Protocol #9512-139, the ill-fated attempt at gene therapy that led to the September death of a young patient, gives the biomedical community pause for thought. But it does not change the fact that gene therapy is a relevant and important area of investigation that is ready for testing in carefully designed human trials. By accepting the lessons that this death teaches, the gene therapy community could emerge stronger and better-prepared to advance this emerging field of medicine.

Since its formation in 1975, the US Recombinant DNA Advisory Committee (RAC), the National Institutes of Health office responsible for overseeing recombinant DNA technologies, has advocated the open and public discussion of advanced therapeutic products and protocols.

Gene therapy, which has only been around for 10 years, has grown up in this environment. As researchers have attempted a bewildering variety of protocols in an attempt to demonstrate the first successful treatment of a patient with gene therapy, the culture of public oversight and disclosure has served them well.

From the early days in which the concept of introducing a gene to compensate for a missing or defective gene seemed wonderfully simple and attainable, through the period when researchers were forced back to the bench to better understand the nature of transgene and vector expression, persistence, toxicity and immunity, to the current efforts to design more-sophisticated delivery systems, empiricism has played a very important part in the great gene therapy experiment. It would not have been a good idea to foist this level of experimentation on the public. But by openly discussing the limitations and problems of the technology, the gene therapy community has been able to carry a general acceptance and at times enthusiasm for a technique that has for a decade failed to deliver.

But now that same community must also publicly acknowledge the first death clearly attributable to the experimental therapy, and a willingness to take on board the lessons that this first gene therapy tragedy offers.

This process has begun with an open RAC meeting (8–10 December 1999), at which James Wilson and his University of Pennsylvania team that conducted the trial gave an honest and open explanation of the events surrounding Jesse Gelsinger's death (see News, page 6). This alone has highlighted some important lessons for all gene therapists. Although most of these observations have only become clear with the benefit of hindsight and the level of scrutiny that follows a tragedy of this magnitude, they are worth examining.

Jesse Gelsinger suffered from a partial deficiency of ornithine transcarbamylase (OTC) and was given 6 × 1011 particles per kilogram (the highest amount permitted under the trial protocol) of an E1/E4-deleted adenovirus bearing the OTC gene. The amount and type of virus and the disease are all very relevant to the safety of the protocol.

As many researchers have struggled to achieve therapeutically relevant levels of transgene expression, one approach has been to transfer larger and larger amounts of viral vector. So far, these escalating doses have been well tolerated. But every therapeutic has a dose limit, and it is now important to find out if the more than six trillion viral copies given to Gelsinger was simply too much.

Furthermore, adenoviruses are known to deliver only transient expression. Yet OTC is a key enzyme in the urea cycle and as such is needed throughout life. Many would question whether an adenovirus was the right vector to use in these circumstances, as repeat administration would have been called for and repeat administration is known to give rise to immunological rejection of this virus.

Whereas these and other aspects of the trial warrant further invesigation, perhaps the darkest cloud on the horizon is the broader issue of secrecy versus disclosure.

As the RAC has struggled to hang on to the concept of public disclosure, more commercially minded gene therapy researchers have attempted to circumvent the rules, claiming that full disclosure would require them to provide trade secrets and would therefore put them at a competitive disadvantage. Others have argued that full disclosure of all adverse events, including those not considered linked to the gene therapy, would be misinterpreted by the press and the public and would unnecessarily scare the public. And some have indulged in amateur semantics, advocating the reporting of only ‘serious’ as opposed to ‘severe’ adverse effects.

The death of Jesse Gelsinger is a clear wake-up call. Like most medical interventions, gene therapy is potentially dangerous. It is also in its infancy. This combination makes it paramount that a single central agency with the resources and expertise to interpret the data, receive reports of all adverse events. To do otherwise not only runs the risk of missing unsuspected or surprising adverse events but also sends a dangerous message to the public, who are unlikely to have confidence in a procedure from which they are excluded. And if the importance of this combination of good science and public confidence is doubted by the gene therapy community, they need look no further than the ongoing debacle over genetically modified crops.

Confidence is growing that successful gene therapy protocols (probably for the treatment of hemophilia) are just around the corner. This leaves just enough time to pick up the pieces left by the first attributable gene therapy death and convince the public that the gene therapy community put safety ahead of secrecy.