And downs.

A few years ago, investigators found the broken gene that causes cystic fibrosis, the fatal disease that has since attracted the interest of many in the gene therapy field. By introducing working copies of the cystic fibrosis gene into the lungs of patients, physicians hoped to correct the mutation and prevent the mucus secretions that lead to infections in the lungs. According to animal studies and test-tube experiments with human cells, the most promising delivery vehicle for those genes was an adenovirus, a virus that infects lung cells.

In the most thorough test of this strategy in humans so far, however, investigators have found that adenoviruses are extremely inefficient at shuttling the cystic fibrosis gene into nasal cells with characteristics similar to those of lung cells. The viruses successfully delivered their cargo to less than 1 percent of nasal cells, report Michael R. Knowles of the University of North Carolina at Chapel Hill and his colleagues in the Sept. 28 New England Journal of Medicine. Higher doses of the adenovirus did not help: They simply irritated the patients' noses, the researchers note.

Gene therapy researchers argue that they are constantly developing better gene-carrying viruses and should be able to overcome these problems. "It's important to remember that gene therapy is truly in its infancy and that the current tools are quite crude," Jeffrey M. Leiden of the University of Chicago wrote in an accompanying commentary.

That same issue of the journal contained more bad news on gene therapy. Genetically engineered muscle cells injected once a month for 6 months into the biceps of 12 boys with Duchenne muscular dystrophy did not restore any of the strength stolen by the disease, report investigators from Ohio State University in Columbus and the North East Wales Institute in Clywd. Researchers had hoped that by injecting the gene for a skeletal muscle protein called dystrophin into the boys' muscle cells, they could make up for the genetic defect that robs Duchenne patients of this protein.

The mixed results presented in the last 2 months are neither surprising nor discouraging for a field that is barely 5 years old, concludes Ronald G. Crystal of the New York Hospital--Cornell Medical Center in New York City. Investigators have shown that they can insert genes into humans, though inconsistently, and that those genes can function, he says. Studies such as the ADA experiments show conclusively that gene therapy can provoke in the body biological responses appropriate to the disease targeted. But Crystal acknowledges that no one has unarguably cured a disease yet.

"Gene therapy is based on solid science. And gene therapy is going to work. And it is going to revolutionize how we treat patients. When that will occur and in what form it will occur we don't know," says Crystal.