Genomes and Gene Therapy: ASGT president ponders each field's effect on the other

If the size of a society provides a reading of the health of a field, gene therapy may be heading for a fruitful stretch. The American Society of Gene Therapy (ASGT) attracted 600 abstracts and 1,700 participants to its inaugural meeting in Seattle last year, compared to 1,000 abstracts and 2,100 participants to its second annual meeting in Washington, D.C., last month. In a conversation with News Editor Paul Smaglik, James Wilson, ASGT president and director of the University of Pennsylvania's Institute for Human Gene Therapy, predicted that the gene therapy field will soon score some clinical successes. The following interview has been edited for length and clarity. Q: There will be at least one "rough draft of the human genome" next year. What implications will the completion of that project have for gene therapy? A: It's not how will the genome project affect gene therapy, but how can gene therapy affect the genome project. We're going to have all our genes identified--150,000 or whatever the most recent estimates are. In fact, we've been addressing the potential of that information over the last few years. So what is its potential? The potential is diagnostics or therapeutics. The traditional strategy for developing therapeutics has been in the pharmaceutical industry, which has embraced [genomics] and redesigned [its] R&D efforts around genomics-based drug discovery. The potential targets for drug therapy have expanded exponentially, but the process by which you develop a drug is still the same--it's just as long as it was before, and just as expensive. ... So where does gene therapy come in? Gene therapy could potentially provide a way ... to accelerate the development of new drugs, or actually create opportunities for developing drugs .... If you have all these genes, then we figure out what their function is--and that's based on proteins. Then once you know what the protein does, you have to understand its biology [how does it function?], and then you develop small-molecule drugs. With gene therapy, where we know what the proteins are, we could actually put the gene encoding a protein into a vector, and then use that to express the protein in animals to understand its biology. But the vector is a way to deliver the drug, which is a protein. So it's a drug-delivery tool. And in fact, you never have to manufacture the protein. You just have to figure out how to manufacture vectors. If you have a gene and you have a vector, you have all the tools you need to take some of these genes and make them therapeutic by putting them into vectors. The only turn on that is, you probably are going to have to be able to regulate the expression of the protein, and that's some of the work we've been doing. Indeed, the genome effort and [the] gene therapy [field] come together in a very powerful way. But I think the combination of the two will create a new paradigm for developing therapeutics. It's a way to short- circuit the development of therapeutic proteins themselves. The interesting thing is, you'll never have to purify the protein under this paradigm. That's a big plus. Q: There seems to be a great deal of excitement over the convergence of physics, chemistry, biology, and computational sciences. Could you comment on that convergence in the context of gene therapy? A: One thing that we struggle with a bit is information technology and genetics. In the genome, [that's] incredibly powerful. We can assemble, abstract, evaluate huge quantities of information, volumes of information, about patients, for example, [and] about diseases. Then we can also discern genetic mutations. The combination of those two is essentially correlating variation in our genome with disease. But to do that right, you need the genome project to come to its completion, you need high-technology characterization of genomes for genetic variation mutations, and then you need information technology in assembling large amounts of information about the patient populations. If you can put that all together, ... that's really how we're going to understand what our genome does. Having said that, there are obviously some challenges. In particular, there are some major ethical dilemmas about privacy of information. Correlating genetic variation with clinical phenotypes is going to take a long, long time. [For example], how much work did it take us to draw a correlation between serum cholesterol and heart disease? So let's talk about the genetic predisposition to colon cancer. There probably are a large number of genes that contribute and there are environmental components as well. Any one person will have a different mix of genetic and environmental contributors, but to ... develop enough information to be able to tell a patient that, as a result of this part of your genetic report card, you have X percent chance of developing cancer--that's going to take a lot of work. Unfortunately, I think what's going to happen is the genome will be identified, there will be genes that, in a worst-case scenario, clearly predispose to something; they'll be declared disease-causing genes. But not everybody who gets that particular defective gene will get that disease and, in fact, there may be other versions of that defective gene. But patients will want these tests, and doctors will want to prescribe them, companies will want to sell them. [Genetic screening] is going to happen, but we won't have all the correlative data to know what it means. Q: Do you sense any recent shift in gene therapy research? A: It's going in a lot of different directions. But importantly, it's in a phase now where there will be some clinical trials rolling out over the next 18 months that I think are really going to show some positive clinical effects. These are going to be in niche applications that have been cleverly designed. I suspect there will be some hemophilia trials that will be successful, where you put vector into muscle or liver--there must be three or four of those in development. The beauty of that clinical experiment is, if the gene is taken up and it expresses a protein, then you'll know that because you can just measure a blood sample from a patient before and after therapy. Or you can see if they bleed or clot. I suspect there will be some cancer successes. Early successes would be in regional [tumors]. I'm also intrigued by the angiogenesis stuff. It looks encouraging. They're in studies where they're starting to look at true efficacy. Those vectors may not be the right vectors and those genes may not be precisely the right genes, but it's an interesting application, where you only need a little bit of expression for a short period of time. Q: Why are you optimistic about success now? A: If you do a literature search for gene therapy papers and you just look at the numbers, ... it's going up like that [stretches his arms apart as wide as possible]. The clinical trials represent the application of technology. The trials ... in 1990 through 1996 were all based on technology that was on the shelf when we started working on this field in the late '80s. That technology actually will likely have successful clinical application. But we didn't know enough about it to know its limitations and its advantages, so we were applying it in the wrong places. Now that we learned more about the technology, we can apply it in the right place. For example, adenovirus doesn't give stable expression, so you should not use it for genetic disease. But adenoviral vectors may be useful when you need transient expression, such as in angiogenesis. So we learned and we adapted, and we now are applying technology where it ... ought to [be applied]. There are other vector systems now that have come forward because of this investment in research. Q: Last year you were very excited about one of those newer vectors--adenoassociated virus. What has happened with that vector since? A: It looks very attractive. It's very stable, safe, doesn't cause any inflammation. Q: What about the production problems? A:I think we're as far along as anybody. We've put a huge effort into that. We've gotten it to the point where we can make enough for a Phase I trial in most applications in humans. I think, just as of last month, we've modified that enough to scale it up into 10-liter bioreactors. If we can do that then we have enough to support a Phase II study. So, yes, I'm confident that we've solved the production problems. But it's been a long, long road.

Genomes and Gene Therapy: ASGT president ponders each field's effect on the other

Interested in reading more?

Become a Member of