U.S. DEPARTMENT OF ENERGY HUMAN GENOME PROGRAM, ORNL.GOV/HGMIS
Science and medical journals are so 20th century.
The Internet changes everything, they say. Well, maybe not everything, yet. The number of articles in medical and scientific periodicals is still fundamentally a product of the number of paper pages funded by the publishing entity. And subscription-only access has proven to be virtually unyielding, despite the leadership of open-access pioneers BMJ, JMIR, Medscape, BioMedCentral, PubMedCentral, and PLoS. The much criticized peer-review system remains virtually intact since its adoption in medical journals in the 18th century.
But now, mature Internet technologies coupled with exploding genomic data may lead to inexorable transformation in scientific publishing, parallel to the change that is rocking biomedical disciplines themselves. Young companies are challenging publishing giants, providing innovative ways of measuring an article’s import, filtering, and redefining what constitutes a “paper,” with the idea of publishing smaller bits of data faster.
Alexander Pope said that “the proper study of mankind is man.” Our rapidly changing knowledge about the uniqueness of my genome tells me that “the proper study of me is me.” And the burgeoning genetic, molecular, and biochemical knowledge about the workings of individual cells suggests that the proper study of my cancer is my cancer. This is the essence of the new era of personalized medicine, championed by Francis Collins, director of the NIH, and now a focus of the $500-million-a-year Patient Centered Outcomes Research Institute, funded by the US 2010 health care reform bill. Adding the patient to the equation of decision-making is a revolutionary step and has begun to advance the practice of medicine and the definition of evidence. What are the publishing processes that will best support these developments?
Cancer is hundreds or thousands of unique molecular diseases.
Certainly, technologies that enable social media should, with reasonable safeguards, become central to scientific communication. It was the notion of blending these technologies with those surfacing in the genomic revolution—in particular the idea that cancer is hundreds or thousands of unique molecular diseases demanding targeted and combination therapies—that spurred the formation of Cancer Commons, a 21st-century, nonprofit publishing platform.
Cancer Commons is a hub of open-science communities in which physicians, patients, and scientists collaborate on models of cancer subtypes to more accurately predict individual responses to therapy. “Verticals” will be developed one at a time, starting with melanoma (now live). We’ll match patients with the best possible individualized treatments based on genomic information, and record and disseminate outcomes as soon as they are available. At the core of each vertical is a “molecular disease model” (MDM) that defines subtypes of the cancer, each hyperlinked to pathways, diagnostic tests, approved and experimental therapies, and clinical trials. Subtypes and therapies are continually revised based on new tumor types and outcomes. The MDMs thus serve as living review articles, maintained and updated by cancer experts.
Additional planned components of the platform are a range of clinical observations, data, case reports, treatment hypotheses, and actionable research findings—some peer-reviewed and some not-ready-for-prime time information that may be helpful to late-stage patients. Some of this content will be validated and incorporated into guidelines as standards of care; other hypotheses and observations will be rejected or revised as new data points appear.
Technology challenges include developing a giant adaptive search engine and semantic wiki to capture and analyze the “N-of-1” experiences of tens of thousands of physicians and hundreds of thousands of patients with their targeted therapies. Delivering this information will be many new and yet-to-be-invented social media tools that maintain patient privacy while providing highly specific details. Apps will be developed, such as the newly launched Melanoma Therapy Finder, a freely available tool for patients and doctors.
Naturally, the many challenges of this venture will be met only by engaging the larger cancer research community, mainstream oncologists, and their patients. The success of 21st-century publishing requires high-decibel tweets.
George D. Lundberg MD, is Editor-in-Chief of Cancer Commons, Editor-at-Large of MedPage Today, and a consulting professor at Stanford. He was Editor-in-Chief at the AMA for 17 years, overseeing 39 journals, where he also served as Editor of JAMA. After leaving the AMA, he founded The Medscape Journal of Medicine, and was Editor-in-Chief of eMedicine from WebMD.