CANCER IS NOT LIKE other diseases. Cancer works from within. Because of genetic flaws or irritants such as tobacco smoke, cells that were once normal start to veer off track, until they begin reproducing wildly, ultimately creating tumors. In other words, fighting cancer is like fighting depression or a civil war—it is basically a matter of fighting oneself. Chemotherapy, for example, floods the system with chemicals that kill cells as they divide, exploiting cancer’s habit of proliferating like crazy. But chemotherapy kills all dividing cells, even healthy ones, and the side effects can be devastating. For decades patients have had few alternatives. This year, however, a series of major breakthroughs suggests that a revolution in cancer care is taking place. One of the critical developments has been wrought by an unassuming Houston surgeon named Jack Roth, who has been correcting the genetic errors that cause tumors to flower in the first place. Roth, in essence, is making tumors disappear.

Tall, pale, and quietly intense, the 53-year-old Roth works in a tidy office at the University of Texas M. D. Anderson Cancer Center, where he is the head of thoracic and cardiovascular surgery. Although his groundbreaking experiments have brought him superstar status, he is shy in a hyperintellectual way. He has a calm expression and a slightly detached air, as if social contact with other human beings might be an alarming idea—often the hallmark of a good surgeon. Roth grew up in Indiana, majored in economics at Cornell, and was headed for a career in business or law when an uncle who was a neurosurgeon at Boston City Hospital invited him one summer to work as a scrub technician. His uncle would bark out the name of an instrument, and Roth would hand it over. From that point on, he wanted to work in a hospital. “I just found it very interesting,” he says. “It was much more exciting than economics, I can tell you that.” He thrived in the strange environment. “In medicine, you learn to be not only compassionate but also somewhat removed. Otherwise, if you are too emotionally involved, it can cloud the decision-making process.”

After attending medical school at Johns Hopkins, he obtained a research fellowship and completed residencies in general and thoracic surgery at UCLA and then went to the National Cancer Institute in Bethesda, Maryland. The bulk of his career has been an attempt to unravel the secret of how cancer ravages the body. “Well, it’s just the mystery of the process,” he says, explaining what intrigued him. “You could treat some patients, and they would survive, but you could treat others, and they would die. And nobody really understood why.” Roth had the good fortune to enter the field in the seventies, just as cancer research started to go through one of its seismic upheavals in understanding. During President Nixon’s war on cancer, hundreds of thousands of tax dollars subsidized the effort to puzzle out how cancer functions at the cellular level. For the first time, it became clear that flaws in certain genes could cause the disease.

Basically, cancer is a mutiny. It starts when healthy cells fail to replicate according to the body’s rules, instead swerving onto an idiosyncratic path. Many mutations are benign, and some are corrected by the body. Cancer results when a slew of deviations accumulates, making a cell unredeemable. Then it may enact a perverse exaggeration of the life cycle. Malignant cells start to proliferate wildly; more important, they refuse to die. In recent decades scientists have come to realize that certain genes determine whether cells become cancerous—one of these genes, p53, is called “the guardian of the genome,” because it acts like a cop, policing the nucleus of a cell to make sure that the chromosomes obey the rules. The most critical moment in a cell’s history occurs when chromosomes—the blueprints for life—split in two, creating an exact copy for a daughter cell. Mistakes in the chromosomes can cause dire consequences. For that reason p53 makes a cell self-destruct if anything goes wrong, preferring a suicide mission over the creation of a mutant cell. When p53 works properly, it prevents tumors from growing. But sometimes, as a precancerous cell starts to go bad, p53 itself becomes deranged. Then, like a city in the hands of a corrupt police department, things spiral out of control. Other genes play similar roles, but p53 is the gene most frequently mutated in human cancers. In fact, defective p53 may be involved in as many as half of all cancers.

Roth formulated the idea behind his current work in the early eighties after observing how cancer and malfunctioning p53 went hand-in-hand. Reinstate a working copy of p53, he speculated, and you could reverse the malignancy process. Get rid of corruption in the police force, and law and order will be restored. It was a good idea, but controversial and ahead of its time: The field of biotechnology was still in its infancy, and there was no consensus on how to safely get a new gene into a living cell.

Still pondering the approach, Roth moved to M. D. Anderson in 1986. One of the first three comprehensive cancer centers designated by the National Cancer Institute, M. D. Anderson was attracting thousands of patients from all over the world—each one looking for some break from the mounting chaos within. Roth found himself surrounded by the finest minds in his profession, all devoted for twelve or eighteen hours a day to the same Sisyphean challenge. Despite the time they spent trying to keep their patients alive, they still lost two out of every three.

In the early nineties molecular biologists began using viruses to perform gene therapy. Working with scientists at Introgen Therapeutics, an Austin biotech company he had co-founded, Roth developed a technique to replace the defective p53 gene, using the common cold virus as a vehicle to carry a healthy version of the gene into the malfunctioning cell. At M. D. Anderson, Roth then successfully replaced defective p53 in animals. “This not only corrected the p53 defect but it also caused the cancer cells to self-destruct,” he recounts. “I was astonished to see that. Cancer cells have all these different genetic deletions and defects, and here we were correcting just one and having this profound effect.” When he proposed trying the concept in people, however, he ignited a raging controversy: Health experts still feared that the viruses used to ferry the genes could inadvertently wreak havoc, and cancer specialists were skeptical that fixing one gene would stop a systemic disease.

After more than a year, Roth finally won the right to proceed. In 1994 he made medical history by inserting a long hypodermic needle into the chest of a furnituremaker. Roth had established that a tumor in the man’s lungs contained defective p53. Depressing the plunger of the needle, he injected the growth with a genetically altered virus. Ordinary viruses invade a cell and usurp its machinery, turning the cell into a factory that churns out more virus; Roth’s altered virus wormed its way into the nucleus of the cancer cells and reinstated a working copy of p53. After several injections, the man’s tumor started to shrink. “Here he was, the first patient we were treating, and it looked like the treatment was working,” says Roth.

The treatment may have worked locally, but it didn’t save the life of the furnituremaker. He died four months later from tumors that demolished his kidneys. (Because of federal restrictions governing the experiment, Roth was not allowed to treat other parts of the man’s body with gene therapy.) Eight other patients enrolled in that first experiment, and every one of them died—they were so sick that their fate was not a surprise. Still, the trial was considered a success: It proved that gene therapy was safe and that p53 was being successfully inserted into cancer cells.

In Roth’s next experiment, several patients, who also had terminal cancer, experienced seemingly miraculous turnarounds. One was 73-year-old Verna Clouse, whom I met during a checkup last year. Clouse has silver-blond hair, blue eyes, and a personality like a bear hug. In the examining room she whipped off her aqua-green track suit without a qualm. “I have five sisters,” she announced, “and I’m so old it doesn’t matter.” I smiled, imagining how she got along with the reserved Dr. Roth.

Four years earlier, a routine CAT scan had revealed a spot on Clouse’s lung. She had smoked for 52 years and was already wheezing from emphysema; radiation and chemotherapy had only slowed the cancer down. It seemed certain she would die—a doctor told her, “You know, there’s no cure for cancer.” One day Clouse was watching TV when she saw a brainy-looking guy talking about p53. It was Roth. Clouse enrolled in his clinical trial of the gene therapy’s effect on lung cancer. By her second visit to M. D. Anderson, her tumor was shrinking. Every month afterward, it shrank more.

“Here’s our star patient!” Roth called heartily as he entered the room. “Your x-rays look great! You still have that little abnormality, but it’s even less now. I think it’s probably just a scar from the radiation.”

Later I asked Roth if Clouse had been cured. “Well, we don’t use that word,” he said. “She certainly has done very well.” p53 is not a cure per se—it does not rid the body of cancer. Rather, when used in combination with radiation, chemotherapy, and surgery, it greatly enhances a doctor’s ability to control life-threatening tumors, at least for a while. Unfortunately, after more than a year’s reprieve, Clouse’s cancer recently metastasized and showed up in her lymph glands. She is fighting it just as gamely as she fought her lung cancer, but her prognosis is once again bleak.

Roth is soldiering on too. After the p53 therapy showed such promise with lung cancer, Introgen quickly applied for permission from the Food and Drug Administration to use it against other forms of the disease. In June Introgen announced that the results of a clinical trial involving head and neck cancers had been so compelling that the company was asking the FDA to put the p53 therapy on the fast track for approval. (The treatment successfully shrank some tumors by more than 50 percent and stopped the growth of others entirely, without any serious side effects; twice as many patients survived as Roth had expected.) If the FDA grants the request, Introgen is likely to become the first biotech company in the country to receive federal approval to market a gene therapy for cancer.

Altogether it has been a year of astounding developments in the field: Deaths from cancer started dropping for the first time ever, the incidence of new cases has been falling, and reports of another new, highly successful approach, involving anti-angiogenesis drugs (which cut off the blood supply to tumors), caused a flurry of excitement. Cancer comes in so many permutations that no single treatment is likely to prove a cure-all, but these emerging therapies clearly mark a watershed moment. All of a sudden experts are predicting that cancer may soon be a disease people can live with, like diabetes or AIDS. For years Jack Roth and the white-coated doctors at M. D. Anderson have fought a losing battle. Now, despite their inculcation against the blurring effects of sentiment and their allegiance to cold facts, many are beginning to believe they may actually defeat the most tenacious disease afflicting the human race.