Harold Varmus Outlines Goals for Precision Medicine in Cancer

February 26, 2015

The concept of precision medicine is not new, said NCI director Harold Varmus, MD’66, during his talk last week as part of Columbia University’s Distinguished Lectures in Precision Medicine, but the ability to sequence an entire genome for under $2,000 and advances in computing should allow precision medicine to make faster progress in the 21st century.

Dr. Varmus told the audience he foresees “a large computationally based network of information that informs a different way to characterize and name diseases. A taxonomy built on a new set of facts about the origins and mechanisms of disease that then prompts more accurate diagnoses, better treatments, better outcomes—over what I envision will be many decades.”

Much of the progress being made in precision medicine today is in cancer, which will account for about a third of President Obama’s new Precision Medicine initiative.

The success of imatinib (aka Gleevec), a drug for chronic myeloid leukemia that was designed to hit the cancer’s genetic driver, is often cited as precision medicine’s “poster child.” The drug can restore a normal life expectancy to a middle-aged patient.


“That’s quite a remarkable feat, but don’t be misled by this,” Dr. Varmus warned. “In every other cancer that we’ve looked at in the last decade and a half, there’s a much more complex array of mutations. And the effects of targeted therapies are unpredictable and often last only a short period of time.”

Several things must be done to make other diseases as treatable as chronic myeloid leukemia.

“We’re not done characterizing tumors at a genetic level,” he said. “We’d like to be able to identify the common genetic drivers of each tumor type, and we’re not there yet. We still have a few thousand more tumors to look at.”

Better ways of diagnosing cancer and monitoring treatment are needed, perhaps with a “liquid biopsy” that can detect cancer DNA that is circulating in a patient’s bloodstream.

And new analyses of clinical trials are required that can account for the multiple subtypes within a single type of cancer.

Despite the challenges, the precision medicine approach to cancer has its successes, Dr. Varmus said. “In lung adenocarcinoma, for example, the array of mutations is complex but we can see that there are some important commonalities. When those genes are studied in depth, it’s been possible to identify a number of therapeutic agents that have had a beneficial effect in patients with widespread disease.

“Tremendous benefits have accrued from carrying out genetic analyses of this tumor type, which 10 to 15 years ago was nearly impossible to treat effectively once the disease expanded beyond the surgeon’s reach.”

Dr. Varmus appealed to researchers at Columbia and other universities to help with the legal and economic issues of precision medicine, including analyzing the costs and benefits of genetic testing and re-imagining patient-consent processes to encourage more cancer patients to share their health and genetic data for the benefit of future cancer patients.

“We like the idea of telling every cancer patient that they are potential cancer information donors,” Dr. Varmus said. “Cancer is one of our society’s greatest problems, and this is an act of altruism that will help everyone.”


Harold Varmus