Venter and Topol on the True Revolution in Medicine

Introduction

Eric J. Topol, MD: Hello. I'm Dr. Eric Topol, Editor-in-Chief of Medscape. I'm really thrilled to have with me Dr. Craig Venter [founder and CEO of the J. Craig Venter Institute and Synthetic Genomics Inc.]. We're going to be discussing genomics in medicine and all sorts of things. As you know, Craig is really a hero of mine, a friend of mine, but also the most accomplished scientist of our era. It's really a great privilege to have a chance to sit down with you today.

J. Craig Venter, PhD: It's good to be with you.

Genomics in Medicine: 13 Years Later

Dr. Topol: You've been thinking a lot about genomics over many years. I still remember that White House meeting in 2000 with President Clinton and [current National Institutes of Health (NIH) director] Francis Collins. At that time, of course, it was said that genomics was going to have a big impact on medicine. Now we're here 13 years later. What do you think? Where are we going to go with genomics in medicine? What's holding us back?

Dr. Venter: Well, lots of things. Everything from how we fund science to the rate of change in medicine. Fortunately, as you know, we've had a huge technology change in the past decade -- not a lot of genomic advances but a lot of technology changes. We had a giant building filled with 300 sequencing machines that sequenced the human genome in 9 months. We had another giant building with a computer that was 1.5 teraflops. Both of these were 4-story buildings. Today, as you know, each of those buildings can be replaced with small boxes, and that 9-month period can be replaced with merely a week or two. So things have changed dramatically.

The Human Genome Project^[1] ended up costing about [$3 billion] globally. Our team spent around a mere $100 million on the project, but those efforts are replicable in terms of affecting patients. Now, it probably costs less than $10,000 to sequence a genome if we're being realistic about it. That's starting to change the number of genomes that can be done. We're starting to get actionable data. If you have lung cancer, as you know, the most important thing is to sequence the cancer gene, which determines whether Pfizer's crizotinib [a personalized drug for non-small cell lung cancer] will work on your type of tumor. There's probably no more important information to have, but these are just dribs and drabs coming out.

Cancer and Genomic Disease

Dr. Topol: That's actually one of the questions about cancer or genomic disease. If somebody has cancer, it's now at the point where you get whole genome sequencing of the tumor, the germline, and find out what the driver mutations are and how to develop precision therapy based on those mutations. As you may know, 11 of the 12 drugs approved by the US Food and Drug Administration (FDA) for cancer over the past year cost over $100,000 each. So, what you're talking about in terms of sequencing costs plummeting, do you think that sequencing cancer tumors is a logical way to move the standard of care in the years ahead?

Dr. Venter: It really depends on the pharmaceuticals. The Pfizer drug was discovered almost by accident in that the clinical trial failed. Then they found that if you had a certain point mutation you had a 60% chance of tumor regression with crizotinib. I think it needs to be part of drug testing and drug development in the future because these markers don't need to be causal, only coincidental. And that understanding needs to be part of how the clinical trials are done from the outset. If we use genomics and incorporate it into every part of medicine, we'll start to build the databases to make it so that these [personalized drugs] aren't rare events, but we need large numbers to do so. That's the problem. Having my genome, having Jim Watson's genome, and having a few other people's genomes doesn't change the course of medicine. We need tens of thousands of sequenced genomes.

Dr. Topol: Maybe millions, right?

Dr. Venter: Yes.

Tearing the Walls Down

Dr. Topol: Why don't we tear the walls down and have all of the sequencing that is being done pulled together? Wouldn't that be a way to accelerate sequencing in the same way that technology is accelerating?

Dr. Venter: Well, there are problems. The sequencing technology today, even though it's faster and cheaper, is still not up to snuff. In fact, the most accurate genome sequence was my genome, which was sequenced in 2007. It was the first and only truly diploid genome for which we could separate half the types. So, although we've gotten faster and cheaper, the technology has gotten less accurate. It's going to be a real challenge for the FDA to get whole genome sequencing up to diagnostic-quality standard.

Dr. Topol: That is such an important point you made about your diploid genome. In general, we don't do de novo assembly. We have a shaky human reference genome. What are your thoughts about that one?

Dr. Venter: It's done in a way to avoid the [hefty] costs but it's also the real part of genomics that's needed. What you want to know about yourself is your complete diploid genome, which means, "here's the sequence of the chromosomes that you got from your mother; here's the one you got from your father." Because, in doing that, we can actually find what's called compound heterozygotes. As you know, that is totally different from simple sequencing. You have this single-nucleotide polymorphism (SNP) variation, and probably most genetic diseases, like most human traits, are a combination of what we get from both parents. Without the complete diploid genome, we won't know the true impact of the genetics associated with disease.

World's First Carbon-Neutral Research Building

Dr. Topol: So much of what's missing today is the rare or low-frequency variant alleles that we're hopefully going to find a lot more of in the future. But you're now building a new institute here in La Jolla, California, which I drive by all the time. It looks really fantastic. When is that going to be ready and what are you going to do in there?

Dr. Venter: It's a unique model. We're going to be the first truly independent research institute on the University of California San Diego (UCSD) campus. It's part of, but also independent of, UCSD. It's going to be putting new things into action. It's going to be the world's first zero-carbon or carbon-neutral research building. It's not so hard to do with an office building, but it's very hard to do when you have computers and fume hoods and are following Occupational Safety and Health Administration (OSHA) regulations for air changes. It's really trying to put all of these things into action to show that you can build an environmentally interactive building and do high-level research. We're going to do all of the things we currently do. There is going to be a lot of synthetic genomics research and a lot of environmental genomics. We're going to continue our ocean sampling and similar techniques that we've used to start this new field -- the microbiome.

The Microbiome's Role in Health and Disease

Dr. Topol: I wanted to get into that because it is such a hot area. Again, you were ahead of the curve. Years ago, you were pushing on the microbiome. And now, particularly the gut microbiome seems to be linked with obesity, diabetes, even hypertension. Where are we going to go with the microbiome? Are we all going to start having gut microbiome sequencing?

Dr. Venter: It's not just the gut. It's the oral cavity. It's the skin. It's the vagina. It's really important because, as you know, we have 20,000-plus human genes. We have about 10 million bacterial genes. So, the complexity of the bacterial metabolism affects what's circulated in our blood. We all have about 50 chemicals circulating through our brains right now from bacterial metabolites of human chemicals, as well as chemicals from our diet. Nobody has ever looked or even asked the question of what they do because we didn't know they were there. It's not surprising that it affects medicine. We are humans in a microbial environment. We're breathing microbes and viruses right now. They're part of our whole physiology. Again, we need large populations to understand that. But almost everywhere people look now, using our techniques with the microbiome, they're finding clear associations. There is going to be microbial replacement therapy, but hopefully, instead of the kind of antibiotic therapy we've been using, we can go to very targeted therapy. We're developing synthetic phage at my company, Synthetic Genomics, to target very specific bacterial populations. For example, if you want to get rid of H pylori, you don't wipe out your entire bacterial populations. You get rid of just the H pylori.

Dr. Topol: Or for C difficile,maybe we can get past fecal transplants?

Dr. Venter: Exactly.

Dr. Topol: That would be nice.

Dr. Venter: We can simply define those populations and do it with a simple pill or suppository in a very defined population. It will work for that individual, instead of just using average medicine practices for the average population.