One is narcoleptic, has high cholesterol, and takes vitamin C. Another struggles with depression and anxiety, is allergic to cats, and wears bifocals. A third has flat feet, type O-positive blood, and the ghost of a scar from a basal cell carcinoma removed a decade ago.
Such details may seem like stray jottings from patients’ medical records, or perhaps snippets from dinner-party patter on age-related woes. In fact, they’re data from the PGP-10—a group of ten volunteers who agreed in 2006 to have their personal medical information made public as part of the Personal Genome Project. The brainchild of George Church—he of the narcolepsy and vitamin supplements—the project aims to sequence the genomes of 100,000 people. By making the very personal very public, Church hopes that researchers will plumb the information to “develop preventive approaches and cures for disease.”
The foundation for the PGP was laid in 1964, when researchers at Cornell University and the U.S. Department of Agriculture sequenced the first gene by determining the order of the four bases, or nucleotides, that make up our DNA: adenine, thymine, cytosine, and guanine. Nucleotide chains form the basic recipe for every human.
That discovery had a lasting impact on the nine-year-old Church, now a professor of genetics at Harvard Medical School. A decade later, he was at a computer—rare in biology circles at the time—entertaining himself in a way that perhaps only a budding scientist might. “One day I started typing in all the known DNA and RNA sequences,” Church says, describing the numerous combinations of bases. “I thought folding them up was really cool.”
Soon, an idea blossomed. “I wondered whether we could sequence all the genes in all people,” says Church. It was, admittedly, a vague and overwhelming plan—but one he thought worth pursuing. Church switched from x-ray crystallography to genetics and began his quest in earnest. By the early 1980s, he and Harvard colleague Walter Gilbert—whose pioneering methods for sequencing the nucleotides in DNA had already earned him a Nobel Prize—had published a strategy for isolating sequences from mouse DNA. That paper, “Genomic Sequencing,” caught the eye of administrators at the U.S. Department of Energy, who requested a meeting with Church, Gilbert, and other researchers.
From that 1984 meeting, the concept for the Human Genome Project was born. “We proposed sequencing a complete human genome,” says Church. Through a process that married computer technology with laboratory science—namely electrophoresis, which separates molecules like DNA using chemical solutions and a charged electric field—the researchers would map the 3 billion base pairs of the genome. “The Department of Energy leaders acted immediately,” Church says. “They didn’t ask Congress; they didn’t ask anybody’s permission. They just started writing checks.”
To many, it seemed an audacious goal. With so much to sequence, critics charged that the project was impossible to achieve and the cost prohibitive. Yet by 2003, researchers—now funded by the National Institutes of Health—had sequenced the majority of the human genome. The price tag? Roughly $3 billion, or about a dollar a gene.
Sequencing the human genome is one feat. Interpreting it, however, is quite another. The media attention surrounding the project’s recent tenth anniversary included commentary from detractors as well as proponents. While biologists continue to be impressed with the project’s findings, others point out that the Human Genome Project has yet to deliver on its ultimate promise: to identify the causes of—and propel treatments for—major diseases such as cancer and Alzheimer’s. But perhaps such criticism is misplaced: In a recent interview with Science Watch, HMS geneticist David Altshuler ’90 argues that the project’s primary goal “is not to predict disease, nor to personalize medicine. It’s to understand the biological systems that underlie common diseases....We’re still in the early days.”
Church contends that the Human Genome Project has had a huge impact on prediction, enabling the development of more than 2,000 genetic tests. “Some genome centers focused too much on the old methods and on common variants,” he says of the project’s goals. “If we had prioritized technological advances earlier, then we would have seen an even greater impact.” In fact, those advances—significantly expanded since 1984—are what allowed Church to initiate the Personal Genome Project.
By 2003, anticipation began to build around another concept: Could every individual someday have his or her own genome sequenced? With an estimated cost of $100 million per person, the prospect was clearly out of reach. Yet by 2006, technologies that Church dubs next-generation sequencing—which virtually eliminated the need for electrophoresis—had slashed costs. “Today,” he says, “we can sequence one person’s genome for just $1,500. Someday, it may be so inexpensive that people could have their sequencing covered by third parties, such as insurance companies.”
With affordable technologies in hand, Church and his colleagues were ready to begin the PGP in earnest. But with whose genomes? Like any study, the PGP needed volunteers. The project’s scientific and social implications—essentially exposing intensely private information for all to see—led the HMS review board on human subjects to request that the PGP’s first group of volunteers be well versed in genomics to ensure that their consent was truly informed.
The first volunteer was one the review board suggested: Church himself. The list of other pioneering participants reads like a Who’s Who of the scientific and technology worlds. Volunteers include John Halamka, chief information officer of HMS; Harvard psychology professor Steven Pinker; and Esther Dyson, an investor in information technologies.
For Dyson, the decision to participate in the study came naturally. With a mathematician mother, physicist father, and science-historian brother, an appreciation for discovery runs in her family. “I’m reasonably healthy and wasn’t concerned about keeping my data private,” she explains. “To me, the genome itself is less interesting than what we do with it.”
It is, in fact, how the PGP data will be used that has scientists alternating between enthusiasm and concern. The open nature of the project—volunteers must be willing to lay bare even the tiniest details of their medical data—has lost some of its shock value in today’s Facebook- and Twitter-obsessed culture. Many of us already share our quotidian crises, romantic adventures, photographs, and stray musings on the Internet. It may not be such a leap to make our health information available, too. “The PGP data are useful scientifically,” Dyson says. “Why keep them private?”
The benefits of this tell-all approach are clear to Church. “We’re trying to determine how different genes, diseases, and environments interact,” he says. “That’s one reason we want a large pool of volunteers and why we want to make both genomic and trait information available to everyone. If we try to control who has access to that data, then we limit who can make breakthroughs using it.” A researcher studying liver health using the PGP data might, for example, make a major advance in the treatment of cirrhosis. But, says Church, that breakthrough just might be made by “the last person you would think to entrust with that data”—a computer programmer, perhaps, or another nonscientist. When it comes to discovery, such open access helps level the playing field.
Church cites this same philosophy when asked about his involvement with companies currently involved in genomic sequencing. He advises more than a dozen such ventures, in the hope that one—or a combination—will improve sequencing technology and drive costs down even further. “It’s pretty much like a race to the bottom,” he says. “Not that I like races so much, but it’s much healthier and more beneficial in the long run to have competition.”
Full exposure makes sense from a scientific—and capitalistic—standpoint, but how does it affect participants? Church and his colleagues make every effort to ensure that volunteers understand the full scope of their involvement. To qualify for the study, they must score 100 percent on a test that walks them through the consent process. The test makes clear that volunteering means giving up any control over who views their medical data, which can include conventional electronic medical records and imaging tests like MRI scans, as well as RNA, metabolic measures, and other information drawn from blood, saliva, and skin samples.
“We don’t want to exclude people,” Church explains, “but we do want them to know what they’re getting into. They have to understand that anyone can have access to all their genetic and medical information.”
That knowledge can be a curse—and a blessing. Just ask PGP volunteer number 6, whose sequencing revealed an allele that put him at elevated risk for hypertrophic cardiomyopathy. Church and his colleagues notified the volunteer and recommended that he see a cardiologist for an echocardiogram to determine whether his heart’s ventricular wall had begun to thicken, a sign of the disease. On one hand, such information—or knowledge of a gene that raises the risk of, say, breast cancer or diabetes—may be startling. On the other, it enables volunteers to take action to stave off full-blown disease. That aspect of the project holds particular appeal for Dyson. “The data that the PGP uncovered can help put our health in perspective,” she says. “They show that a disease risk doesn’t have to be our destiny.”
Of even greater value may be the data’s promise in advancing the field of personalized medicine. Although pharmacogenetics—the study of how and why specific medications appear to work only in certain populations—is in its early stages, genetic tests have already been developed to help determine which drugs, at what dosages, can help treat certain cancers, HIV/AIDS, psychoses, heart disease, and other conditions. Researchers recently identified, for example, a genetic variant in about 30 percent of people that decreases the effectiveness of the anticoagulant clopidogrel, or Plavix; these patients benefit from higher doses of the drug. Ideally, says Church, PGP data will help guide patients and their physicians in treatment decisions.
Truth to Power
With the potential benefits of open access come risks. At the top of the list of concerns: the possibility of discrimination by employers and insurers who know an individual’s personal genetic information. The Genetic Information Nondiscrimination Act of 2008 has made a leap forward in preventing such issues by making it illegal for employers to make hiring or promotion decisions based on genetic information. The act also bars health plans from charging higher premiums based on genetic information.
Church hopes that as more volunteers join the PGP—the plan is to expand enrollment to 100,000 participants—privacy concerns will fade. He cites PatientsLikeMe, a medicine-meets-social-networking website where visitors with such conditions as HIV/AIDS, neurodegenerative diseases, and depression share information about themselves and their health. “These used to be very stigmatizing conditions,” he says. “Now people see that the benefits of educating others and helping them cope with disease outweigh the hazards of revealing private knowledge. I believe the same thing has happened with the PGP.”
Dyson agrees. “The truth,” she says, “is never a risk.”
Jessica Cerretani, a former assistant editor of Harvard Medicine, is now a full-time freelance writer.