In the war on cancer, researchers have long struggled to recruit the body’s immune system to attack tumors while leaving healthy cells alone. This strategy-basically that of vaccination-would offer a huge improvement over existing treatments such as chemotherapy and radiation, which kill healthy and diseased cells indiscriminately and carry harsh side effects. But repeated attempts to find a cancer vaccine have failed, largely because scientists have had a crude understanding of the molecular mechanics of the immune system and cancer cells.

Today, armed with a growing understanding of how to manipulate the immune system, researchers are offering the glimmer of a hope that cancer vaccines could soon become part of the cancer-fighting arsenal. More than 50 cancer vaccines are being tested in the United States, Canada, and Europe against several types of cancers, including melanoma and kidney, lung, breast, and prostate cancers. Several are in the final stages of human trials, at least two of which are expected to conclude within a year. In some of the trials, a few patients have seen their cancers go into remission, while in other patients, the vaccine slowed the spread of the disease. If all goes well, the first cancer vaccine could be ready for general use in three to five years.

“We are learning more and more about how to turn on the immune system and how to regulate it to get an antitumor effect. That’s why we have so many potential new vaccines being pursued,” says Antoni Ribas, an oncologist  developing a melanoma vaccine at the University of California, Los Angeles.

Cancer vaccines, unlike conventional vaccines, are designed not to prevent disease but rather to treat it. The immune system is normally relatively tolerant of cancer, and the trick is to get it to see tumors as the enemy. Approaches include manipulating tumor proteins and specialized immune cells in the lab, so that when they are put back into the body, they “teach” the immune system to see the tumors as foreign. Earlier attempts at cancer vaccines often used whole tumor cells, which present a mixed bag of unknown proteins to the immune system. Newer vaccines use a smaller number of extracted and purified proteins, which some researchers say promise to be more effective and predictable.

So far, successes have been modest but promising. One of the efforts that’s gone farthest is a prostate cancer vaccine from Dendreon of Seattle, WA. In one trial, researchers found that a subset of patients with less aggressive tumors went a median of seven weeks longer than untreated patients before the disease progressed. A second trial focusing only on patients with less aggressive tumors is under way. If results are favorable, the company expects to file for regulatory approval with the U.S. Food and Drug Administration by late next year. Another company, Antigenics of New York City, is now conducting the final stage of human trials of kidney cancer and melanoma vaccines. The company expects initial results from the kidney cancer trial later this year.

At their most optimistic, of course, cancer researchers hope the new vaccines will stop and shrink tumors. That has happened in a few isolated cases in some trials involving melanoma and kidney cancers. But more likely is that these new vaccines will be part of a larger treatment strategy, a way to mop up cancer cells after surgery or chemotherapy, or to reduce the need for chemotherapy or radiation in the first place.

In this respect, vaccines are seen as the next up-and-coming cancer therapy, following a group of drugs that use proteins called monoclonal antibodies. Monoclonal-antibody drugs, six of which have made it to market, bind specifically to cancer cells and either slow down their growth or mark them for destruction by the immune system. But patients have to take the drugs continuously, and they’re expensive.

With a vaccine, by contrast, patients would in theory only have to get a few shots; their own immune systems would do the rest. David Urdal, president and chief scientific officer of Dendreon, says cancer vaccines today are where monoclonal antibodies were ten years ago. “We’re now at that threshold with the cancer vaccine,” he says.

Some recent cancer vaccines have proved disappointing, however. A breast cancer vaccine from Biomira in Edmonton, Alberta, failed recently in advanced trials. And although a melanoma vaccine from Corixa of Seattle is approved in Canada, the FDA said approval in the U.S. would require a second trial, which Corixa may or may not attempt.

If other cancer vaccines work better, they might still pose problems. Vaccines could prime the immune system to go after not only cancer cells but also healthy ones. And some vaccine strategies require the isolation and purification of immune cells or tumor proteins from individual patients, which may make the treatments prohibitively expensive and labor intensive. But perhaps most worrisome is that, as vaccine developers well know, cancer cells are cunning creatures. “They have a lot of escape mechanisms to overcome obstacles generated by a vaccine,” said Steven Rosenberg, a researcher at the National Cancer Institute in Bethesda, MD, who is testing vaccines and other ways of recruiting the immune system to fight cancer.

And even proponents say cancer vaccines probably won’t be enough to fight cancer alone. Vaccines would have to work in combination with other therapies, bombarding tumors from all sides. Still, after decades of frustrated research efforts, recent advances are raising cautious hopes that these new therapies will soon take their place in the cancer-fighting arsenal.