In what could prove to be one of the first major drugs to emerge from the human genome project, a chemical that singles out a specific gene thought to increase the risk of heart attacks has shown promising results in a preliminary test.

Last year DeCode Genetics, a gene-finding company in Reykjavik, Iceland, discovered among Icelanders a variant gene that doubled the risk of heart attack and stroke. Because a drug that inhibited the gene had been developed, although for a different purpose, DeCode was able to begin a clinical trial almost immediately.

Results of the trial, published today in The Journal of the American Medical Association, are that the drug seems safe and shows preliminary signs of effectiveness. The trial did not last long enough to have any effect in preventing heart disease, so researchers looked instead at a set of biological changes known to predict heart attack. The biomarkers of heart attack, as they are known, were generally reduced by the drug.

The researchers, led by Dr. Hakon Hakonarson, said they did not know whether the drug's ability to send the biomarkers the right way would translate into a decreased risk of heart attack. To answer that question, DeCode plans to start a new trial in the fall, hoping to see the first results in two years, its chief executive, Dr. Kari Stefansson, said.

If the drug, known as DG031, should prove effective, it will be among the first major fruits of the human genome project, the $3 billion effort to sequence the genome that was advertised to Congress as the royal road to develop treatments for common diseases. An editorial in the journal calls the test "one of the first human trials to attempt to translate genomic findings into clinical practice," at least for heart disease.

The drug works by inhibiting a gene known as FLAP, which is switched on by white blood cells when they detect infection. The gene helps activate chemical signals that cause inflammation, part of the body's defense mechanism.

People with the variant gene linked to heart disease produce larger amounts of the chemical signals. This was doubtless an advantage in the days before antibiotics. But when people grow older, the stronger inflammatory response raises the risk of heart attack, Dr. Stefansson said, possibly by rupturing the fatty deposits that build up in the arteries.

In Boston, Dr. Christopher J. O'Donnell, the author of the editorial and an expert on the genetics of heart disease at Massachusetts General Hospital, praised the study but said there were uncertainties about the rationale for the drug. One is that although some cardiologists theorize that inflammation is indeed a contributory cause of heart attacks, others regard it as just a symptom. If it is a symptom, a drug that reduced inflammation would do nothing to prevent heart attacks.

Dr. Stefansson said he had discovered the gene variant by scanning the genome for links with heart disease without any expectation about what kind of gene might be found. The variant, and by inference the inflammation it causes, must be a cause of heart attacks, not an effect, he said, or it would have shown no link to disease.

Several experts expressed reservations last year about a statement by DeCode that the FLAP gene was linked to heart disease. The reservations grew in part out of a larger debate among medical geneticists as to what standards of proof should be required for detecting genes that cause complex diseases like cancer.

Reflecting that debate, Dr. O'Donnell wrote in his editorial that further research should be conducted to confirm the link between the gene variant and heart disease. He said in an interview, however, that the new tests lent some empirical support to the DeCode position.

An original critic, Dr. David Altshuler of the Broad Institute in Cambridge, Mass., said the new trial had nothing do with whether FLAP variants had a role in heart attack, a question that could be resolved only by further studies.

Dr. Stefansson, while conceding no weakness in his original study, said he was waiting to publish three further studies, all of which confirmed the association. The studies were conducted at three locations in the United States.

If the drug proves effective, Dr. Stefansson said, he expects that it could be taken as widely as the statin drugs. The average risk for a man older than 40 of having a heart attack at some time in his life is 49 percent, Dr. Stefansson said, and although just 33 percent of Americans have the at-risk variant, many more might gain a protective effect from the drug.