all the information, none of the junk | biotech • healthcare • life sciences

Ready or Not, Genomics is Coming to a Hospital Near You

Xconomy National — 

First came genomics for scientists. Then came genomics for Big Pharma. But this year, at a faster pace than many industry insiders anticipated, we’re seeing genomics enter the world of real-life care for patients.

This idea would have sounded like science fiction a couple years ago. It took researchers 13 years and about $3 billion to complete the Human Genome Project, to come up with one complete genome that carries the instructions for making cells. A decade later, companies like Illumina (NASDAQ: ILMN) and Life Technologies (NASDAQ: LIFE) are in a fierce competition to sequence genomes at industrial scale, for as little as $1,000 apiece in a single day.

This breathtaking advancement is enabling scientists to ask questions they never could have dreamed of asking a few years ago, about what makes people different and what causes disease. A deeper understanding of the human genome paved the way for two important personalized cancer drugs last year from Roche and Pfizer, and it is helping guide many more therapies in the works.

It’s heady stuff, moving forward at an incredibly fast pace.

Much of this genomic advancement makes for good headlines, but is still quite remote from the average patient on a visit to see their physician. But this year we are seeing an increasing wave of demand emerging from hospitals, clinics, and leading physicians who want to take advantage of the new fast and cheap sequencing tools to see if they can be used to help patients.

Steve Jobs and Christopher Hitchens were high-profile examples last year of cancer patients who got their genomes sequenced, in the hope of learning more about the genetics of their tumors, in order to better fight them. And even though there were no miracles to report on in the Jobs and Hitchens cases, the movement toward “medical genomics” or “clinical genomics” has started to take off. Cambridge, MA-based Foundation Medicine, which performs DNA sequencing on an array of 200 genes implicated in cancer, has seen a surge of demand among physicians ordering its test and summary reports, even before the company was ready to begin marketing. Cambridge, MA-based Knome, which provides genomic analysis software as a service, is announcing this week that it has gotten so much demand from physicians that it is beginning to offer a new version of its software geared for physicians treating patients.

Consider the story at Foundation Medicine. This small company, backed by Kleiner Perkins Caufield & Byers and Google Ventures , among others, got set up with its first commercially licensed DNA sequencing lab late last fall. The company wasn’t in position in the early days to receive large volumes of tumor samples, sequence them, and deliver useful reports back to clinicians. The company was, and still is, generating the bulk of its revenue from providing its service to drugmakers like Novartis, Sanofi, and Celgene, who use it to help inform their drug development strategies, CEO Mike Pellini says.

But even without actively promoting its service to physicians, Foundation learned that demand is increasing. More than 175 oncologists ordered genetic tests from Foundation Medicine for about 400 different patients before the company began active marketing, says Pellini, a former GE Healthcare executive. Foundation now expects that number to grow, as it officially rolled out its tumor-sequencing service earlier this month at the American Society of Clinical Oncology (ASCO) meeting in Chicago.

Foundation Medicine CEO Mike Pellini

“A couple things have surprised me in the past year, and at the top of the list is the clinical demand we’re seeing,” Pellini says. “We always expected this approach would ultimately be adopted by oncologists in the academic and community settings. But the speed of adoption appears to be faster than many of us anticipated.”

Like Foundation Medicine, Knome is also seeing demand grow from doctors. Some of the country’s leading institutions, like the Mayo Clinic, Johns Hopkins, and MD Anderson Cancer Center are making aggressive pushes to incorporate genomic screening into care of cancer patients. Without an active marketing effort, Knome CEO Martin Tolar says his company has had inbound requests from 50 different hospitals that want to start sequencing patients’ genomes, and who need help from software to interpret these massive files with 6 billion DNA datapoints per person.

It’s not just Mayo and Hopkins, he says, but also regional hospital systems in Pennsylvania, Virginia, and elsewhere. Some of the hospitals are thinking big, about analyzing thousands or tens of thousands of genomes per year, Tolar says.

To deal with the surge in demand, Knomeis rolling out a new version of its software this week specifically for clinicians, mainly for those who treat cancer patients and genetic disorders in children.

Knome CEO Martin Tolar

“It has been incredible, like a firestorm,” Tolar says. “Last fall, I thought it would be 3-5 years before the clinics would be ready for that. They are saying to us they want it now.”

There is good reason to be excited about genomics, but this is the point in the story where everybody should just take a deep breath and come to their senses. Just because we can sequence the genome of a patient’s tumor, and compare it to the genome from healthy cells, doesn’t mean we get any information that a doctor can use to save someone’s life.

It’s possible that scanning the genome could lead a physician to a bright new insight on how to treat an individual’s cancer, and an individual might even be lucky it would work. But such tests might also cause physicians to try new regimens that don’t work, and actually cause harm in a dying patient’s final days. People shouldn’t get too exuberant with some false sense of hope about what genomics can tell doctors at this point.

“People who are in desperate medical situations look for all kinds of information about what they can do, and this is just the latest thing,” says Doug Hawkins, the associate division chief for hematology/oncology at Seattle Children’s Hospital. He says genome sequencing in its current form “is an incredibly powerful tool” for generating new data about the biology of health and disease, but careful studies will need to be done before it can provide a reliable guide for physician decision making.

Part of the reason, he says, is that tumors are loaded with vast number of genetic abnormalities, and it’s not always clear which ones are really harmful. The genome instruments also spit out terabytes of data, which creates a massive computing challenge for storing, accessing, visualizing, and interpreting the data. Essentially, too often doctors don’t know yet what all that data means.

Doug Hawkins, associate division chief of hematology/oncology, Seattle Children's

“If you have enough money, you can do genomic testing for everyone,” Hawkins says. “But the real proof will be in whether you can achieve better outcomes. Until then it’s just another expensive test.”

Even researchers doing some of the most important genomics work are being careful not to overplay their hand. “Our capacity to generate data is outstripping our ability to interpret it in ways that are useful to physicians and patients,” a team of University of Washington genomic researchers said last week in a high-profile paper in Science Translational Medicine which was covered by the New York Times.

What’s awesome about this movement toward “medical genomics” is the opportunity it provides for learning more about cancer, rare childhood disorders, and much more. Foundation Medicine’s Pellini says his company is interested in finding ways to help physician customers more easily share what they are learning about cancer care with their peers from this kind of genomic screening.

Exciting as it may be, Pellini doesn’t want to see people running around and promising magical insights will come from his company’s tests, when the data can’t back up the claims. “We continue to want to make sure that the science and the data support the decision-making of clinicians. It’s extremely important to us,” Pellini says.

That said, what we are really talking about here is a vast new ability to conduct clinical research. Cancer physicians will be able to form new hypotheses on how to treat individual patients, and then test those hypotheses, according to the scientific method.

“What is science driven by? It’s driven by hypotheses,” Pellini says. “I know some clinical decisions will be based on hypotheses. It’s how we learn in the medical community. People want this. It’s extremely important. But it’s also extremely important that the oncology community continues to follow the standard of care. The types of oncologists we are dealing with are following standard of care. The problem is that the standard of care has brought them to the point where there are no more good options.”

Genomics has the potential over the long term to identify some new options for treating patients. It has the potential to make clinical trials faster, shorter, and more likely to succeed—which is something the pharma industry desperately needs. It has potential to help scientists unravel some very tough biological riddles, like what causes Alzheimer’s.

But what it can’t do as of today is cure anybody’s cancer. The families of Steve Jobs and Christopher Hitchens can testify to that. Let’s all stay excited about the potential of genomic medicine, but not get too carried away about what it can actually do today.

By posting a comment, you agree to our terms and conditions.

  • Great article Luke. 

    I am definitely bullish on the upcoming movements towards more genomics in the clinic. I was recently at a personalized medicine/genomic conference and heard an excellent talk by Zack Kahone on Incidentalome – he is supportive of personalized medicine but his talk was about the need to be ultra careful of analyzing genomics before moving full force into the the clinic For Zack Kahone informatics software for integrating biology & bedside:  https://www.i2b2.org/about/index.html
    For concept of Incidentalome (mentioned by Zach Kahone) http://en.wikipedia.org/wiki/IncidentalomeAlso, Mike Snyder from Stanford gave a terrific talk on the topic of his own experience with personalized medicine and the importance of measuring the other ‘omics’ 
    For Mike Snyder personal genomics project:http://snyderlab.stanford.edu/

  • Salydrc

    I think they have “overly” geneticized things, as evidenced by the Swedish scientist – I cannot find his name, who looks to have proved some sort of “micro” evolution. One cannot also rule out environment, both physical and emotional, culture, climate, diet, etc. etc. as factors to disease. I wish I had studied this 20 years ago.

  • John Waslif

    The 4th Annual Personalized Medicine Conference in Boston on September 12-13 is exploring just this subject. Foundation Medicine will be presenting, along with other leaders in the field of personalized medicine (www.personalized-medicine-conference.com)