Firm hopes Big Data can personalize health care
GNS aims to help MDs know which treatment will work the best for each patient
When Colin Hill’s father was diagnosed with later-stage prostate cancer last summer, he was treated the same as every other patient with the illness.
This standardized approach bothered Hill, who believes medicine should approach each patient’s illness as unique, with medication tailored to the person’s history and biology.
“You show up to the hospital, and it’s like Groundhog Day,” Hill said, with patients being cared for the same way, over and over again. “It’s this outdated standard of care created for this hypothetical average patient. But no one’s an average patient.”
A genetic analysis of the tumor in his 69-year-old father, Foster Hill, found he had a genetic variant of the cancer that does not usually respond well to the hormone therapy Lupron, the current standard of care. But not knowing what else would work, doctors gave Foster Hill Lupron anyway. Luckily, the treatment seems to be helping, and his father’s outlook is much improved.
Hill hopes a Kendall Square company he founded 13 years ago, GNS Healthcare, will eventually improve medical care for his father — and for countless others — by providing personalized treatment. GNS is among the leaders in using Big Data analytics to learn more about diseases, patients, and treatments.
The company is deploying enormous computing power to produce a more complete understanding of treatments for rheumatoid arthritis, diabetes, cancers, and other illnesses.
For example, it is working with the Dana-Farber Cancer Institute and Mount Sinai Medical School to build a computer model of multiple myeloma, so researchers can better understand what works well for patients today, as well as develop more effective treatments for the blood cancer. It is involved in a similar collaboration with Brigham and Women’s Hospital and several other partners to learn more about multiple sclerosis.
Harvard Medical School recently agreed to use a GNS computing platform to analyze how cells replicate or transform into different types, for insights into conditions such as cancer and neurodegenerative diseases, Hill said.
And the company has a partnership with the Centers for Medicare & Medicaid Services and Health Services Advisory Group to assess health care quality measures, as well as other recent deals with pharmaceutical companies, hospitals, and advocacy groups, and the insurance giants Aetna and Blue Cross Blue Shield.
“It’s exciting times for us,” Hill said, after 13 years of developing his approach to analyzing health care. “We are now in the thick of things.”
Hill did not always have such an absorbing interest in science. He went to college at Virginia Tech — mainly to play tennis. “I was more serious than I was good,” he quipped.
But while there, he became fascinated by physics and chaos theory — the idea that complex patterns could result from simple rules.
His imagination was stoked by a summer job in 1996 at Santa Fe Institute, an interdisciplinary research center focused on highly complex issues. Hill then went on to graduate from McGill and then Cornell University.
By then, the Human Genome Project was becoming a reality, capturing the attention of many scientists, including a young Hill.
“That’s when two and two came together,” Hill said. “It was like ‘Oh yeah, the stuff we’re doing, though it’s pretty theoretical, it is going to be the thing that links these pieces together,’ ” including chaos theory, genetics, Big Data, and health care.
Now he’s in the middle of the so-called Big Data revolution in health care. Companies such as GNS have an enormous capacity for crunching troves of information on patients, diseases, and medical outcomes collected by medical providers, insurers, and other big players.
Hill likes to say he wants GNS to capture the “data exhaust thrown off” by every interaction a patient has with the health care system, from the doctor’s office to the hospital to the pharmacy. A bad reaction to medication is a data point worth having; ditto for other side effects, as well as results of all kinds of procedures and interventions — bad or good.
While such data sets are getting easier to find, Hill said, genetic information is still too expensive to be truly useful.
“What we don’t have yet for my dad and for other men with prostate cancer is a large coherent set of data on prostate cancer patients that includes the molecular level,” Hill said.
With information aggregated from thousands of cases, Hill said, GNS uses artificial intelligence algorithms developed out of chaos theory to determine what treatment made the crucial difference for each patient, and with it what is likely to work best for the next patient — rather than simply trying one medication after another, as is often done today.
“What we’re trying to get at is not just patterns and trends, but reverse engineer the mechanisms that gave rise to the data,” he said. “We’re trying to find the cause-effect relationships within the data.”
This ability to predict results is what sets GNS apart, said Dr. Atul Butte of Stanford University, one of the academic leaders of the Big Data movement. That’s the “nifty part of their technology,” Butte said.
Alexis Borisy, a partner at the Boston life sciences venture capital firm Third Rock Ventures, said GNS is in the vanguard of Big Data companies analyzing health care information.
“They’ve had a chance to learn, refine, and they’ve kept with it so they’ve had a lot of experience to build on,” Borisy said.
He and Hill have known each other since they were young business executives more than a decade ago, and he said he has tremendous respect for Hill’s intelligence, persistence, and communication skills.
“He is one of the visionaries in the space,” said Borisy, who also serves as chairman of Foundation Medicine, a molecular information company, and interim chief executive of Warp Drive Bio, both of Cambridge. “I think it’s fair to say he’s kept GNS growing and building by the force of his personality and the force of his efforts.”