The completion of the human genome about a decade ago brought with it the promise of healthcare tailored to our personal genetic codes. But genomics has, so far, only trickled into the clinic in rare instances.
So, when will genomic medicine arrive in most doctors’ offices and hospitals?
Predicting medical advancements is dangerous business, as the past has shown. The official long-range plan casts the bulk of the transition at least a decade in the future, and experts have begun preparing for the range of issues raised by the technology.
In 1999, while the sequencing of the human genome was underway, the head of the U.S. government's sequencing team, Francis Collins, who is now director of the National Institutes of Health, described a hypothetical scenario just a decade away, when "the field of pharmacogenomics has blossomed." In it, a high cholesterol reading for a 23-year-old man, John, triggers a battery of genetic tests—some related, some not—to his risk of cardiovascular disease. Based on the results, he is prescribed a preventative drug regime, stops smoking, and plans to begin annual colonoscopies at 45.
But by 2010, of course, this scenario had not arrived in primary care offices. Instead, that year in the journal Nature, Collins discussed major accomplishments, but tempered expectations: “Remember that genomics obeys the First Law of Technology: We invariably overestimate the short-term impacts of new technologies and underestimate their longer-term effects.”
While innovations in personal medicine have yet to affect the vast majority of patients, they are beginning to find their way into practice in certain cases. For instance, doctors can choose a cancer treatment based on what they see in the sequence from a tumor; sequencing can aid the search for the causes of rare, undiagnosed diseases; and tests for certain disorders, such as autism, scan for many changes within a genome.
In the strategic plan for genomic medicine published in 2011, Eric Green and Mark Guyer of the National Human Genome Research Institute included a diagram illustrating progress in different domains within genomics. This figure—“completely hypothetical,” according to Guyer, NHGRI’s deputy director—shows the bulk of progress in healthcare beyond 2020, with scattered accomplishments occurring earlier.
“I think we are actually well ahead of the plan, which is uncommon in science,” says Gail Jarvik, head of medical genetics at the University of Washington. Already, genomics is being used to determine the genetic basis for diseases in cases in which that would not have been possible five years ago, she says. Jarvik predicts genomics will soon be used to diagnose and treat more common diseases: "Now we are applying it to cancer patients. In five years, I expect we will be applying it to heart disease."
Making a prediction
The door for pharmacogenomics—matching a patients’ genomic variation to a pharmaceutical—is already open. In certain cases, doctors screen patients for genetic variants to avoid dangerous reactions. For instance, people with a particular version of the HLA-B gene can suffer life-threatening reactions to the HIV drug Abacavir.
As part of its promise for personalized drug treatments, pharmacogenomics is seen as having the potential to reduce harmful effects caused by drugs. A research group led by Ramy Arnaout of Beth Israel Deaconess Medical Center created a mathematical model to predict when pharmacogenomics would take a significant bite out of these outcomes, and what it would cost to link genetic variants to enough adverse outcomes from the most common prescriptions to make it possible to cut the overall burden in half.
Based on what it took to link genetic variants to adverse outcomes for six drugs and one drug class, and a simulation involving drugs prescribed at Beth Israel, their estimation: Between $1.5 billion and $6 billion, and approximately 20 years. The cost varied depending largely on the degree to which genetic variation contributed to the risk of a bad outcome.
Medicine has already moved from a single-gene approach to looking for multiple genes' roles in disease, such as in cystic fibrosis. "It is not that difficult to say let's go to all genes and all the regions between the genes," Arnaout says. "Once we can connect genotypic information with phenotypic reaction—who had what adverse event on what drug—we will find the walls will fall rather quickly."
Preparing to practice
Now that genomics has begun to trickle into the clinic, with more applications on the way, healthcare providers must figure out how to best handle this new technology. NIH’s Clinical Sequencing Exploratory Research program has funded half a dozen projects across the country exploring the host of issues raised by the arrival of genomic medicine in the clinic.
At the University of Washington, researchers are sequencing the exomes, or protein-coding regions in the genome, of colorectal cancer and polyposis patients (a related condition). “To be honest, a bigger part of the project is: What about the rest of their genome?” says Jarvik, who is the principal investigator.
With the exome sequence comes plenty of information, a great deal unrelated to the medical issue, such as colon cancer, at hand. Some of the variations uncovered in a sequence are meaningful, some are not, and the significance of some is unclear. The question of what, and how to tell a patient, is crucial.
For instance, at the University of Washington, a committee has drawn up a list of disease-causing variations that prompt further review and discussion if they pop up in someone’s sequence, Jarvik says. Doctors must be certain that a particular variant does cause disease before they inform a patient that he or she carries it, she says.
Not the only cause of disease
“We have to remember, the stakes are very high in clinical medicine, and we can hurt people when we misapply and prematurely apply complex tests,” says James Evans, professor of genetics and medicine at the University of North Carolina, Chapel Hill, and principal investigator for another project funded by the NIH Clinical Sequencing Exploratory Research program. For the foreseeable future, he sees limits on the benefits of whole-genome sequencing for patients.
“Don’t get me wrong, I do think in given situations, in given clinical contexts for certain people, genomics will literally be lifesaving. It is right now. My point is for most of us, most of our genome is both uninterpretable and boring. … I think that will persist for quite a long time because most of the diseases we get are not predominantly genetic.”
See the nygenome blog's ongoing coverage of Bringing Genomics to the Bedside here.
Wynne Parry is a journalist based in New York. She is a regular contributor to LiveScience.com, and her work has appeared in The New York Times, Scientific American, Discover Magazine and the New York Post.