Advances in the War on Cancer

By Claudia Wallis

New research finds promising leads in the genes of tumor cells

What causes nice, healthy, law-abiding cells to go berserk, proliferate wildly and thus produce the phenomenon called cancer? The answer, scientists have long suspected, lies in the genetic material of the cells. Somehow genes, composed of the molecule deoxyribonucleic acid (DNA), have been made abnormal, perhaps by such environmental factors as cigarette smoke and radiation.

In recent months, scientists in three fields have made remarkable progress in documenting how these genetic changes take place. Cell geneticists, studying chromosomes, the discrete units into which genes are packaged, have begun to associate specific abnormalities--an extra chromosome, a missing piece of a chromosome--with specific types of cancer. Virologists have clarified the role that viruses can play in altering DNA. And molecular biologists have used the new tools of genetic engineering to pinpoint precisely which gene, out of the tens of thousands present in every human cell, is responsible for causing a tumor.

Even more exciting is the fact that work in these three disciplines has begun to converge. The mood was electric when 400 scientists gathered last month at the University of Chicago for the Bristol-Myers Symposium on Cancer Research. Researchers reeled off findings that the journals have not been able to keep up with. Said Conference Chairman Janet Rowley, a geneticist at the University of Chicago: "Ten years ago, few of us had any notion the progress would be so rapid, even explosive."

The first breaks in cancer genetics came from the field of virology. Scientists have known since 1908 that a virus could cause malignant tumors in chickens. Over the decades, it was found that viruses could cause tumors in mice, cats, cows and a menagerie of other species. But not until 1980 did anyone identify a virus that causes cancer in human cells: Dr. Robert Gallo of the National Cancer Institute in Bethesda, Md., isolated a virus that can transform normal human white blood cells into the malignant type found in a rare cancer called T-cell leukemia. The same virus was found last year to be responsible for a relatively high rate of both T-cell leukemia and a form of lymphoma (cancer of the lymphatic system) in Japan. Says Gallo: "There is strong evidence that this virus will be important for a number of human cancers."

Researchers are also learning just how a cancer virus can alter normal cell DNA. Some cancer viruses contain cancer-causing genes, or oncogenes. When these genes are isolated and then transferred into healthy cell cultures in the laboratory, they create malignant cells. In the past decade, more than 15 oncogenes have been found in cancer viruses.

Yet many cancer viruses do not appear to contain oncogenes. These viruses seem to cause cancer in a less direct way, perhaps by altering genes that are already in the normal cell. Investigating this possibility, Microbiologists J. Michael Bishop and Harold Varmus of the University of California School of Medicine at San Francisco made an astonishing discovery: genes almost identical to the cancer-causing genes in viruses can be found in the normal cells of all manner of creatures, from fruit flies to humans. The supposition is that these harmless genes can easily be turned into the dangerous genes they so closely resemble. In this sense, says Bishop, "cancer may be part of the genetic dowry of every living cell."

Exactly what these potentially devastating genes are doing in healthy cells is not clear, but the fact that they are present in so many species suggests that they perform an important function. "God knows, they haven't been conserved through evolution just to make tumors," says Varmus. These genes may play a role during fetal development and then become quiescent, unless activated or altered by carcinogens or cancer viruses. They start out as "normal, good genes," explains Molecular Biologist Robert Weinberg of the Massachusetts Institute of Technology, "but they become damaged during an individual's lifetime. When damaged, they assume a new role: directing the cell to grow abnormally."

Weinberg has shown that only a minuscule amount of damage is needed to turn one of these normal genes into an agent of cancer. A gene from a normal human bladder cell contains about 6,000 chemical constituents or bases. The difference between this gene and one that produces bladder cancer involves only one of those 6,000 bases. Says Weinberg: "That very subtle change led to the creation of a tumor."

Competing laboratories are racing to identify the genes capable of inducing cancer in human cell cultures. About a dozen such genes have already been isolated from leukemic cells and from tumors of the lung, bladder, colon and breast. Many of these genes are nearly identical to oncogenes isolated years earlier from cancer viruses. Moreover, certain tumors (colon and lung carcinomas, for example) were found to contain the same oncogene. This suggests that perhaps several dozen genes are responsible for producing the 100 or more known forms of cancer.

Since all of the work with human cancer genes has been carried out in laboratory dishes, the role played by oncogenes in causing tumors in the body itself remains unproved. Evidence is, however, mounting. Researchers have known for two decades that certain forms of cancer are associated with certain visible changes in the 23 pairs of chromosomes found in human cells. Recent studies suggest, for example, that in some lung cancers a piece is often missing from chromosome No. 3. Better-documented changes occur in certain leukemias and lymphomas. In one form of chronic leukemia, a piece of chromosome 22 changes place with a piece of chromosome 9. In most patients with a cancer known as Burkitt's lymphoma, a piece of chromosome 8 has changed places with a piece of No. 14.

How or why these "translocations" might cause cancer has never been known, but the first real insight was announced at the Chicago conference. Researchers representing teams at Harvard and at Philadelphia's Wistar Institute reported that they had analyzed the piece of chromosome 8 involved in Burkitt's lymphoma. Both found that it contained a gene virtually identical to a cancer-causing gene isolated years earlier from a virus. The oncogene is located at the precise point where the fragment of chromosome 8 broke off. Thus the first link has been made between a known oncogene and an easily detectable change in the chromosomes of cancer patients.

Before these new discoveries can be applied, more must be learned about the precise function of oncogenes in the cell. Every gene in nature carries a code, or recipe, for creating a specific protein. If a single gene is responsible for producing cancer, it must be a result of the protein that the gene creates. Researchers are just beginning to figure out what kinds of proteins oncogenes make. Once they do, scientists may be able to develop tests to detect tiny traces of these proteins; such tests could allow the diagnosis of cancer at a far earlier stage than is now possible. Identifying the proteins made by cancer genes will enable biologists to create antibodies that specifically attack cancer cells and, perhaps, to block the creation of those proteins, thus stopping the cancer process itself.

--By Claudia Wallis. Reported by Jeanne Saddler/Washington and Dick Thompson/San Francisco

With reporting by Jeanne Saddler, Dick Thompson

This file is automatically generated by a robot program, so viewer discretion is required.