Although there is evidence that such a process occurs in animals, scientists have been unable to prove a similar virus exists in humans.

Nonetheless, the work has produced a wealth of biological insight. One is the idea that one oncagene, by means of a single protein product, can cause a cell to become cancerous.

Working with chickens, Erikson startled the scientific community when he identified the protein produced by the oncagene in one type of virus.

"He really opened up the field of oncagene proteins," Rohrschneider says. "He was able to identify the gene but now must use this to see how it turns a normal cell to a cancerous one."

Erikson made his first headway in the field while at the University of Colorado in the 1970s when he began working with chickens and viruses.

Although he spent most of his youth in the West as an undergraduate at the University of Wisconsin at Madison and as a graduate student at the University of Colorado Medical School, he eventually made his way to Harvard, which is noted for attracting the best cancer researchers from all over the world.

The large number of cancer researchers at Harvard and "the high quality of graduate students here" were enough to lure him East, he says.

According to Erikson, the idea that small changes within a cell could induce oncagenesis eventully allowed him to actually identify the oncagene and the protein it produces.

Shortly thereafter, J. Michael Bishop and his colleagues at the University of California at San Francisco found a virtually identical set of genes in cells of vertebrates including humans. And, in each case, a similar protein to Erikson's was produced.

Since then about 20 different oncagenes have been isolated from viruses that cause leukemia and other forms of cancer in animals such as chickens, mice, cats and monkeys. And once an oncagene is identified, researchers are able to discover its products within a few weeks, Frikson says.

In the past few years a second line of attack has also uncovered oncagenes as scientists have found genes in various tumor cells that can transform them into cancer cells. Presumably, each oncagene was instrumental in generating the tumor from which it was isolated.

"Right now, we may be nearing the end of identifying oncagenes," Erikson says. "This is the first step. Next we must understand how oncagene products change the way a normal cell behaves."

In particular, Erikson is focusing not only on the functions of oncagenes but on identifying the areas of a cell that they attack.

"Until we understand these pathways, we won't understand cancer," he says.

Once oncagenes are understood, the final step will be to design a drug that inhibits the protein formation process. This, however, may be the most difficult part of the puzzle, Cantley says, since all cells have these proteins, and if they are attacked and destroyed, normal body cells may fall victim to the drugs as well.