NEWS
Leroy Hood Looks Forward to P4 Medicine: Predictive, Personalized, Preventive, and Participatory By Charlie Schmidt
jnci.oxfordjournals.org
bioethicist at New York’s NYU Langone Medical Center, said there’s no guarantee people will change behavior just because they’re more aware of their personal risk factors.
“In about 5 years, we’ll be able to sequence an entire human genome in 15 minutes for roughly $500.” P4 took shape at the California Institute of Technology, where Hood taught from 1968 to 1990. His research teams produced the protein sequencer, protein synthesizer, DNA synthesizer, automated DNA sequencer, and inkjet DNA synthesizer—all crucial to the Human Genome Project. The Human Genome Project’s fundamental accomplishment, he said, was to “democratize the network of 20,000 genes in the human body,” meaning they were now available to anyone with the right tools. Meanwhile, computers and the Internet brought about other networked data sets, such as electronic medical records and personal networks arising from social media. According to Hood, P4 unites them all into a “network of networks” that extends from DNA; to cells, tissues, and individuals; and finally to human communities. “They’re seamlessly integrated,” he said. “So what we learn about molecular networks can be applied right out to social networks.” The goal is to mine this integrated system for “actionable possibilities” to keep people healthy. Hood said that signal-tonoise ratios—the challenge of extracting useful signals from huge amounts of data— will be a problem. But on the basis of his earlier research, he’s confident it can be done. For example, four gene networks can
explain a neurodegenerative illness in mice, and 13 proteins in blood can distinguish cancerous from benign lung nodules with 90% accuracy. Hood cowrote a 2013 Science Translational Medicine article describing the latter example. According to Hood, measurements from the blood, genome, and gut microbiome offer the best information. Flush with $100 million from the government of Luxembourg, Belgium, he and ISB are testing that hypothesis in 100 Seattle-area volunteers (Hood calls them pioneers). ISB will sequence their genomes; analyze saliva, urine, blood, and stool samples; and evaluate parameters such as sleep patterns and blood pressure. ISB plans to expand the study by a factor of 10 annually, until it grows into a prospective, longitudinal evaluation of 100,000 people that could last 30 years. Measurements so far (none genomically based yet) reveal nutritional Leroy Hood, M.D., Ph.D. problems (especially vitamin D deficiency), evidence of chronic inflammation, lipid imbalances that predispose to heart disease, and elevated glucose—a risk factor for diabetes. Hood describes each problem as having actionable potential for improving health. Clay Marsh, M.D., executive director of the Center for Personalized Health Care at Ohio State University in Columbus, said that over time these measures create around the individual a frame of reference that predicts disease. “Leroy’s view is that in complex systems, everything is connected, but some features JNCI | News 1 of 8
Downloaded from http://jnci.oxfordjournals.org/ at Akdeniz University on December 23, 2014
A
t the American Society of Clinical Oncology (ASCO) annual meeting in Chicago, genetics pioneer Leroy Hood, M.D., Ph.D., predicted, “In about 5 years, we’ll be able to sequence an entire human genome in 15 minutes for roughly $500.” Hood cofounded Seattle’s Institute for Systems Biology (ISB) and believes the availability of cheap sequencing data and other personal information is bringing health care to a tipping point. Instead of reacting to disease, he said, health care systems will soon access a “virtual data cloud” that envelops individuals, simplifying disease prevention. Made up of genomic data, clinical measures, and lifestyle information, that data cloud is central to Hood’s new health care paradigm, “P4 medicine”: predictive, personalized, preventive, and participatory. Hood claims that P4 will demystify illness and optimize individual wellness. “And I wouldn’t bet against him,” said Clifford A. Hudis, M.D., chief of Memorial Sloan– Kettering Cancer Center’s Breast Cancer Medicine Service in New York and ASCO’s immediate past president. Having led the development of molecular technologies used in high-throughput research, Hood has secured his global reputation. A member of the National Academy of Sciences, the National Academy of Engineering, and the Institute of Medicine, he has won many awards, including the National Medal of Science. But Hood’s ideas raise challenging questions. P4 faces an uncertain regulatory environment. Hudis said that tools used to guide medical decisions can be construed as devices, and the U.S. Food and Drug Administration has no bright line separating what constitutes a device from what doesn’t. Moreover, Art Caplan, Ph.D., a
NEWS
shifts and you can’t get good food in your neighborhood, then you’re not going to benefit from P4.” To Hood, that is a trivial generalization. “The whole mantra of eating less, exercising more, and staying out of the sun has gotten us nowhere because what people need is more targeted information about what they can do stay healthy,” he said. Experience with the pioneers, he said, shows that vitamin D levels among those who were deficient shot up to normal after they began taking supplements. “This shows what can happen when you educate people more appropriately.” But what happens when a risk factor doesn’t have an obvious response? Although managing elevated glucose is straightforward, Hudis said, a prediction of amyotrophic lateral sclerosis is not. Yet P4 could help in this respect, Hudis said. By
studying networked systems, scientists may be able to extract lifestyle patterns associated with worse outcomes in that and other diseases. “This is why we invited Leroy Hood to speak at ASCO,” Hudis said. “P4 redefines what we traditionally thought of as walled-off domains, and we’re especially interested in what we can learn from mining the [electronic medical records] of people who aren’t on clinical trials. Think of how Google anticipates what you’re typing—not by doing prospective or randomized experiments, but rather by observing the typing behavior of billions of people. It’s not about p values but about trends and observations. Only about 3% of the population participates in clinical trials. And we might learn a lot from studying outcomes in 325 million Americans.” © Oxford University Press 2014. DOI:10.1093/jnci/dju416
Cancer Screening in Older Adults: Risks and Benefits By Sue Rochman
W
ho should undergo cancer screening? How often? Which tests are beneficial? Although screening strives to reduce cancer deaths, different answers to these questions have led to inconsistent screening guidelines from the U.S. Preventive Services Task Force and leading medical organizations. Even so, most agree that screening is not beneficial in people with limited life expectancy. An October 2014 study in JAMA Internal Medicine (doi:10.1001/jamainternmed.2014.3895) examined prostate, breast, cervical, and colorectal cancer screening patterns among 27,404 older adults (aged ≥65 years) with different life expectancies. Screening remained common among people at very high risk of dying within 9 years—and it continued among those expected to live less than 5 years. Among participants with a very high mortality risk (≥75% chance of dying within 9 years):
2 of 8 News | JNCI
• 55% of men had recently undergone prostate cancer screening; • 38% of women had recently undergone screening for breast cancer and 31% had recently undergone cervical cancer screening; and • 41% of women and men had recently undergone colorectal cancer screening. No one knows precisely why people with a very high mortality risk get screened, said study senior author Ronald C. Chen, M.D., M.P.H., associate professor of radiation oncology at the University of North Carolina at Chapel Hill. Part of the problem, he said, could be patient driven: “Patients think screening is a good thing, and the idea that there is a point in a person’s life where their limited life expectancy may make screening unnecessary is a new concept for many patients—and it may be a difficult idea for patients to accept.”
The topic can also be hard for physicians to raise. “To tell someone their life expectancy is now limited and screening is not helpful is a potentially difficult discussion to have,” said Chen, “and primary-care physicians may not have enough time allotted to have those important discussions.” A study in the April 8, 2013, JAMA Internal Medicine describes such challenges a doctor may face. Among 33 adults aged 63–91 years, many patients viewed stopping screening as a monumental decision. Some patients found it upsetting to hear they should stop screening. Others said advice to stop cancer screening might lead them to change doctors. Chen said it is difficult for patients to see a test that is not typically dangerous as harmful. The harm comes when a patient or family members learn that a screening test is abnormal, which then leads to more tests downstream, including invasive Vol. 106, Issue 12 | December 10, 2014
Downloaded from http://jnci.oxfordjournals.org/ at Akdeniz University on December 23, 2014
are more important than others,” said Marsh, who cofounded the P4 Medicine Institute with Hood. “The important signals are mixed up in the noise, so we’re developing tools to look for convergence and simplicity.” But will individuals capitalize on this information to keep illness at bay? Caplan said he’s doubtful, especially with the billions companies spend to promote unhealthy products, such as processed food. “We’ve known for a long time that being healthy requires lifestyle changes,” Caplan said. “We just need people to walk more, drink less [alcohol], eat better, reduce stress, and sleep more. Even without mapping a single gene, we should be able to make a huge dent in wellness now, and we’re not doing it. As with many things, the wealthier will respond better to personalized medicine. But if you’re poor and working three
Leroy Hood Looks Forward to P4 Medicine: Predictive, Personalized, Preventive, and Participatory By Charlie Schmidt
jnci.oxfordjournals.org
bioethicist at New York’s NYU Langone Medical Center, said there’s no guarantee people will change behavior just because they’re more aware of their personal risk factors.
“In about 5 years, we’ll be able to sequence an entire human genome in 15 minutes for roughly $500.” P4 took shape at the California Institute of Technology, where Hood taught from 1968 to 1990. His research teams produced the protein sequencer, protein synthesizer, DNA synthesizer, automated DNA sequencer, and inkjet DNA synthesizer—all crucial to the Human Genome Project. The Human Genome Project’s fundamental accomplishment, he said, was to “democratize the network of 20,000 genes in the human body,” meaning they were now available to anyone with the right tools. Meanwhile, computers and the Internet brought about other networked data sets, such as electronic medical records and personal networks arising from social media. According to Hood, P4 unites them all into a “network of networks” that extends from DNA; to cells, tissues, and individuals; and finally to human communities. “They’re seamlessly integrated,” he said. “So what we learn about molecular networks can be applied right out to social networks.” The goal is to mine this integrated system for “actionable possibilities” to keep people healthy. Hood said that signal-tonoise ratios—the challenge of extracting useful signals from huge amounts of data— will be a problem. But on the basis of his earlier research, he’s confident it can be done. For example, four gene networks can
explain a neurodegenerative illness in mice, and 13 proteins in blood can distinguish cancerous from benign lung nodules with 90% accuracy. Hood cowrote a 2013 Science Translational Medicine article describing the latter example. According to Hood, measurements from the blood, genome, and gut microbiome offer the best information. Flush with $100 million from the government of Luxembourg, Belgium, he and ISB are testing that hypothesis in 100 Seattle-area volunteers (Hood calls them pioneers). ISB will sequence their genomes; analyze saliva, urine, blood, and stool samples; and evaluate parameters such as sleep patterns and blood pressure. ISB plans to expand the study by a factor of 10 annually, until it grows into a prospective, longitudinal evaluation of 100,000 people that could last 30 years. Measurements so far (none genomically based yet) reveal nutritional Leroy Hood, M.D., Ph.D. problems (especially vitamin D deficiency), evidence of chronic inflammation, lipid imbalances that predispose to heart disease, and elevated glucose—a risk factor for diabetes. Hood describes each problem as having actionable potential for improving health. Clay Marsh, M.D., executive director of the Center for Personalized Health Care at Ohio State University in Columbus, said that over time these measures create around the individual a frame of reference that predicts disease. “Leroy’s view is that in complex systems, everything is connected, but some features JNCI | News 1 of 8
Downloaded from http://jnci.oxfordjournals.org/ at Akdeniz University on December 23, 2014
A
t the American Society of Clinical Oncology (ASCO) annual meeting in Chicago, genetics pioneer Leroy Hood, M.D., Ph.D., predicted, “In about 5 years, we’ll be able to sequence an entire human genome in 15 minutes for roughly $500.” Hood cofounded Seattle’s Institute for Systems Biology (ISB) and believes the availability of cheap sequencing data and other personal information is bringing health care to a tipping point. Instead of reacting to disease, he said, health care systems will soon access a “virtual data cloud” that envelops individuals, simplifying disease prevention. Made up of genomic data, clinical measures, and lifestyle information, that data cloud is central to Hood’s new health care paradigm, “P4 medicine”: predictive, personalized, preventive, and participatory. Hood claims that P4 will demystify illness and optimize individual wellness. “And I wouldn’t bet against him,” said Clifford A. Hudis, M.D., chief of Memorial Sloan– Kettering Cancer Center’s Breast Cancer Medicine Service in New York and ASCO’s immediate past president. Having led the development of molecular technologies used in high-throughput research, Hood has secured his global reputation. A member of the National Academy of Sciences, the National Academy of Engineering, and the Institute of Medicine, he has won many awards, including the National Medal of Science. But Hood’s ideas raise challenging questions. P4 faces an uncertain regulatory environment. Hudis said that tools used to guide medical decisions can be construed as devices, and the U.S. Food and Drug Administration has no bright line separating what constitutes a device from what doesn’t. Moreover, Art Caplan, Ph.D., a
NEWS
shifts and you can’t get good food in your neighborhood, then you’re not going to benefit from P4.” To Hood, that is a trivial generalization. “The whole mantra of eating less, exercising more, and staying out of the sun has gotten us nowhere because what people need is more targeted information about what they can do stay healthy,” he said. Experience with the pioneers, he said, shows that vitamin D levels among those who were deficient shot up to normal after they began taking supplements. “This shows what can happen when you educate people more appropriately.” But what happens when a risk factor doesn’t have an obvious response? Although managing elevated glucose is straightforward, Hudis said, a prediction of amyotrophic lateral sclerosis is not. Yet P4 could help in this respect, Hudis said. By
studying networked systems, scientists may be able to extract lifestyle patterns associated with worse outcomes in that and other diseases. “This is why we invited Leroy Hood to speak at ASCO,” Hudis said. “P4 redefines what we traditionally thought of as walled-off domains, and we’re especially interested in what we can learn from mining the [electronic medical records] of people who aren’t on clinical trials. Think of how Google anticipates what you’re typing—not by doing prospective or randomized experiments, but rather by observing the typing behavior of billions of people. It’s not about p values but about trends and observations. Only about 3% of the population participates in clinical trials. And we might learn a lot from studying outcomes in 325 million Americans.” © Oxford University Press 2014. DOI:10.1093/jnci/dju416
Cancer Screening in Older Adults: Risks and Benefits By Sue Rochman
W
ho should undergo cancer screening? How often? Which tests are beneficial? Although screening strives to reduce cancer deaths, different answers to these questions have led to inconsistent screening guidelines from the U.S. Preventive Services Task Force and leading medical organizations. Even so, most agree that screening is not beneficial in people with limited life expectancy. An October 2014 study in JAMA Internal Medicine (doi:10.1001/jamainternmed.2014.3895) examined prostate, breast, cervical, and colorectal cancer screening patterns among 27,404 older adults (aged ≥65 years) with different life expectancies. Screening remained common among people at very high risk of dying within 9 years—and it continued among those expected to live less than 5 years. Among participants with a very high mortality risk (≥75% chance of dying within 9 years):
2 of 8 News | JNCI
• 55% of men had recently undergone prostate cancer screening; • 38% of women had recently undergone screening for breast cancer and 31% had recently undergone cervical cancer screening; and • 41% of women and men had recently undergone colorectal cancer screening. No one knows precisely why people with a very high mortality risk get screened, said study senior author Ronald C. Chen, M.D., M.P.H., associate professor of radiation oncology at the University of North Carolina at Chapel Hill. Part of the problem, he said, could be patient driven: “Patients think screening is a good thing, and the idea that there is a point in a person’s life where their limited life expectancy may make screening unnecessary is a new concept for many patients—and it may be a difficult idea for patients to accept.”
The topic can also be hard for physicians to raise. “To tell someone their life expectancy is now limited and screening is not helpful is a potentially difficult discussion to have,” said Chen, “and primary-care physicians may not have enough time allotted to have those important discussions.” A study in the April 8, 2013, JAMA Internal Medicine describes such challenges a doctor may face. Among 33 adults aged 63–91 years, many patients viewed stopping screening as a monumental decision. Some patients found it upsetting to hear they should stop screening. Others said advice to stop cancer screening might lead them to change doctors. Chen said it is difficult for patients to see a test that is not typically dangerous as harmful. The harm comes when a patient or family members learn that a screening test is abnormal, which then leads to more tests downstream, including invasive Vol. 106, Issue 12 | December 10, 2014
Downloaded from http://jnci.oxfordjournals.org/ at Akdeniz University on December 23, 2014
are more important than others,” said Marsh, who cofounded the P4 Medicine Institute with Hood. “The important signals are mixed up in the noise, so we’re developing tools to look for convergence and simplicity.” But will individuals capitalize on this information to keep illness at bay? Caplan said he’s doubtful, especially with the billions companies spend to promote unhealthy products, such as processed food. “We’ve known for a long time that being healthy requires lifestyle changes,” Caplan said. “We just need people to walk more, drink less [alcohol], eat better, reduce stress, and sleep more. Even without mapping a single gene, we should be able to make a huge dent in wellness now, and we’re not doing it. As with many things, the wealthier will respond better to personalized medicine. But if you’re poor and working three
