Molecular Revolution

By Claudia Wallis

Last week's miracle-in-mice may have launched a thousand premature hopes, but there's no doubt in the minds of cancer researchers today that a new era is dawning in the treatment of the U.S.'s No. 2 killer. Three decades ago, the Federal Government's "War on Cancer" underwrote basic discoveries about the ways broken-down genes lead to malignancies. Now that work is beginning to pay off. "The black box that was the cancer cell has been opened," says Dr. Bert Vogelstein, a world-renowned investigator of cancer genes at the Johns Hopkins University in Baltimore, Md. "As researchers, we feel a tremendous amount of hope, probably for the first time in the history of cancer treatment."

In pharmaceutical-company research departments and academic labs around the country, scientists are feverishly at work on drugs that target the products of specific genes--the very genes that make a cell cancerous. The hope is that these treatments will be more effective, longer lasting and far less toxic than traditional chemotherapy and radiation--treatments that inspire dread so deep that they are almost as feared as cancer itself.

"Because of the early success with chemotherapy in some forms of leukemias and lymphomas," says Dr. Dennis Slamon, of the Revlon/UCLA Women's Cancer Research Program, "we have been slugging at cancer that way for 25 years. We didn't make any significant inroads, and in some cases, we ended up killing people. Now we are beginning to look specifically at what's broken in a cancer cell and trying to target that."

What's broken in cancer cells is genes, usually genes that control some aspect of cell growth and division. Hundreds of genes play a role in this process, and more than three dozen have been identified as playing a role in cancer. Some are like accelerators, telling cells to grow, grow, grow. Others put the brakes on growth. Some regulate steps in cell division to make sure that DNA is copied correctly from mother to daughter cell. Some play executioner, killing mutant cells in which the copying has gone awry. Cancer is caused by errors in these genes, usually multiple errors. Though some of these errors may be inherited, most are acquired during years of living. Sunlight, cigarette smoke, environmental toxins and aging itself help these errors accumulate.

Since every gene holds the recipe for a vital protein, corrupt genes mean corrupt proteins: too much of one protein, too little of another, or a misshapen protein that doesn't function properly. The new generation of cancer drugs takes aim at these defective proteins, blocking them, disrupting them in myriad ways. Unlike old-fashioned chemotherapy drugs, the new substances don't poison the tumor--an approach that usually causes collateral damage to healthy cells. Instead, they aim to halt the processes that make a cancer cell act like a cancer cell in the first place.

Barbara Bradfield, 55, is living proof that this can work. A teacher turned homemaker in La Canada Flintridge, Calif., Bradfield is one of the lucky cancer patients who have already benefited from the new generation of gene-based treatments. She was 47 years old when she discovered a large lump in her breast. Tests showed that the malignancy had spread to her lymph nodes. Bradfield got the works: a double mastectomy and six months of chemotherapy, followed by radiation and then more chemo. It bought her 18 months of symptom-free life. Then one hot August night, she recalls, "I went to rub my neck, and there was a tumor about the size of a marshmallow." Bradfield was already depressed--her daughter had just died in a car accident--and she never wanted to face chemo again. "I thought I was probably going to die, and I didn't want to die bald and throwing up."

As it happened, Bradfield's tumor cells had a characteristic present in about 30% of breast-cancer cells: too many copies of a gene known as HER-2/neu. This gene makes a protein that helps relay the signal telling cells to divide. Having too much of it is associated with an especially rampaging, hard-to-treat cancer. Once this form of breast cancer metastasizes, a patient typically has just six to 12 months to live.

Bradfield's doctor put her in touch with UCLA's Slamon, who was testing a brand-new antibody that targeted the HER-2/neu protein. Although Slamon was using the antibody in combination with chemotherapy--and Bradfield was loath to go back to chemo--the combined therapy proved miraculous in her case. Sixteen small tumors in her lungs melted away. By 1993 she was in remission, and still is. "I got to be at my son's wedding," she exults. "The gift is that I'm here!"

The antibody is not a panacea. It didn't work as well for Bradfield's fellow guinea pigs in the initial study. But results of a just completed trial with 470 women do show it to be a significant improvement over chemo alone for women with this awful form of breast cancer. The details of the study will be revealed by Slamon this Sunday at a meeting in Los Angeles of the American Society of Clinical Oncology. Manufactured by Genentech under the name Herceptin, the drug is on a fast track for approval by the FDA, perhaps before year's end.

Herceptin, if approved, will join the lymphoma drug Rituxan, also an antibody, as the first of the new gene-based therapies to make it to market. Rituxan, made by IDEC Pharmaceuticals, was approved late last year.

Other substances in the works may be further from the market, but they are in some ways even more exciting. Several of them take aim at a growth-signaling protein made by a gene called RAS (for rat sarcoma, the cancer in which it was first discovered). In about 30% of cancers, the RAS protein is stuck in an "on" position, mindlessly ordering the cell to divide again and again. It plays a role in 90% of pancreatic cancers, 50% of colon cancers and 25% of lung cancers. Dr. Edward Scolnick discovered the RAS gene in rats while working at the National Cancer Institute in 1978. Now president of Merck Research Laboratories, he is overseeing the development of a drug that stops the RAS protein from sending its malignant message. Several other big drug companies, including Bristol-Myers Squibb, Johnson & Johnson and Schering-Plough, are testing similar drugs. "We think the odds are that if you treat people with a good RAS drug, you will produce some clinical benefit," says Scolnick. Finding a new kind of cancer therapy based on gene discoveries like his own is, Scolnick admits, "my fondest hope."

Other drug companies are targeting another common cancer gene: one that codes for a protein called the EGF (epidermal growth factor) receptor. This receptor, which takes in growth signals and relays them to the RAS protein, is found in abnormally high numbers on the surface of about 40% of tumor cells, including about 90% of lung-cancer tumors, some prostate tumors and other malignancies. Researchers at M.D. Anderson Cancer Center in Houston are testing an antibody to EGF receptors in patients with advanced head and neck cancers. But most other groups, including teams at drug makers Pfizer, Novartis and Zeneca, are using smaller molecules that, unlike antibodies, could ultimately be taken orally. "We have a very exciting tablet that is taken once a day," says Dr. George Blackledge, head of new cancer projects at the Zeneca Group. Testing on patients is still at a very early stage. "We have to be cautious," he says, "but potentially this could be an effective new treatment for the most common type of lung cancer."

At the Memorial Sloan-Kettering Cancer Center in New York City, Dr. Mark Malkin is working with a substance that targets a receptor for another growth factor called PDGF (platelet-derived growth factor). This receptor studs the surfaces of cells in certain ovarian, prostate, lung and brain tumors. Malkin has been testing the drug, SU101, on patients with an extraordinarily deadly brain tumor called glioblastoma. Median survival for a patient found to have this cancer is 14 months.

So far, the drug, manufactured by Sugen, appears to slow or arrest tumor growth in about a third of glioblastoma patients, but it's too soon to say how long the benefits will last. Side effects appear to be mild. "We have one patient who's been on it for two years and three months," says Malkin. "His tumor is still there, but it's stable. He's alive; he's at work. For someone with recurrent glioblastoma, that's remarkable."

Malkin is quick to point out that one growth-factor inhibitor isn't going to cure cancer. Cancer is a complicated disease. Tumors usually are made up of different types of cells, expressing different genes, sensitive to different growth factors and therefore responding to different drugs. "When you are trying to kill cancer cells, you're always likely to need combination treatment," says Merck's Scolnick. Like AIDS treatments, the new generation of cancer drugs will need to be combined with older drugs and possibly with one another to be most effective.

If the promise of these drugs holds up, however, cancer treatment in the 21st century will bear little resemblance to today's chemotherapy. Drugs will be precisely tailored to the individual tumor, and the cancers themselves will be described not by the site they attack--breast cancers, lung cancers, etc.--but by the genes they express. The National Cancer Institute is at work creating a DNA library of tumor types, a long-range project called C-GAP (Cancer Genome Anatomy Project). But it will be years before this library can be put to practical use. "It took 20 years to make testing for hormone receptors routine in breast-cancer patients," notes UCLA's Slamon. It will take at least a decade to make testing for HER-2/neu, RAS and other genes routine for cancer patients in general.

It's also possible that the new generation of drugs emerging from the labs won't work very well, or that the much vaunted lack of major side effects will prove to be an illusion. All the enzymes and growth-factor receptors blocked by the new drugs play a role in normal cell division as well as in cancer. So disrupting them could cause harm. "Whether the therapy is going to be a major advance, a modest improvement or a disappointment is not clear," says Dr. J. Michael Bishop, molecular biologist at the University of California, San Francisco, who shared a 1989 Nobel Prize with Dr. Harold Varmus for their pioneering work on oncogenes. But Bishop is impressed that the field is moving so swiftly, and most researchers are convinced that they are at least on the right track. Says Joseph Schlessinger, a New York University scientist who helped develop SU101: "Early in the next millennium, we will significantly extend the life expectancy of cancer patients. I have no doubt about that."

Though patients long desperately for a "cure," extending life is the more realistic goal in treating cancer. The newer drugs, unlike chemotherapy agents, are "cancer stoppers," not "cancer killers," says Malkin. Chances are that they will have to be taken for many years, or even for the rest of a patient's life. But if such drugs can slow or stop the growth and spread of malignant cells, then cancer can be transformed from an acute and deadly disease into a chronic and manageable one. That doesn't make as sexy a headline as a cancer cure, but it's still the difference between life and death.

--With reporting by Lawrence Mondi/New York and James Willwerth/Los Angeles

With reporting by Lawrence Mondi/New York and James Willwerth/Los Angeles