IN NOVEMBER 1995, PAUL MADISON, a retired state tax examiner from Northport, Alabama, learned that the strange speech problems he had been having were the result of a brain tumor. Over the next several months, he began an exhausting schedule of radiation and chemotherapy. His wife, Alligene, who now has half a dozen spiral notebooks full of handwritten observations on his condition, accompanied him on every appointment. On April 12 the Madisons drove to the Kirkland Clinic at the University of Alabama in Birmingham to see whether the tumor’s rampant growth had been arrested. But Paul’s doctor gave them disheartening news: It was growing again. “It’s very aggressive,” he said. “It has changed quite a bit since the last scan.”

Before the appointment, the Madisons had read an article about suicide gene therapy, an experimental remedy that alters the genetic makeup of a tumor’s cells so that the tumor stops growing. In desperation the Madisons seized on the possibility that the therapy—which was being tested at several clinics around the country, including the University of Texas M. D. Anderson Cancer Center in Houston—might be Paul’s salvation. They didn’t have much time, because the kind of tumor that afflicted Paul is almost always fatal and grows very quickly, and his first symptoms had appeared eight months earlier. “Mother, pack your bags,” the Madisons’ daughter Denese said when she heard the test results from Birmingham. “We’re going to Houston in the morning.”

Paul, who is 68, began the treatment at M. D. Anderson in May, and since then, tests have shown no signs of growth in his tumor. The Madisons are convinced that he has been cured. They talk about gene therapy with the fervency of religious converts describing their faith—or rather, Alligene does, as Paul still has trouble speaking.

“She . . .” Paul begins, and then he gets stuck.

“Speaks for me?” Alligene asks.

He nods.

“The good Lord opened the doors and windows for us,” says Alligene. “We are just so grateful that the tumor is not growing anymore.”

Yet Alfred Yung, the neuro-oncologist overseeing the clinical trial at M. D. Anderson, is more measured in his assessment of the therapy’s possibilities, for he knows firsthand that it doesn’t always work. Of the three patients in his care who received the treatment, only Paul Madison has survived. And that, in a nutshell, is the state of gene therapy today: wildly promising but still far from reliable. While the public seems to expect doctors to be able to fix chronic medical problems by tinkering with genetic material like a mechanic under the hood of a car, years of research need to be done before gene therapy fulfills its potential. Then again, if it can someday be perfected, it will indeed resemble a miracle—it will be an alternative for people who cannot be treated by more-conventional means.

Like most other patients who enter a clinical trial, Paul arrived at M. D. Anderson after a protracted ordeal. In his case the first indications that something was wrong were subtle; he had always been articulate, but about a year and a half ago he began having trouble saying some words. “I remember he had trouble with the word ‘pretty,’” says Alligene. “He would say ‘pity’ instead. He couldn’t get the r’s out. He more or less quit saying the word ‘refrigerator.’ He would just point at the refrigerator instead.”

The Madisons’ family doctor ran a battery of tests, but none of them revealed the cause of Paul’s problem. Finally, the doctor conceded that he couldn’t make a diagnosis and referred Paul to a neurologist, who ordered a magnetic resonance imaging (MRI) scan of his brain. Before the results came back from the lab, however, Paul’s condition abruptly worsened. On September 24, 1995, at about eight in the evening, Alligene saw him go into the bathroom. She grew concerned when she heard the water run for a long time and a lot of splashing. “It sounded like a flood,” she says. Alligene discovered that Paul was washing his face over and over; for a brief period, he was unable to speak. “He could not say a single word. I could see in his face that he didn’t know what was happening to him.”

Alligene took Paul to the nearby Druid City Regional Medical Center, where the medical staff suggested that he had had a stroke. When the results of the MRI finally came back, Paul’s neurologist noticed a small mass in the left parietal lobe of his brain—an area that governs the ability to speak—but he assumed that it was scar tissue left over from a stroke that he decided Paul must have suffered in the past, causing his initial speech problems. Then, several days later, Paul experienced another episode similar to the one that had sent him to the hospital; his face started twitching, and he couldn’t speak. This time, doctors recognized the occurrence for what it was—a seizure—but still thought the underlying problem was a stroke.

It wasn’t until November, when Paul was referred to the Kirkland Clinic, that his illness was correctly diagnosed. Stroke specialist Thomas Head discovered that an unusual amount of oxygen was being absorbed from Paul’s blood in the vicinity of the mass that the doctors in Druid City had found. In other words, the mass was growing, so it couldn’t be scar tissue. Head sent Paul to see a second specialist, Stephen Rosenfeld, who confirmed the presence of a tumor. After another MRI scan showed signs of further growth, a neurosurgeon performed a biopsy, and the findings were grim: Paul had a glioblastoma, the most malignant type of brain tumor. Because of the tumor’s location, however, surgery wasn’t an option; and even if it had been, there was no guarantee that all of the cancer could have been removed—glio-blastomas are fiercely tenacious. So that December, Paul began radiation and chemotherapy. For thirty consecutive weekdays, he drove to Druid City for the treatments. In February the Madisons were elated to learn that an MRI scan showed the tumor to be a tiny bit smaller. “We want to stay aggressive,” Rosenfeld told them. “Let’s go through another round of chemotherapy.” Before the next set of treatments was over, however, Paul started having serious problems. “He could hardly walk, and he was dropping things,” says Alligene. “His speech got to where we almost couldn’t understand him. Finally I said, ‘I really think we’ve got to go back.’” That was when Rosenfeld discovered that the tumor had started growing again.

Three days later, the Madisons and their daughter showed up at M. D. Anderson and announced that Paul wanted to enter the clinical trial of suicide gene therapy. M. D. Anderson has a worldwide reputation for its expertise in conducting trials, and every year hundreds of patients arrive there seeking access to therapies that aren’t available anywhere else. In 1995 M. D. Anderson administered more than four hundred clinical trials involving nearly 5,300 patients. While some trials are concerned with therapies developed at M. D. Anderson, others are sponsored by research centers or drug companies that want to test new approaches they have developed. The suicide gene therapy trial is an example of the latter: It grew out of work done independently by scientists at the National Institutes of Health and Massachusetts General Hospital. Mass General sold its research to one of the pioneers of gene therapies, a Maryland-based biomedical company called Genetic Therapy, which in turn contacted clinics around the country and asked them to participate in a trial. Doctors at M. D. Anderson were interested because it was a chance to try out a therapy they are likely to use in the future and because it gave their patients an alternative weapon against cancer.

The first patient to receive the suicide gene therapy at M. D. Anderson was 52-year-old Chris Pope. In the spring of 1995 Pope—a Houston Chronicle distributor and an avid outdoorsman—began to feel unusually sluggish. Then he started losing his sense of equilibrium, leading him to speculate that he had an inner ear infection. That May, an ear specialist discovered that Pope could not quickly touch his own nose, or the doctor’s nose, on command. An MRI scan subsequently revealed a large tumor deep in the middle of his brain, and a biopsy revealed that it was a glioblastoma. Because it could not be reached through surgery, Pope immediately began radiation and chemotherapy. By early October, however, the tumor was continuing to grow, and he decided to try the suicide gene therapy. “Chris always had to be the best at everything, so we said, ‘Of course you’d be the first,’” says his wife, Carol. “We decided that even if it didn’t work, there was some meaning, because the doctors could learn more about gene therapy through him. We had no false hopes at all.”

Soon after, Pope was outfitted with a metal halo that guided a needle precisely to his tumor. Through it, Yung’s team injected a retrovirus with DNA from the herpes virus spliced into it. The idea was to infect the tumor’s cells as they replicated, integrating genetic material from the herpes virus into the DNA of the quickly reproducing tumor cells. Then, fifteen days later, Pope was to start intravenous infusions of ganciclovir, an antiviral drug used to treat herpes. The ganciclovir was supposed to mistake the tumor cells for herpes and kill them.

The neat logic of suicide gene therapy—turning the tumor against itself—initially entranced researchers, but like so many new approaches, it doesn’t always work like it is supposed to, and there have been unforeseen side effects. Several days after the surgery, Pope started to feel terribly nauseous and couldn’t eat. Yung was baffled by the reaction until he discovered that the virus had inflamed the membranes surrounding Pope’s brain, causing him to develop chemical meningitis. Once Pope began the infusions of ganciclovir, the nausea cleared up, but he did not survive the glioblastoma. An MRI performed shortly after the therapy was administered indicated that the tumor had continued to grow despite the treatment, and he died on January 13, 1996. Nevertheless, Carol Pope is convinced that it extended her husband’s life. “There is no doubt in my heart that it gave Chris two more months to live,” she says. “In that time, one of our sons got married, the Chronicle presented Chris with an award, and a friend in Atlanta we hadn’t been able to see was able to spend Christmas with us. I look at those two months now and I think how incredible a time it was.”

Curtis Miller, a sixty-year-old oilman from Austin, was the second patient to receive suicide gene therapy at M. D. Anderson. On March 2, 1995, Miller went into convulsions while riding in a car that his daughter Kim was driving. Kim, who is a nurse, first thought he was having a stroke, but she soon realized it was a grand mal seizure. On March 15, after an MRI scan detected a small tumor in Miller’s brain and a biopsy revealed that it was a glioblastoma, a neurosurgeon at Austin’s St. David’s Hospital tried to remove the growth. But as often happens with glioblastomas, the tumor grew back. Miller then went to M. D. Anderson for two rounds of chemotherapy and a second surgery; when they failed to defeat the tumor, he entered the suicide gene therapy trial. “We knew it was experimental, but it’s a hope that you grab onto,” says his wife, Diane. “We were very excited. We felt he had a better than fifty-fifty chance.”

But the gene therapy did not curtail the growth of Miller’s tumor. After an MRI scan revealed that the treatment hadn’t succeeded, Miller underwent surgery for the third time, and it failed too. He died this past August 5. “They don’t know why it didn’t work,” says Diane. “The theory is sound—it’s just so new. But we wanted to help. This kind of cancer is so devouring and so evil; if just one little thing we did helps one other person, it was worth it.”

Last May 2, Paul Madison became the third participant in M. D. Anderson’s gene therapy trial, and his family feared the worst. Several days after the virus was injected into his brain, while they were staying with Alligene’s sister in Houston, Alligene was unable to wake Paul up in the middle of the night to give him a dose of medication. “He couldn’t respond to me at all,” she recalls. “He was just gurgling a little bit—it sounded like he was strangling.” Alligene called an ambulance to rush Paul to M. D. Anderson, where Yung ordered an MRI. The scan showed a broken blood vessel between the skull and his brain; it was causing Paul’s brain to swell to a dangerous degree, and it had quickly sent him into a coma. Still, Yung thought the condition could be treated. “There’s quite a lot of fluid there,” he told Alligene, “but we can fix this.” Paul was rushed into surgery, where the fluid was removed, alleviating the pressure on his brain. The operation was successful, and Paul’s condition stabilized in time for him to begin the infusions of ganciclovir in mid-May.

On June 25 the Madisons returned to M. D. Anderson to see how Paul was responding. “I think it’s time for hugs,” Yung told them: Miraculously, an MRI scan showed no growth in Paul’s tumor. Later, when Paul came back for follow-up tests, the news remained good. Because the tumor has weakened the right side of his body, Paul must rely on a wheelchair or a walker to get around, and he still has a great deal of trouble speaking. The brain doesn’t repair itself, so he will not regain the use of the part that was damaged; however, he is working with a speech therapist to learn to speak using other parts of his brain. These are complications he’s willing to accept, for unlike everyone else who came to M. D. Anderson for the suicide gene therapy, Paul is still alive. In October, when he last saw Yung, he got out of his wheelchair to shake his hand. When Yung told him that the latest MRI scan again showed no signs of growth, he was so relieved that he started to cry.

The Madisons want everybody to know about their experience so that others can benefit from gene therapy. “This needs to be told,” says Alligene. “We hope to enlighten other people.” But of course, it’s important to remember that the therapy appears to have helped only one out of three patients at M. D. Anderson. “This is our frustration—not all tumors react the same way,” Yung says. While he considers a survival rate of 33 percent surprisingly good for glioblastoma, he does not yet share Alligene’s unabashed enthusiasm for the new treatment. “The important thing with gene therapy is not to oversell it,” he says. “Patients think it is going to be their salvation, and when they don’t get the results they are waiting for, they are very disappointed. There are a lot of problems we need to solve before we can call this a good or an excellent treatment.”