Putting a tumor suppressor back to work.

The gene known as p53 has been called the guardian of the genome. When a cell turns cancerous, for example, p53 often turns on and directs the cell to refrain from dividing or, more drastically, commands the cell to commit suicide.

Yet not all cancer cells meekly obey p53's orders. Many silence the gene by deleting or mutating the two copies that most cells possess. Indeed, more than 50 percent of all tumors contain cancer cells having no functional p53.

What would happen if scientists popped working copies of p53 into such tumors? This tantalizing gene therapy strategy has now had its first test in humans, and investigators report that it temporarily prompted tumor regression in three out of seven cancer patients who completed the study, and it checked tumor growth in three others.

These encouraging results were far from a cure, however. All seven men, each afflicted with widespread lung cancer unresponsive to conventional therapies, eventually died of the cancer or complications relating to it.

Still, the primary intent of this initial p53 gene therapy trial was to establish the strategy's safety, and in that respect, the trial appears a success. "There weren't any toxic effects associated directly with the treatment," says Jack A. Roth of the University of Texas M.D. Anderson Cancer Center in Houston.

Roth and his colleagues packaged p53 genes into harmless but infectious viruses. They injected those viruses into inoperable, life-threatening lung tumors that had not responded to radiation or chemotherapy. The hope was that the viruses would infect cells in and around the tumor and deliver a cargo of p53 genes. Cancer cells, no longer free of p53's influence, would undergo apoptosis, a form of cellular suicide. In contrast, healthy cells receiving an extra copy of p53 would continue a normal life.

With studies in test tubes and in animals, Roth and other researchers had shown that this strategy bore promise. In the human trial, biopsies of the treated tumor sites showed evidence of increased apoptosis in six of the seven patients, Roth's group reports in the September Nature Medicine.

In several of the patients, the number of cancer cells that died was greater than the estimated number of virus-infected cells. Roth's group had reported a similar result in their earlier studies on animals.

"We see tumor regression that exceeds expectations. That is very surprising," says Roth.

Roth suggests that cells receiving a p53 gene from a virus may induce apoptosis in neighboring cancer cells. "This is an area of very active research, and we're in the process of identifying the factors involved," says Roth.

Not all cancer researchers have observed this bystander effect, notes Larry M.

Kaiser of the University of Pennsylvania Medical Center in Philadelphia.

Kaiser's group has almost completed its own initial clinical trial of the p53 gene therapy strategy.

In that study, as well as in a new one being conducted by Roth's team, the researchers are using a p53-delivery virus different from that used in Roth's initial trial. The new virus, a member of the cold-causing adenovirus family, infects cells more efficiently, says Kaiser.

Another planned modification of the gene therapy strategy is to combine it with radiation or chemotherapy. Some cancer cells that receive a working p53 gene may not immediately commit suicide, but the added gene may make them more sensitive to conventional treatments, explains Roth.

While p53-carrying viruses may aid the treatment of localized tumors, they will likely have difficulty reaching and eliminating cancer cells that have spread beyond the initial tumor site, notes Curtis C. Harris of the National Cancer Institute in Bethesda, Md.