Medicine Gets Precise

HMS experts discuss the future of medical care and the power of getting personal 

From left, Martha Bebinger, George Church, and Isaac Kohane. Photo by Aaron Washington

From left, Martha Bebinger, George Church, and Isaac Kohane. Photo by Aaron Washington

Precision medicine. It’s one of the hottest topics in modern health care, but for many in the general public it remains a somewhat murky term.

What does it really mean?

During HUBweek, a series of events aimed at celebrating “the world-changing work, art and thinking being imagined and built in Greater Boston,” George Church, HMS professor of genetics, and Isaac Kohane, head of the HMS Department of Bioinformatics—both key figures in this emerging field—were asked to lend clarity and dispel some of the mystery around precision medicine.

“I would say this is what will make us healthier, live longer and with more vitality. We will do this by measuring more than we’ve ever measured, from the genome to the microbiome to social networks,” said Church.

It constitutes a departure, he added, from traditional physician practice.

For the uninitiated, precision medicine can be loosely defined as a movement away from a one-size-fits-all health care approach to one that uses the latest computational advances to assess each individual patient’s genetics, family history and any relevant environmental considerations—all with the hope of providing far more precise therapies.

Kohane compared it to Netflix.

Netflix, he said, knows you. It knows the entire history of what you and viewers like you watch. Thus, it can accurately predict the nuances of all your entertainment preferences.

Comparatively, he said, today’s doctors know very little about their patients. Data is available, he said, but it should be distributed more widely in order to precisely stratify patients into specific groups.

Why did it take us until 2015 to fully grasp this?

“Medicine is a conservative practice,” Kohane said.

Martha Bebinger, a health care reporter for NPR-affiliate WBUR in Boston, moderated the discussion between Church and Kohane before a crowd of more than 200 at the HMS Martin Conference Center during the eight-day HUBweek, sponsored by MIT, Harvard University, Massachusetts General Hospital and the Boston Globe.

Bebinger asked Kohane if there was a particular “oh my God” moment that led him into the field of bioinformatics.

“The whole point of this is to escape the fate of our ancestors.”—George Church

Kohane replied that for him that moment began 30 years ago and hasn’t let up yet. After completing graduate studies in computer science and beginning his medical training, he said, he was shocked at how alien the whole idea of computation and analysis was to medical diagnosis.

To Sequence or Not to Sequence

Bebinger asked if it’s a good idea for people to get their genomes sequenced today, given that computer platforms that can analyze and make sense of this data have not been fully developed.

Here was perhaps the only point in the evening where the Church/Kohane discussion verged on sparring.

For Kohane, the answer depends on how serious a patient’s condition is. If she has problems that are not being answered by general medicine, he said, then it might make sense to get the information. The problem is, he said, that we don't know what much of the genome means.

“George will disagree,” he added.

“I will,” Church said, “but in a very supportive way.”

For Church, the more genomes sequenced, the better. He said that excellent software for genetics does exist and it is probably adequate enough to provide useful information for genetic counselors.

He said increased sequencing may help shed light on the thousands of diseases that many people might unknowingly be carrying genes for, diseases that are individually rare but collectively common.

And of course, he said, there are genetic components to certain kinds of actionable cancers, such as breast, ovarian, stomach and colon cancer, and more sequencing may help increase understanding on how to treat them

Bebinger asked how long precision medicine will likely be in this transitional period that every emerging field at some point occupies, where the new technologies are not yet fully embraced.

Speaking specifically to the issue of genome sequencing, Church said we are at a tipping point.

“We’ve removed the barrier of cost. Next barrier is education,” he said, citing the early days of the Internet as an example.

Back then, people needed just one reason to get on the web, and when they did, they realized that there were thousands of reasons to do it. Once the cost became affordable, the remaining curve was purely educational. That is where we are now regarding genome sequencing, Church said.

Privacy concerns, he emphasized, are not what is holding people back.

Kohane agreed, and further stated that for the few people who are concerned with privacy, if they really understand how much their privacy has already become dissipated by existing medical records they would be far less concerned.

“The whole point of this,” Church reminded the audience, “is to escape the [genetic] fate of our ancestors.”

While the cost of genome sequencing itself has plummeted, the price of targeted therapies has not. They remain prohibitively expensive.

“We will be in big trouble for a short time,” said Church.

“The most personalized drugs are the most expensive. The costs will come down and there will be alternatives,” he predicted. “We saw huge drops in diagnostics.”

Alluding to the recent cost spike for the generic drug Daraprim, Kohane pointed out that it’s clear you don’t need precision medicine to drive prices up.

“Market forces must intervene,” he said.

At the conclusion, Bebinger asked Kohane, “What are the top three things we need to happen in order to move forward?”

“First, physician education,” he said. “They need to learn practical genomics. Two, we need to bring real, practical computing to the point of care. And three, sharing. Broad sharing of data, so we can find out what all this genomic data means.”