness and amputation — even determine if the patient’s siblings and children are likely to develop diabetes later in life.

The first step toward that kind of complex, individualized care might simply be a change in the way we sort and classify disease. A new report from the National Research Council is calling for a “new taxonomy” that would define diseases more precisely by their underlying molecular causes rather than their traditional physical signs and symptoms.

Dr. Susan Desmond-Hellmann, co-chair of the committee and chancellor of University of California, San Francisco, believes the new definitions and molecular-level view of disease will lead not only to more effective care but to new research, more collaboration and the development of innovative drugs.

There’s reason to believe she might be right. As the former head of product development at Genentech, Desmond-Hellmann is credited with bringing the first gene-targeted therapy drug — Herceptin for breast cancer — to the public more than a decade ago.

She sat down with NewsHour Correspondent Hari Sreenivasan to discuss how the simple act of re-categorizing how we think of disease could lead to better outcomes for patients.



At the moment, physicians refer to the International Classification of Diseases — a 100-year-old system used to track and diagnose disease and determine reimbursement for care. It’s a cumbersome taxonomy based mostly on signs and symptoms, and one not well-structured to incorporate new information on the molecular basis of disease.

Outside the laboratory, that affects diagnosis and treatment.

In the time it takes for biomedical research information to make its way to doctors and patients, wasteful health care expenditures are carried out for treatments that are only effective in specific sub-groups.

Classifying disease by particular molecular pathways might also encourage the sharing of information between scientists who study the same molecular structures but different diseases — people who traditionally have not known of each other’s findings.

All of this research-swapping would take place through an “information commons” that links molecular data, medical histories, and health outcomes to individual patients.

New data would be continuously funneled into the network from the research community, enriching the reference materials that doctors use to diagnose and treat disease. At the same time, biological information would be extracted directly from the medical records of consenting patients and used in future research.

“We need to embed our scientific enterprise into the normal course of clinical care,” Desmond-Hellmann told Sreenivasan. “Imagine a world where it isn’t like, ‘Here’s where you take care of patients and over here — in a completely separate place — is research.’ We’re saying, ‘Take that research and embed it in the normal course of clinical care.’”

Essentially, research and treatment would truly become complimentary sides of a two-way street.