The hearings held by the US National Institutes of Health (NIH) last week into problems that have arisen in adenovirus-based gene-therapy trials revealed clear breaches of protocol by some of the field's leading researchers, and inadequacies in the state of experimentation (see page 707). Adenoviruses are one of several types of vectors being tried in clinical gene-therapy studies. The case of Jesse Gelsinger, whose death during a gene-therapy trial led to last week's hearing, reveals how poorly understood are the body's responses to those vectors in particular. But the uncertainties and the violations revealed by the hearing should not halt the pursuit of the adenovirus approach, whose particular advantages include the fact that they can infect non-dividing cells.

Given gene therapy's highly experimental and controversial nature, demanding standards of openness are appropriate. Last week's hearings underscore the desirability of giving the NIH's Recombinant DNA Advisory Committee (RAC) more power to examine adverse events as they arise, thus ensuring a public airing of problems.

The NIH has made commendable efforts to make itself accessible — all RAC meetings are open to the public and their minutes are on the web. Given the controversy of the past few months and the worries and even hostilities that experiments in genetic manipulation can engender, the NIH can usefully go further. It should establish itself as a model provider of readable and easily navigable information, providing on its website a substantial overview of the genetic basis of diseases and the state of understanding of respective gene therapies. Considering the amount of unreliable information offered to the public on the web (see page 722), and especially the interests of those members of the public more directly concerned — patients, for example, or their relatives and dependents — the NIH should use the website routinely to give detailed but well-signposted and comprehensible information on the state of trials, covering progress and problems worldwide. Such a programme of enhanced access requires a small budgetary commitment compared with experimental funds. But it could come to play a major role in sustaining both informed consent and public trust.