Vol 7. Issue 20 / July 2, 2007
From Science to Medicine: An Interview with Eric Topol
Earlier this year, leading cardiologist Eric Topol, previously of the Cleveland Clinic, took on an array of new responsibilities as chief of Genomic Medicine and Translational Science at Scripps Health, senior consultant in the Division of Cardiology at Scripps Clinic, and professor of translational genomics at the La Jolla, California, campus of The Scripps Research Institute. Here, he speaks with Mika Ono of News&Views about the goals for his new research program and the challenges of bridging the gap between today's science and tomorrow's medicine.
News&Views: It looks like you are settling into your lab here at Scripps Research.
Eric Topol: Yes, the whole team is here now, so it's very exciting actually. We have four senior investigators. In addition to me, there's Nick Schork, a world-renowned statistical genomics expert, who joins us from the University of California, San Diego, with his group of 10 lab members. Kelly Frazer joins us from Perlegen, where she headed up genomics. She's internationally known for the so-called "Hap Map," which maps the genome across ancestries picking out variants in the genome inherited in blocks (called haplotypes). Then there's Sarah Murray, who joins us from Illumina, a high-flying genome science company in San Diego. She was responsible for designing content for their 500K and 1 million SNP chips, which have transformed genome-wide association studies worldwide. [SNPs are single nucleotide polymorphisms, individual differences from the norm of a single base in the DNA.] So our team is assembled. They are phenomenal people and we're having a great time here.
Could you tell me about your research?
As you know, we are part of Ernie Beutler's Molecular and Experimental Medicine Department at TSRI [The Scripps Research Institute]. Our theme is "from gene discovery to changing the future of medicine." In other words, we not only want to make the discoveries, but also to take them all the way to having an impact on health care—from gene discovery to gene-based clinical trials. It's a unique program. There's none quite like it anywhere we know of.
What makes it so unique?
The continuum. Most centers focus on gene discovery, hunting for genes and then putting them out there for whoever wants to take subsequent steps. We're going to try to go all the way. I've had the chance of doing this with new molecular entities, such as t-PA (abciximab), but not with gene discovery. The technology has been changing dramatically. While we're not quite yet at the point of resequencing the whole three billion base pair human genome routinely, that's coming. But we do have the ability to look at anywhere between a half million and a million key variants in the genome from one person to the next. One remarkable resource we have is Ernie Beutler's work with Kaiser South. He has accumulated almost 40,000 samples of DNA, so that presents a great opportunity for us. We're also going to be prospectively developing new cohorts, of course working with Scripps Health—first at Scripps Clinic but also at Scripps Mercy, Scripps La Jolla Memorial, and all the Scripps Health facilities.
How are you going about doing that?
We have Institutional Review Board approval to draw DNA and create a biorepository of clinical data. We'll eventually include information from thousands of patients.
You're basically correlating DNA sequences with medical charts? With disease?
With a specific phenotype. For example, we could take all the heart attack patients coming to various hospitals and collect their DNA, knowing they qualified (that they really did have a heart attack), and see what genes are truly linked to that very important disease. But we want to go beyond heart disease to cancer, neurodegenerative disease, and also, importantly, health. What are the genes that account for health?
So you will have a normal population as well?
Yes, a normal, healthy elderly population, because you don't really know someone is healthy until he or she reaches 80-plus without developing important diseases. We're going to be working on that in a big way in the weeks and months ahead.
Are you fishing as well to see if anything surprising comes up?
At TSRI, there are already dedicated interests in certain areas that we want to take advantage of. For example, John Griffin, on the floor below us, has worked very hard on clot disorders. He has hundreds of patients with vein thrombosis issues, and we hope to work with him. We haven't even started rolling yet, but there are many investigators at TSRI who have laid a lot of groundwork. We've been meeting with some of them, like John Yates and Ben Cravatt. They're terrific and enthusiastic. In order for us to go from gene discovery to clinical trials, we have to understand the function of the genes and also be involved in drug discovery programs. Hugh Rosen's group can also be a phenomenal resource to help us in drug discovery efforts.
It sounds like you're extending a lot of genetic work that's already going on at TSRI.
Right. The Scripps Genomic Medicine Program—a joint program between Scripps Health and TSRI—is a formal agreement to work together on such projects as whole genome association studies, looking at 500,000 to 1,000,000 SNPs in the genome of large numbers of patients. We are planning to develop a kind of power resequencing of the genome, bringing in some technologies that have not yet been used or exploited at TSRI. Some of our work is actually hypothesis-free. We'll look at the whole genome to find areas of interest for resequencing key regions in hundreds of individuals with a specific phenotype.
Does Scripps Genomic Medicine exist as an entity now?
Yes. We have an identity, and the collaboration of Scripps Health and TSRI is part of our logo. It's a powerhouse because you have one of the largest private biomedical research facilities in the world and this incredible health system with access to a million plus individuals of diverse ancestry. We want to bring it all together and really advance the field.
Give me the history of how Scripps Genomic Medicine was created and how you came here.
I was recruited back in October. I started January 1, with the notion of developing this program under the heading of the Scripps Translational Science Institute.
How does that dovetail with Scripps Genomic Medicine?
Let me try to draw that out. The Scripps Translational Science Institute pulls a number of elements together. On one end is Scripps Genomic Medicine—this is really the discovery engine. Scripps Genomic Medicine is set up to find important genes. Through collaborative efforts in functional genomics, we hope to set the foundation for drug discovery, which is one of the strong suits of TSRI.
But eventually, potential therapies need to go into clinical trials. Provisionally, we've called this branch of the organization SACT or Scripps Advanced Clinical Trials, but it may wind up being called AIM, which is Advanced Individualized Medicine. We're working on this and it's starting to gel. There's big industry collaboration. Here, we'll do gene-based trials. For example, a rapid genotype would say that you qualify for a trial because you have the variant of interest and you would get a drug that is targeting a specific biologic alteration, shutting a gene down or revving it up. Or this could be protein-based, such as targeting a misfolded or unfolded protein. These trials will start with drugs and programs that already exist. Now, we'll do the genotyping back here, but the infrastructure of the trials requires large numbers of clinical research coordinators. That's a whole additional operation. The Coordinating Center that we are forming will hopefully be spearheading many gene- and protein-based randomized clinical trials around the world.
There is the potential for us to recruit additional investigators as needed. For example, let's say we have a really exciting discovery in Alzheimer's, and we find that even though there's incredible neuroscience depth at TSRI, we don't have a way to take it forward. In this case, we have the option of recruiting a translational investigator to lead this effort.
Could you elaborate on the link between the genomics and the drug discovery parts of your program?
Yes. Most of the genes that have been discovered have been reported in just the last few months. Recently, more diseases have been unraveled than in all of history. This week was prostate cancer. Last week was osteoarthritis. The week before was autism. About a month before that was diabetes. It's happening really fast.
By "unraveled," you mean genes were discovered?
The genes involved are becoming known, but the specific alterations within the genes and how this connects to the disease susceptibility or protection needs to be elaborated before this can go forward to improve medicine. Key gene markers have been discovered accounting for anywhere from 10 to 80 percent of a disease. Macular degeneration, a leading cause of blindness, is one example. Probably 80 percent of this disease's genetics is now known. But precisely how the complement factors induce susceptibility for blindness is still unknown. For diabetes, there's a gene called TCF7L2, a transcription factor. The only problem is no one has a clue what it does. We just know that this gene accounts for 25 percent of the population's risk of diabetes, which is enormous. Are we going to be able to figure out how it works? What is it about that gene variant that makes people so susceptible to diabetes? If we can figure that out then we can hopefully fashion a drug and eventually take it to clinical trial to prevent diabetes in select patients for whom alterations in TCF7L2 are driving the process.
How do you go about figuring out how it works?
There are lots of ways you can do that. This is where functional genomics comes into play, involving sophisticated molecular and cell biology experiments, transgenic mice, and a variety of probing techniques.
What is the bottleneck?
Not all genes are going to be amenable to therapy, but the scientific environment is changing rapidly. For example, now we're in the era of small interfering RNA, siRNAs, where you can basically knock down any gene. That approach is really exciting and with improving delivery techniques may soon become available. I wouldn't be surprised if siRNA and other novel biologics turn out to be really important.
Another promising area is vaccine development. If we know a gene is really important in, say, driving heart attacks and we have a pathway, we might fashion a vaccine for those people who are susceptible. This is going to be a challenge, but it only takes one success to make a big impact.
Interestingly, many genes that are important to each disease had never been theorized to be involved. Pinpointing the underlying biology like this opens up new doors for prevention and improved therapy.
It sound likes part of the problem with this step is the diversity. The key mechanism could be any number of things in the whole biological system.
Exactly. As you point out, already with gene-gene interactions or gene environmental interactions, it's a complex system biology story here, and you're hoping to get one that is a dominant driving force that you can manipulate or exploit.
Have there been success stories in individualized medicine?
There are many. The disease CML, chronic myelogenous leukemia, was a horrible disease. It was incurable. Now you can give a medicine called Gleevec, which is extraordinary because it's directed specifically at the disease's biologic underpinning. There's an almost 90 percent cure rate. Then, in about 20 percent or more of breast cancer cases, the Herceptin 2 or HER2 receptor is positive and very responsive to Herceptin as a medicine. That has been a tremendous boost for breast cancer success. It's only a fraction of breast cancer, but in those women who are HER2-positive it's a major advance.
Individualized medicine is taking off. In fact, we hope to do an important gene-based trial in heart failure later this year. Some people don't respond well to medicines and others do. There is a particular genotype that appears to predict whether a specific medicine could be useful. We're working on coordinating that trial.
So individualized medicine might involve an existing drug, but better identify the patients who would get the most out of it?
Exactly. These are examples of individualized medicine. It is going on now.
How is what you are doing now at Scripps Health and TSRI different from what you were doing before?
Well, for the immediate past year before coming here I was in a department of genetics teaching and doing research, but prior to that I spent 15 years at an institution where I ran the heart center. I started a new medical college, which was fun and exciting, but this is totally different. I'm having more fun than I've ever had in my professional life. This is a totally new program, starting from scratch, pulling pieces together, and assembling the great parts that are here. It's really fantastic. But I'm also involved in medicine. I get to walk across the street and practice cardiology.
Do you find that practicing medicine gives you insight into the research?
Yes, then you figure out what really is important day-to-day. What are on patients' minds? What are our frustrations and unmet needs? If you're not connected to that, it's hard to know where to put your energy and priorities.
So at the moment you want to get the word out about your program and let people know you're here?
Yes. We've met with several TSRI investigators individually and I hope to heighten awareness of our program so we will continue to make those connections. TSRI is such a remarkable brain trust, a powerhouse of talent, that we really look forward to getting collaborations going.
For more of Topol's thoughts on the future of genomics, look out for the upcoming commentary, "The Genomics Gold Rush," in the July 11, 2007, issue of the Journal of the American Medical Society (JAMA).
Send comments to: mikaono[at]scripps.edu