When Bezwoda returned to the annual cancer meeting in Atlanta in May 1999, he was clearly triumphant. He rose to the podium confidently, feigning irritation that his name had been mispronounced during the introduction, and flashed his opening slides. As Bezwoda presented the data—his monotone voice washing over the vast sea of faces in front of him—a spell of silence fell over the audience. The wizard of Wits had worked magic again. At Witwatersrand hospital, young women with high-risk breast cancer treated with bone marrow transplants had showed staggeringly successful results. At eight and a half years, nearly 60 percent of patients in the megadose/transplant arm were still alive, versus only 20 percent in the control arm. For patients treated with the Bezwoda regimen, the line of survival had plateaued at about seven years with no further deaths, suggesting that many of the remaining patients were not just alive, but likely cured. Applause broke out among transplanters.
But Bezwoda’s triumph felt odd, for although the Witwatersrand results were unequivocally spectacular, three other trials presented that afternoon, including Peters’s, were either equivocal or negative. At Duke, embarrassingly enough, the trial had not even been finished because of low accrual. And while it was too soon to assess the survival benefits of transplantation, its darker face was readily evident: of the three-hundred-odd patients randomized to the transplant arm, thirty-one women had died of complications—of infections, blood clots, failed organs, and leukemias. The news from Philadelphia was even more grim. The megadose chemotherapy regimen had not produced a hint of benefit, not “even a modest improvement,” as the investigators glumly informed the audience. A complex and tangled trial from Sweden, with patients divided into groups and subgroups, was also headed inexorably toward failure with no obvious survival benefit in sight.
How, then, to reconcile these vastly disparate results? The president of the American Society of Clinical Oncology (ASCO) had asked a panel of discussants to try to hammer all the contradictory data into a single cohesive shape, but even the experts threw up their hands. “My goal here,” one discussant began, frankly befuddled, “is to critique the data just presented, to maintain some credibility in the field, and to continue to remain friends with both the presenters and the discussants.”
But even that would be a tall order. On and off the stage, the presenters and discussants bickered about small points, hurling critiques at each other’s trials. Nothing was resolved and certainly no friendships were made. “People who like to transplant will continue to transplant, and people who don’t will continue not to,” Larry Norton, the powerful breast oncologist and president of the National Alliance of Breast Cancer Organizations (NABCO), told a journalist from the New York Times. The conference had been a disaster. As the exhausted audience trickled out of the massive auditorium in Atlanta, it was already dark outside and the warm, muggy blast of air provided no relief.
Bezwoda left the Atlanta meeting in a hurry, leaving behind a field awash with confusion and tumult. He had underestimated the impact of his data, for it was now the sole fulcrum on which an entire theory of cancer therapy, not to mention a $4 billion industry, rested. Oncologists had come to Atlanta for clarity. They left exasperated and confused.
In December 1999, with the benefits of the regimen still uncertain and thousands of women clamoring for treatment, a team of American investigators wrote to Bezwoda at Witwatersrand to ask if they could travel to Johannesburg to examine the data from his trial in person. Bezwoda’s transplants were the only ones that had succeeded. Perhaps important lessons could be learned and brought back to America.
Bezwoda readily agreed. On the first day of the visit, when the investigators requested the records and logbooks of the 154 patients in his study, Bezwoda sent them only 58 files—all, oddly, from the treatment arm of the trial. When the team pressed for records from the control arm, Bezwoda claimed that they had been “lost.”
Mystified, the team probed further, and the picture began to turn disturbing. The records provided were remarkably shoddy: scratched-out, one-page notes with random scribbles written almost as an afterthought, summarizing six or eight months of supposed care. Criteria for eligibility for the trial were virtually always missing in the records. Bezwoda had claimed to have transplanted equal numbers of black and white women, yet nearly all the records belonged to poor, barely literate black women treated at the Hillbrow Hospital in Johannesburg. When the reviewers asked for consent forms for a procedure known to have deadly consequences, no such forms could be found. The hospital’s review boards, meant to safeguard such protocols, certainly had no copies. No one, it seemed, had approved the procedure or possessed even the barest knowledge of the trial. Many of the patients counted as “alive” had long been discharged to terminal-care facilities with advanced, fungating lesions of breast cancer, presumably to die, with no designated follow-up. One woman counted in the treatment arm had never been treated with any drugs. Another patient record, tracked back to its origin, belonged to a man—obviously not a patient with breast cancer.
The whole thing was a fraud, an invention, a sham. In late February 2000, with the trial unraveling and the noose of the investigation tightening around him every day, Werner Bezwoda wrote a terse typewritten letter to his colleagues at Witwatersrand admitting to having falsified parts of the study (he would later claim that he had altered his records to make the trial more “accessible” to American researchers). “I have committed a serious breach of scientific honesty and integrity,” he wrote. He then resigned from his university position and promptly stopped giving interviews, referring all questions to his attorney. His phone number was unlisted in Johannesburg. In 2008, when I tried to reach him for an interview, Werner Bezwoda was nowhere to be found.
The epic fall of Werner Bezwoda was a terminal blow to the ambitions of megadose chemotherapy. In the summer of 1999, a final trial was designed to examine whether STAMP might increase survival among women with breast cancer that had spread to multiple lymph nodes. Four years later, the answer was clear. There was no discernible benefit. Of the five hundred patients assigned to the high-dose group, nine died of transplantation-related complications. An additional nine developed highly aggressive, chemotherapy-resistant acute myeloid leukemias as a consequence of their treatments—cancers far worse than the cancers that they had begun with.
“By the late 1990s, the romance was already over,” Robert Mayer said. “The final trials were merely trials meant to hammer the nails into the coffin. We had suspected the result for nearly a decade.”
Maggie Keswick Jencks witnessed the end of the transplant era in 1995. Jencks, a landscape artist who lived in Scotland, created fantastical and desolate gardens—futuristic swirls of sticks, lakes, stones, and earth shored up against the disordered forces of nature. Diagnosed with breast cancer in 1988, she was treated with a lumpectomy and then a mastectomy. For several months, she considered herself cured. But five years later, just short of her fifty-second birthday, she relapsed with metastatic breast cancer in her liver, bones, and spine. At the Western General Hospital in Edinburgh, she was treated with high-dose chemotherapy followed with autologous transplant. Jencks did not know that the STAMP trial would eventually fail. “Dr. Bill Peters . . . had already treated several hundred patients with [transplantation],” she wrote, ever hopeful for a cure. “The average length of remission for his patients after treatment was eighteen months. It seemed like a lifetime.” But Jencks’s remission did not last a lifetime: in 1994, just short of her eighteenth month after transplantation, she relapsed again. She died in July 1995.
In an essay titled A View from the Front Line, Jencks described her experience with cancer as like being woken up midflight on a jumbo jet and then thrown out with a parachute into a foreign landscape without a map:
“There you are, the future patient, quietly progressing with other passengers toward a distant destination when, astonishingly (Why me?) a large hole opens in the floor next to you. People in white coats appear, help you into a
parachute and—no time to think—out you go.
“You descend. You hit the ground. . . . But where is the enemy? What is the enemy? What is it up to? . . . No road. No compass. No map. No training. Is there something you should know and don’t?
“The white coats are far, far away, strapping others into their parachutes. Occasionally they wave but, even if you ask them, they don’t know the answers. They are up there in the Jumbo, involved with parachutes, not map-making.”
The image captured the desolation and desperation of the era. Obsessed with radical and aggressive therapies, oncologists were devising newer and newer parachutes, but with no systematic maps of the quagmire to guide patients and doctors. The War on Cancer was “lost”—in both senses of the word.
Summer is a season of sequels, but no one, frankly, was looking forward to John Bailar’s. Sequestered away at the University of Chicago, Bailar had been smoldering quietly in his office since his first article—“Progress Against Cancer?”—had sent a deep gash through the NCI’s brow in May 1986. But eleven years had passed since the publication of that article, and Bailar, the nation’s reminder-in-chief on cancer, was expected to explode with an update any day. In May 1997, exactly eleven years after the publication of his first article, Bailar was back in the pages of the New England Journal of Medicine with another appraisal of the progress on cancer.
The punch line of Bailar’s article (coauthored with an epidemiologist named Heather Gornik) was evident in its title: “Cancer Undefeated.” “In 1986,” he began pointedly, “when one of us reported on trends in the incidence of cancer in the United States from 1950 through 1982, it was clear that some 40 years of cancer research, centered primarily on treatment, had failed to reverse a long, slow increase in mortality. Here we update that analysis through 1994. Our evaluation begins with 1970, both to provide some overlap with the previous article and because passage of the National Cancer Act of 1971 marked a critical increase in the magnitude and vigor of the nation’s efforts in cancer research.”
Little had changed in methodology from Bailar’s earlier analysis. As before, Bailar and Gornik began by “age-adjusting” the U.S. population, such that every year between 1970 and 1994 contained exactly the same distribution of ages (the method is described in more detail in earlier pages). Cancer mortality for each age bracket was also adjusted proportionally, in effect, creating a frozen, static population so that cancer mortality could be compared directly from one year to the next.
The pattern that emerged from this analysis was sobering. Between 1970 and 1994, cancer mortality had, if anything, increased slightly, about 6 percent, from 189 deaths per 100,000 to 201 deaths. Admittedly, the death rate had plateaued somewhat in the last ten years, but even so, this could hardly be construed as a victory. Cancer, Bailar concluded, was still reigning “undefeated.” Charted as a graph, the nation’s progress on cancer was a flat line; the War on Cancer had, thus far, yielded a stalemate.
But was the flat line of cancer mortality truly inanimate? Physics teaches us to discriminate a static equilibrium from a dynamic equilibrium; the product of two equal and opposite reactions can seem to sit perfectly still until the opposing forces are uncoupled. What if the flat line of cancer mortality represented a dynamic equilibrium of counterbalanced forces pushing and pulling against each other?
As Bailar and Gornik probed their own data further, they began to discern such forces counterpoised against each other with almost exquisite precision. When cancer mortality between 1970 and 1994 was split into two age groups, the counterbalancing of forces was immediately obvious: in men and women above fifty-five, cancer mortality had increased, while in men and women under fifty-five, cancer mortality had decreased by exactly the same proportion. (Part of the reason for this will become clear below.)
A similar dynamic equilibrium was apparent when cancer mortality was reassessed by the type of cancer involved. Mortality had decreased for some forms, plateaued for others, and increased for yet others, offsetting nearly every gain with an equal and opposite loss. Death rates from colon cancer, for instance, had fallen by nearly 30 percent, and from cervical and uterine cancer by 20. Both diseases could be detected by screening tests (colonoscopy for colon cancer, and Pap smears for cervical cancer) and at least part of the decrease in mortality was the likely consequence of earlier detection.
Death rates for most forms of children’s cancer had also declined since the 1970s, with declines continuing over the decade. So, too, had mortality from Hodgkin’s disease and testicular cancer. Although the net number of such cancers still represented a small fraction of the total cancer mortality, treatment had fundamentally altered the physiognomy of these diseases.
The most prominent countervailing ballast against these advances was lung cancer. Lung cancer was still the single biggest killer among cancers, responsible for nearly one-fourth of all cancer deaths. Overall mortality for lung cancer had increased between 1970 and 1994. But the distribution of deaths was markedly skewed. Death rates among men had peaked and dropped off by the mid-1980s. In contrast, lung cancer mortality had dramatically risen in women, particularly in older women, and it was still rising. Between 1970 and 1994, lung cancer deaths among women over the age of fifty-five had increased by 400 percent, more than the rise in the rates of breast and colon cancer combined. This exponential upswing in mortality had effaced nearly all gains in survival not just for lung cancer, but for all other types of cancer.
Alterations in the pattern of lung cancer mortality also partially explained the overall age skew of cancer mortality. The incidence of lung cancer was highest in those above fifty-five, and was lower in men and women below fifty-five, a consequence of changes in smoking behavior since the 1950s. The decrease in cancer mortality in younger men and women had been perfectly offset by the increase in cancer mortality in older men and women.
Taken in balance, “Cancer Undefeated” was an article whose title belied its message. The national stalemate on cancer was hardly a stalemate, but rather the product of a frantic game of death in progress. Bailar had set out to prove that the War on Cancer had reached terminal stagnancy. Instead, he had chronicled a dynamic, moving battle in midpitch against a dynamic, moving target.
So even Bailar—especially Bailar, the fiercest and most inventive critic of the war—could not deny the fierce inventiveness of this war. Pressed on public television, he begrudgingly conceded the point:
Interviewer: Why do you think they’re going down a little bit, or plateauing?
Bailar: We think they have gone down perhaps one percent. I would like to wait a little bit longer to see this downturn confirmed, but if it isn’t here yet, it’s coming. . . .
Interviewer: Dr. Bailar?
Bailar: I think we might agree that the cup is half-full.
No single strategy for prevention or cure had been a runaway success. But undeniably this “half-full cup” was the product of an astonishingly ingenious array of forces that had been deployed against cancer. The vaunted promises of the 1960s and 1970s and the struggles of the 1980s had given way to a more grounded realism in the 1990s—but this new reality had brought its own promises.
Sharply critiquing the defeatism of Bailar and Gornik’s assessment, Richard Klausner, the director of the NCI, pointed out:
“‘Cancer’ is, in truth, a variety of diseases. Viewing it as a single disease that will yield to a single approach is no more logical than viewing neuropsychiatric disease as a single entity that will respond to one strategy. It is unlikely that we will soon see a ‘magic bullet’ for the treatment of cancer. But it is just as unlikely that there will be a magic bullet of prevention or early detection that will knock out the full spectrum of cancers. . . . We are making progress. Although we also have a long way to go, it is facile to claim that the pace of favorable trends in mortality reflects poor policies or mistaken priorities.”
An era of oncology was coming to a close. Already, the field had turned away from its fiery adol
escence, its entrancement with universal solutions and radical cures, and was grappling with fundamental questions about cancer. What were the underlying principles that governed the root behavior of a particular form of cancer? What was common to all cancers, and what made breast cancer different from lung or prostate cancer? Might those common pathways, or differences for that matter, establish new road maps to cure and prevent cancer?
The quest to combat cancer thus turned inward, toward basic biology, toward fundamental mechanisms. To answer these questions, we must turn inward, too. We must, at last, return to the cancer cell.
PART FIVE
“A DISTORTED VERSION
OF OUR NORMAL SELVES”
It is in vain to speak of cures, or think of remedies, until such time as we have considered of the causes . . . cures must be imperfect, lame, and to no purpose, wherein the causes have not first been searched.
—Robert Burton,
The Anatomy of Melancholy, 1893
You can’t do experiments to see what causes cancer. It’s not an accessible problem and it’s not the sort of thing scientists can afford to do.
—I. Hermann,
cancer researcher, 1978
What can be the “why” of these happenings?
—Peyton Rous,
1966, on the mystery
of the origin of cancer
“A unitary cause”
It is the spring of 2005—a pivot point in the medical oncology fellowship. Our paths are about to divide. Three of us will continue in the clinic, with a primary focus in clinical research and in the day-to-day care of patients. Four will explore cancer in the laboratory, retaining just a minor presence in the clinic, seeing just a handful of patients every week.