It’s hard to overstate the significance of the bargain-basement price of the human genome. The cost to decipher an individual’s genetic code plunged from $10 million to $1,000 or less in a decade. To appreciate what that means, think of an individual’s genome as a single work of literature, one that encapsulates a life from infancy to old age, with all the unique strengths and frailties that characterize a human. For a genetics researcher, a single genome may make for fascinating reading, but the story it tells is limited. A library containing a million genomes would allow researchers to compare the intricacies of a million lives, potentially leading to precisely targeted therapies for specific diseases and for individual patients.
President Barack Obama’s $215 million precision medicine initiative, announced in January, is intended to accelerate this research. The funding is modest but the aim ambitious: to build a research cohort of a million volunteers willing to share genetic and other medical information to usher in a new era of precision diagnosis and treatment. Driving this change is high-throughput sequencing, a relatively new advance that enables researchers to decipher mountains of genetic information quickly and at lower cost. Gene panels use high-throughput sequencing to assess millions of genetic mutations simultaneously.
Dr. Elizabeth McNally, director of Northwestern University’s Center for Genetic Medicine, who spoke Oct. 19 at the U.S. News Hospital of Tomorrow Conference, is a precision medicine pioneer deeply involved in this research. Northwestern is also home to the NUgene Project biobank, a repository of DNA from 12,000 volunteers and one of 10 collaborating institutions with an early lead in the field. McNally, a cardiologist, spoke to U.S. News about the promise and challenges ahead. (The interview has been edited for length and clarity.)
Can you give us a status report on precision medicine?
I’ve been running a cardiovascular genetics clinic for a number of years, focusing on heart disease that runs in families. When we started, we didn’t have any genetic testing. What really revolutionized bringing genetics into practice were some technological advances in 2006 or 2007, including [high-throughput] sequencing. Now we’re seeing the fruits of this [technology] making its way into practice. Gene panels are coming out, and there’s all this testing to be done. The other thing that happened is that we now have large databases of human genetic sequence to begin to understand how much [genetic] variation there is from person to person. It has surprised lots of people to discover how much is really there. The challenge we face is to make sense of all this variation and to [figure out how to use it] to understand health outcomes and how we manage people with drugs.
To what extent has this information simplified things or complicated them?
Dr. Elizabeth McNally is the director of the Center for Genetic Medicine and a professor in the departments of Medicine and Biochemistry, Molecular Biology and Genetics at Northwestern University.Courtesy Northwestern University
There are way more rare variations than people ever really thought. When I say rare, I’m talking about their presence in 1 of every 100 people, 1 in 200 people, 1 in 2,500 people. That means 85 percent of the variation in anybody’s individual genome is all but unique to them and their families. The combination of these variations is truly unique. Only 15 percent of the variation is common, meaning that it’s found in high frequency in the population.
Can you explain this in simpler terms?
When you look across a room of people, we’re all amazingly different from each other. We’re much more different than we ever imagined. There are unique differences in each of us. When you try to sum that up across a population, it’s hard to do. At the same time, it’s very helpful for managing individual patients. We have lots of great information that’s helping individuals and their families. It’s funny; when I give these talks, they say, ‘You’re just trying to help people.’ That’s what precision medicine is.
What implications does this have for research to associate genetic mutations with disease?
The trouble is that all genome-wide association studies are built on the premise that [diseases are caused by] common variations. That’s why these studies haven’t gotten us too far in using that information to manage individual patients. They’re looking at such a small portion of the genome, and it isn’t that different across populations.
Are these rare variations likely to harbor gene sequences that have a big impact on human disease?
That’s my bias. I admit we’re not in the majority in that point of view. What I do is order genetic testing on individuals who have forms of heart failure. If I order a test of 100 different genes, we [may] find rare [genetic variants] that we think are associated with that person’s disease. Now, across [a big] population that result doesn’t mean very much.
Almost everything I told you about heart failure we’d also say about breast cancer, some other kinds of cancers and neurological diseases. It’s really meaningful for individuals but it’s hard to draw conclusions across populations.
Could you talk about genetic profiling and how it’s used to predict risk, diagnose disease and apply drug therapies?
Again, there’s been a lot of hope that the 15 percent of variation that’s common would determine a lot, in terms of who gets hypertension and how a person responds to drugs. But it turns out that that’s [not the case]. If you decided to use [a genetic testing service that predicts your risk of getting common diseases, it might] tell you that you have a 1.4-fold increased risk of getting macular degeneration or something along those lines, but the overall risk is relatively small. They can’t tell you that you have a huge, huge risk of something because the stuff they test for is so common in the population. [The variants may be there, but they’re not always associated with disease.] Once you begin looking at rare variations … those genes are incredibly powerful within families.
So, if you’re a woman and you carry a BRCA mutation – and we know that less than 10 percent of breast cancer is due to BRCA mutations – your risk is actually pretty high of getting breast cancer over your lifetime. It’s a rare gene, but the risk associated with it is really very high.
If you know you have a higher genetic risk of an inherited disease, how do you act on the information?
What we’re talking about is having a little bit of a view of what an individual’s future might look like. And again, it’s a risk assessment – when you do a genome on someone you’re doing a risk assessment. And, yes, for many types of disorders we can reduce risk. Heart disease – we can reduce risk. If it’s really a bad [heart] rhythm, we can put in [a pacemaker or defibrillator], we can put people on medications and save lives that way. For cancer, we look at things like improving surveillance. Someone who carries a BRCA mutation will typically undergo mammograms at an earlier age, or, as in the case of Angelina Jolie, choose to have her breasts removed, if she’s really concerned. If you’re in a family with colon cancer, and you’re at increased risk, you’re not going to wait until you’re 50 to have your first colonoscopy.
How close are we to genuine precision medicine: an individual diagnosis paired with a personalized treatment plan?
We see a lot of successes happening already. If you have a tumor these days, it’s actually fairly likely that the tumor will be profiled in some way so that … a chemotherapy plan will be put together precisely for that tumor. We have genes now that can be used for guiding how certain drugs are given, such as how much blood thinner you may need to take or whether or not you should be on aspirin. Things like that are evolving and coming into regular practice. And, again, we’re certainly [using] it for a lot of inherited diseases. Genetic mutations and inheritance play a much bigger role in many diseases than we realized. We can counsel family members and give them good advice. We see that in every branch of medicine.
Can you cite diseases where this is becoming commonplace?
Almost any kind of cancer … in part because we know cancer is a largely genetic disease. Lung cancer is a great example. There are some are very clear gene mutations that will change the recommendation for how you get treated and what drugs are given to you. Certain types of head and neck cancer, exactly the same thing, where you’ll see some kinds of markers and you will absolutely change the recommendation for how that individual is treated, whether they’re going to get chemotherapy or surgery upfront. Those are two very different courses of action. Almost every type of leukemia and lymphoma are now [tested] for genetic rearrangements that occur; we have very different treatments for those. We now even consider them different disease from each other.
Does this mean we’re beginning to rethink how we classify these diseases?
What we used to consider as one group of diseases we now understand in many cases are different diseases compiled together. In my area, the heart, we’re starting to see that happen as well. We used to treat all forms of heart failure as one. We’re starting to see, no, there are subtypes of heart failure that depend on what gene mutations you have. In this group, should we think of [implanting a pacemaker and/or defibrillator] sooner because they’re more prone to arrhythmias? There are really clear areas in neurology where this is emerging as well. Epilepsy is a great example. We have all these different drugs to treat epilepsy and it’s now kind of a random guess which drug will be applied to each patient. When we understand the genetics of epilepsy better, we’ll be able to make much more rational choices.