|A key experiment has been approved, but many researchers worry that slipping genes into humans is premature.|
BETHESDA, MD.--Maybe W. French Anderson wouldn't be in the center of a slow-burning controversy if it weren't for the letters. But he can't escape them. Several times a week, new correspondence lands on his desk on the seventh floor of Building 10 at the National Institutes of Health in Bethesda. The letters come from all across the United States and from dozens of foreign countries, and they are all heartrending. Written by the friends, families, and physicians of those stricken with genetic illness, they ask one question: When will gene therapy live up to its promise to cure genetic ills?
Motivated by scientific curiosity and ambition as well as by these appeals, Anderson is searching passionately for an answer. He wants to forge ahead with his research by actually inserting new genes into people. But some scientists worry that the chief of the National Heart, Lung and Blood Institute's Laboratory of Molecular Hematology--and other gene therapy researchers--are moving too quickly, too soon. And the result has been a quiet but vigorous debate in committee rooms from NIH to the Food and Drug Administration. Explains Richard Mulligan, a retrovirus expert at the Whitehead Institute for Biomedical Research in Cambridge, "I think it's absolutely crazy to take the technology we have and think you can do an intelligent experiment on a human. I can guarantee you it won't be successful."
Last month, one small part of the debate was resolved when Anderson received long-awaited approval for an experiment that could eventually lead to gene therapy. But the sentiments of scientists like Mulligan are a sobering reminder of the approach's encounter with scientific reality. As little as three years ago, the notion of curing genetic diseases like thalessemia, Lesch-Nyhan, sickle-cell anemia, and even cancer by slipping in new genes was proclaimed by many--including Anderson--to be a medical miracle just around the corner. The idea seemed so simple and elegant that it captivated the imagination and hopes of researchers and patients alike.
But scientists discovered that gene therapy isn't so simple after all. Even finding a means of inserting a gene into a person--most approaches use a stripped-down retrovirus to deliver the new gene--has proven to be a quagmire of unforeseen problems, from ensuring the procedure's safety to finding the right disease to try it on. And even if they solve these problems, scientists face the monumental task of inducing the inserted gene to produce the right amount of enzyme (or other protein missing in each genetic disease) at the right time.
Partly as a result, a schism that cuts between scientific disciplines has developed about how to proceed. Researchers with medical degrees and a clinical background, like Anderson, favor heeding the desperate pleas of people stricken by genetic disease and proceeding with human trials as soon as possible--even though the science isn't completely understood. Molecular biologists and geneticists doing basic research, on the other hand, contend that their physician colleagues have let the quest to be first override better judgment and sound science.
The division became widely evident last October, when Anderson and principal coworkers Steven Rosenberg and Michael Blaese of the National Cancer Institute received permission from the NIH's Recombinant DNA Advisory Committee (RAC) to transplant foreign genes into humans. The experiment approved by the committee was not designed to actually treat a genetic disease. Instead it would have added a bacterial gene to certain cancer-fighting blood cells (called tumor infiltrating lymphocytes, or TILs) in order to help doctors monitor the cancer treatment's effectiveness. The marker gene is necessary, Anderson and his colleagues explain, because many patients with malignant tumors do not respond to the TILs. By using the gene to track where the cells go and how long they last, the scientists might be able to figure out why the treatment often fails.
In Anderson's planned experiment, the gene would be placed inside a stripped-down virus. Some of the TILs would then be exposed to the virus, causing them to incorporate the now harmless virus--plus the marker gene--into their own DNA. After scientists injected the altered cells into patients, the researchers could track the cancer-fighters' progress by testing for the marker gene (which produces a characteristic enzyme).
The inserted gene, of course, offers no therapeutic hope. But the experiment depends on the same techniques of getting a foreign gene inside the human body as does gene therapy. Anderson, for one, says that the experiment is a vital forerunner. "This is an important first step," he proclaims. "It's getting over the emotional hurdle of putting a human gene into a human." And once 10 marker patients have been evaluated, Anderson and his colleagues fully expect to forge ahead with gene therapy. "Once we've done our 10 patients, we plan to come in with a couple of gene therapy protocols," he says.
When NIH's Recombinant DNA Advisory Committee approved Anderson's experiment last fall, the scientist hoped to immediately put his ambitious plan into action. But, in fact, the committee's decision set off a controversy. Many members of the RAC have no gene therapy expertise. And their 16-5 vote in favor of the experiment went against the expressed wishes of the more-specialized Human Gene Therapy Subcommittee--which reports to the RAC (seven members overlap these two bodies) and which had previously voted to defer the idea until questions about the stripped-down virus's safety were better addressed.
Specifically, the subcommittee members worried that the virus mightcontain so-called helper viruses. These naturally existing strands of DNA have been described as roving spare parts and can potentially restore full operation to a lab-modified virus. Ironically, Anderson had by then already completed studies addressing the virus' safety--but withheld some details until the last moment for fear of jeopardizing impending papers in Science and the New England Journal of Medicine. So when the RAC approved the plan, some subcommittee members were upset. "The gene therapy subcommittee contains the special expertise in the area of gene therapy," explains member Paul Neiman, associate director for basic science at the Fred Hutchinson Cancer Research Center in Seattle. "The RAC needs that expertise before it acts--and it acted without benefit of advice of the committee." Aware of the subcommittee's objections, NIH director James Wyngaarden delayed his approval until new reviews of the safety data could be completed.
Anderson admits withholding information was a mistake. And for months, the scientist has been in and out of committee hearings, pleading his case in an attempt to get final approval. His perseverance finally paid off; last month, Anderson received four key endorsements; from the Human Gene Therapy subcommittee; the full RAC; the National Heart, Lung, and Blood Institute institutional review board; and an FDA advisory committee. Provided no other hurdles pop up unexpectedly, the scientist expects to start his experiment in February or March and finish within a year. "After seven months of intense review, it will be a relief to get started," Anderson says.
Still, the main part of the controversy--whether to go ahead with actual gene therapy--has yet to be resolved. Any attempt to perform such experiments will certainly be carefully scrutinized by many of the same scientists who questioned the TIL experiment.
These researchers have learned from hard experience that the underlying science necessary for successful gene therapy is still in its infancy. One problem is deciding whether to take the risk of something as unproven as gene therapy when other treatments might be available. Anderson, for example, had planned to do his first experiments with ADA deficiency, a life-threatening genetic disease that affects some 20 to 40 persons born each year worldwide. The disease kills because its victims lack a fully functioning gene for an enzyme called adenosine deaminase. Without this enzyme, toxic metabolites accumulate in cells and cripple the body's immune system. Many victims die in infancy.
Until recently, regular and risky blood transfusions or a bone marrow transplant were the only possible treatments--and neither guaranteed success. But in the past few years, the New Jersey firm Enzon Inc. developed a method to directly administer the missing enzyme. While the technique remains experimental, it renders the disease a less viable gene therapy candidate. "This now says you cannot just pick any kid with ADA deficiency for the first trial," states Duke University's Michael Hershfield, who has successfully treated nine children with the new procedure.
Anderson points out that the technique has only been used on the less critically ill ADA patients--and argues that gene therapy still may have a role to play. But should he propose inserting genes in ADA patients, other scientists almost certainly will question such an experiment.
Even more troublesome than finding the right disease for the first gene therapy trials is the technical problem of inserting the genes. Most research centers on using retroviruses as a delivery system. These viruses are adept at infecting cells and reshuffling genes--their own or their host's. So research teams around the world are striving to strip the viruses, which can cause cancer, of their lethal characteristics and put in the genes needed to correct human genetic defects.
The goal is to deliver these genes to blood cells, or more specifically their precursor stem cells in the bone marrow. Explains the Whitehead Institute's Richard Mulligan, "If you could successfully introduce a gene into that cell system, you'd effectively introduce it into the entire blood system--once those infected cells differentiated."
So far, however, the problems with this approach have been insurmountable. The gene-carrying retrovirus splices its genetic message randomly into the cell's DNA, so it's impossible to guide the transplanted gene into a specific spot in a cell's genes--and consequently to ensure it will be properly read. Not even Anderson's most successful primate experiments--in which the animals produced a tantalizing trace of the desired protein for a limited time--have given any indication that stem cells are infected. "We have to know a lot more about the biology than we thought we needed, a lot more about what elements to include in the virus to get it to express," says Dusty Miller, a gene therapy researcher at the Fred Hutchinson Center.
Such huge holes in the science underlying gene therapy only widen the chasm between basic researchers and clinicians. Gene therapy has been tried, without federal permission and with no apparent benefit, on four patients--in 1970, 1971, and 1980. The pioneer was Oak Ridge National Laboratory researcher Stanfield Rogers, who attempted gene therapy on two sisters in Germany who were suffering from a genetic defect that causes progressive retardation. When the first experiment failed, Rogers tried again a year later. But the technique was not promising, and criticism from scientists halted any further treatments. Nearly a decade later, in 1980, UCLA Medical Center hematologist Martin Cline also ventured overseas and tried gene therapy on two young women--a 23-year-old Israeli and a 16-year-old Italian--suffering from beta-zero thalessemia. Neither patient improved, and reviews conducted by both UCLA and the NIH concluded Cline had violated federal and university guidelines by not obtaining federal approval for the trials.
Today, few researchers give a new experiment any chance of achieving better results than the two previous cases. Still, those in Anderson's lab can't help but be affected by the messages in the letters that trickle into the office each week. Says former senior staff fellow Martin Eglitis, who left to join private industry but remains a collaborator, "These people don't see, even if there's a one-in-a-thousand chance, what the hell we're waiting for. It's hard to envision who this will impact other than the patient--and the patient's a goner."
Anderson puts it less dramatically. "The philosophy of some basic scientists is not to go into a human until we've done everything possible in animals and tissue cultures. But our motivation is to get the technology to the bedside as rapidly as we can."
The determination of Anderson and likeminded clinicians to bring gene therapy, however unproved, to patients might encourage those affected by genetic disease, but it adds fuel to the fires of controversy. "I don't think family desperation is any rationale for doing something stupid," says Dusty Miller. Adds Mulligan, "If you jump to the patient at this point you, preempt a series of experiments that could yield information that will eventually be useful for the patient," he asserts.
How will the debate be resolved? Much depends on the outcome of the experiment Anderson has just been cleared to perform--and on the progress in other labs of devising effective methods of gene delivery. But at least some experts believe that gene therapy will be tried again soon--long before its problems are solved. Bob Cook-Deegan, who follows the issue as an Office of Technology Assessment senior associate, predicts Food and Drug Administration approval--the final hurdle for human gene therapy--could come on the heels of receiving a formal proposal from the NIH researchers. Apparently the cold logic of science is no match for impassioned pleas on behalf of the terminally ill.
Robert Buderi is a freelance writer in San Francisco.