LONDON — The Nuffield Council on Bioethics Round Table Group — a panel of experts in patent law, philosophy and biotechnology — has just finished an in-depth investigation into gene patenting that took two years to complete. The result is a discussion paper, called The ethics of patenting DNA, that attempts to set out an ethical framework for gene patenting that will stimulate innovation for the public good but at the same time rein in the worst excesses of commercial exploitation.

The group was set up because of growing concerns that patents are being granted to biotech firms before they have any real idea of a gene's potential. This practice is thought to account for a significant number of the tens of thousands of patents already held by private companies and has been likened to the snapping up of internet domains for sale at vastly inflated prices during the dotcom boom of the late nineties.

To explore the issue, the Nuffield Council on Bioethics brought together a distinguished group of experts from most of the areas involved in gene patenting. Members included Martin Bobrow, head of medical genetics at Cambridge University, Michael Stratton, head of the Cancer Genome Project at The Sanger Institute and High Court judge the Honourable Mr Justice Jacob.

Their brief was to explore whether DNA patents were justifiably being granted and how best to regulate the system in future. The key recommendation appears to be that the rules need to be more rigorously applied so that the numbers being issued for DNA sequences are substantially reduced. The report makes it clear too many patents are being granted, many are of "doubtful validity" and the criteria set down for deciding if they are warranted are not being strictly adhered to. It says patents for DNA sequences should be the exception rather than the rule and just because they are eligible for a patent, doesn't mean "each and every application should be granted."

It also expresses fears that gene patenting could adversely affect healthcare and research by making diagnostic tests and medicines unaffordable or unavailable, slowing down the development of new tests and treatments and inhibiting research. Sandy Thomas, director of the Nuffield Council on Bioethics said: "We are concerned that, for patents involving DNA, the patent system is in danger of not achieving its main goal — to stimulate innovation for the public good. We believe these recommendations would safeguard against the situation."

But many observers will wonder if the call for action comes too late. Thomas said although thousands of patents have been issued, there still remain many sequences of unknown function that, in the present climate, could be patented. "There's still a lot to play for so it's important we look at all the issues now," she said.

So how do the experts involved hope to stem the tide of gene patenting? Mainly by redefining the three legal tests by which an application must be judged — that it is novel, inventive and useful. For example, when it comes to inventiveness, the report argues, in many cases it's now possible to identify a gene simply by using a database compiled by other scientists. This hardly counts as inventive, it says, and patent applications based on this practice should be rejected.

It also questions whether patenting DNA sequences — either as a potential diagnostic test or target for gene therapy — is justified when a gene is associated with a particular disease. The degree of invention involved, it argues, is minimal. "We feel that once the link between gene and disease is identified, the use of the gene for diagnosis or therapy is obvious," says Bobrow. "It therefore does not fulfil the inventive (or 'non-obvious') test. Patents involving a DNA sequence as a diagnostic test for gene therapy should rarely be granted. In the case of a diagnostic test, a patent on the use of the sequence may be acceptable but the patent should be restricted to the test specified."

A key issue facing the group was the patenting of DNA sequences for research purposes, rather than specific therapeutic use. This is a controversial area because, as the panel of experts found, granting patents for research can give claimants untold rewards if the same sequence is later found to have much wider uses. "Some medicines, or therapeutic proteins, can be produced directly from a DNA sequence, for example insulin, and in such cases the test for usefulness is fully met," says Bobrow. "A patent on the DNA sequence itself, to produce a specific protein, is justifiable." But the report adds patents for DNA sequences to be used as research tools "should be discouraged."

Bobrow added: "It is vital that the tests of inventiveness and usefulness be more rigorously applied. If this happens, it should substantially reduce the number of patents granted involving DNA sequences."

The Medical Research Council said the call for a tougher stance on patents was in line with its existing policy. A spokesman said: "Some patents that have been granted have been too broad and we've been concerned about that. They could slow down research. We are supportive of patents for research only if the application can produce data that shows the genetic material involved has a clearly defined function."

The Royal Society will not comment until it has considered the report but is likely to be supportive of the findings. President of the Society, Sir Robert May, expressed concerns about the widespread patenting of DNA sequences when the human genome was published: "Of course, when someone elucidates the actual function of a particular gene, and develops a product based on this knowledge, such invention should be protected by intellectual property rights. But if tricky arguments are successfully used to claim patents on bits of uninterpreted sequence data, then subsequent research that develops products from knowledge of gene structure and function will be clouded by a myriad of legalistic claims for co-ownership."