This May, when the US Food and Drug Administration (FDA) placed a hold on the first clinical trial that proposed using a product derived from embryonic stem cells, some blamed a tragic death in a gene-therapy trial almost a decade ago. Indeed, gene therapy was an ever-present spectre at a meeting held in April, when patients' advocates and industry watchers packed into a meeting of the FDA's Cellular, Tissue and Gene Therapies Advisory Committee.

The purpose of this rare open meeting was to determine how to decide whether products derived from embryonic stem (ES) cells were ready to test in patients. One worry was that the stem cell field might make the same missteps that the gene-therapy field did, which could cause it to suffer the same punishment: becoming a stalled field that is mistrusted by the public and shunned by investors. Some scholars and stem cell critics who followed the tragedies of early gene-therapy trials see history repeating itself.

Not so long ago, gene therapy was the 'Next Big Thing' in medicine. By the early 1990s, scientists knew how to sequence, synthesize and recombine DNA, and a number of disease-causing genes had been isolated. These new technologies, plus breathlessly reported implications for human health, helped propel the first clinical trials of gene therapy in the early 1990s. What better way to cure disease than to replace patients' disease-causing genes with healthy counterparts in the appropriate cells and tissues?

“Our system of protection has not kept up with the number of trials. Stem cell science is just too early. We're ploughing ahead like blind people.” Adil Shamoo, University of Maryland

But the science was young and had only just begun to be tested. Jesse Gelsinger was the first to die. During a gene-transfer trial sponsored by the University of Pennsylvania in 1999, he succumbed to organ failure triggered by an overwhelming immune response to the adenovirus carrier. Gelsinger had a mutation that caused a metabolic disease, but the happy 18 year old was not critically ill at the time; his life was unquestionably cut short by the trial.

According to the Gelsinger family and their supporters, a major problem was the way investigators had described the risks of the procedures. In an article published this year, Paul Gelsinger (Jesse's father) and Adil Shamoo, a bioethicist at the University of Maryland in College Park, wrote that the informed consent “did not mention any serious reactions in humans, and conversations with the doctors led Paul and Jesse to believe the therapy was safe”1. This, despite the fact that work in monkeys had revealed the same kind of reaction believed to have caused Jesse Gelsinger's death.

Retracing old paths

Stem cell trials could easily follow similar paths, says David Magnus, who directs the biomedical ethics centre at Stanford University, California, and who was working at the University of Pennsylvania at the time of Gelsinger's trial. He recalls, “The FDA failed to halt the trial after they received reports of adverse events with the same vector from other trials, and the investigation [following Gelsinger's death] exposed financial conflicts of interests.” He pauses and adds: “And there was a fair degree of investigator hubris.” Here is where Magnus sees the potential for a gene-therapy sequel: a powerful technology in the hands of an overenthusiastic investigator with insufficient oversight.

Following the Pennsylvania tragedy, the FDA slammed the brakes on the nation's gene-transfer studies. But Shamoo, who heads a patient advocacy group created a year before Jesse Gelsinger's death, believes that clinical researchers have not learned their lesson and even feels that trials of autologous, or patient-derived, stem cells have moved very quickly. “Our system of protection has not kept up with the number of trials,” he says. Current systems react to, rather than anticipate, potential dilemmas. “Stem cell science is just too early. We're ploughing ahead like blind people,” he warns.

Three years after Gelsinger's death, tragedy struck again: insertional mutagenesis plagued an ex vivo retroviral stem cell trial treating children with X-linked severe combined immunodeficiency (X-linked SCID). Of about two dozen children enrolled in the experiment, four developed T cell leukaemia and one of these died. Although the study was based in France and Britain, the FDA again placed a halt on all US trials using retroviral vectors in stem cells, but it eased the ban in 2003. The remaining children appear to be SCID-free, and the three with leukaemia are in remission.

Most recently, Jolee Mohr, an otherwise healthy woman with rheumatoid arthritis, enrolled in a phase I/II study assessing the toxicology of a gene carried by an adeno-associated viral vector. She died in July 2007 of multiple organ failure caused by histoplasmosis, a common and usually mild fungal infection. The Recombinant DNA Advisory Committee (RAC) of the US National Institutes of Health (NIH), in Bethesda, Maryland, which was created in 1974 to review gene-transfer protocols in the United States, investigated Mohr's death. In this case, the RAC concluded that the procedure was not likely to be the cause of death, though it could not be ruled out. The FDA allowed the trial to resume. However, the committee's report underscores the uncertainty manifest when death occurs in a first-in-human trial.

To some bioethicists, stem cell trials seem to carry the same baggage as the early years of gene therapy. Potential treatments have been overstated by some researchers and hyped by the media. Transplanted cells carry risks: immunogenicity from allogeneic transplants, unrestricted cell growth, ectopic differentiation or unanticipated effects could endanger trial participants. But researchers' language downplays these risks and can promote a therapeutic misconception. For example, 'gene transfer' is a more accurate term than gene therapy, as no licensed gene therapies exist. However a search of 'gene therapy' interventions in ClinicalTrials.gov yields 330 studies, and the term is mentioned in more than 1,500 protocols. Only 93 trials and 110 protocols use the less loaded term 'gene transfer'. Similarly, the term 'stem cell therapy' is found in an increasing number of registered trials.

Stem cell trials raise other ethical issues. Ethicists worry about balancing the interests of two groups: the donors of cells, gametes and embryos, and the trial volunteers. Among other things, donors must be fully informed of how their biological materials could be used so they can document their consent. If cells are to be banked as a source for transplants in the future, questions about privacy emerge. Years may pass from the time cells are donated before therapies are fully realized. Future patients may be at risk from previously unknown diseases transmitted through donor cell lines. And unlike generalized DNA sequences, transplants come from cell donors, leading to questions such as: How will a donor feel about having his or her genetic information immortalized in a cell line? Potential trial volunteers could have moral reasons to decline a transplant derived from a destroyed embryo. Physicians could also refuse to do transplants for the same reason.

Improving oversight

With new trials on the horizon, there is a debate among policy researchers about how much oversight is appropriate for stem cell trials. US politics has caused a regulatory vacuum at the federal level and no coherent framework of regulation. For research in the United States, stem cell oversight research committees voluntarily follow guidelines from the National Academy of Sciences and the International Society for Stem Cell Research (see When the past catches up with the present) and abide by a patchwork of state regulation and law. For trials involving human patients, institutional review boards must approve the procedure. If cells are given to anyone other than the person from whom they were taken, or if cells undergo significant manipulation in a laboratory, FDA approval is required as well.

Shamoo joins some who favour a stronger system of national oversight. Since its creation the RAC has reviewed gene-transfer trials, but it has no authority to approve a study. Jeffrey Kahn, a bioethicist at the University of Minnesota in Minneapolis and a RAC member, sees the gene-transfer case as “a good trial run for stem cell therapies.” However, if transplanted cells are not genetically modified, the trials are unlikely to be overseen by the RAC.

John Robertson, an ethicist and legal scholar at the University of Texas in Austin reminds us that originally the RAC was born out of concern that recombinant organisms would escape the lab. “We're a lot further down the regulatory road now,” Robertson says. “I'm not sure the dangers are so great that we need another level of review and bureaucracy at this point.”

Alan Milstein, the Gelsinger's lawyer, admits that research universities have beefed up oversight since Jesse's death. If the first stem cell trials are done at top universities, he believes, risk could be managed. But he also points out that these institutions tend to have the most serious problems, and there's little in the way of a safety net. He calls federal regulation a “total failure.”

Precedents for stem cells are still slim. More than three months after the FDA placed a clinical hold on an ES cell product made by Geron, based in Menlo Park, California, the company has not disclosed the agency's concerns, and the FDA is legally barred from discussing them. Jonathan Kimmelman at McGill University in Montreal, Canada, who studies the ethics of gene-transfer trials, argues for the transparency of a national system of review. A click on a link on the NIH's website will display a RAC deliberation; not so for clinical trials reviewed by the FDA, which is bound by confidentiality. “A national review reduces uncertainty, pools disparate findings and calibrates ethical assessments,” he contends.

Ultimately, of course, the issue is whether patients have the information they need to make reasonable decisions. But a patient's assessment of their disease, and the relative risks associated with a clinical trial aimed at ameliorating it, may be different from that of the general public. Newly disabled people (and those who care for them) may underestimate the emotional impact of a recent injury. As a result, patients might be more likely to agree to a trial soon after an injury, whereas the same trial might seem too risky once the person has adjusted to, say, life in a wheelchair. For those living with a debilitating disease, the hope of any incremental benefit from a safety trial — however remote — might outweigh considerations of risk. The risk/benefit calculations are difficult: restoration of bowel function in a spinally injured patient is a significant improvement in their quality of life. On the other hand, an inoperable tumour caused by a transplant may mean a lifetime of peripheral pain.

One thing is clear: policy on the ethics of stem cell research is being set where it happens rather than in the federal arena. At Stanford University, the stem cell oversight research committee regularly faces new challenges. Now that stem cell trials are upon us, the committee will work with Stanford's institutional review board to determine the appropriate levels of risks and benefits for patients.

So until some federal policy is set, states supporting ES cell research are going it alone. The weakness in this approach is that crucial ethical and policy decisions from big research universities are imperfectly communicated to other committees, agencies and nations. However, a strong argument can be made in support of a system that makes policy from the ground level. With the benefit of the various strengths in the field, Stanford University — and other institutions grappling with the same challenges — is in the best position to drive a national discourse about the next big issue facing the ethics of stem cells: the first-in-human clinical trial.