If the government’s role in reducing AIDS mortality represents a great victory for the idea that directed research campaigns can substantially affect the history of an epidemic, the war on cancer is an entirely different matter. For more than half a century, all the weapons of directed medical research—extramural grants to academic researchers, intramural programs of research and drug development, public-private partnerships, and outright support of industry—have been deployed against cancer with far less satisfactory results.
It has been more than three decades since President Nixon signed the National Cancer Act. The government over those years spent more than $50 billion combating what one social historian called “the dread disease.” The National Cancer Institute (NCI), the largest arm of the National Institutes of Health (NIH), created and financed a nationwide academic research establishment dedicated to understanding the biochemistry of cancer. It also devoted a sizable fraction of its resources to a wide-ranging hunt for drugs.
Those efforts transformed the commercial landscape. At the campaign’s outset, the pharmaceutical industry for the most part ignored cancer. But by the early twenty-first century there were dozens of firms pursuing more than four hundred potential cancer drug candidates. Why? NCI-funded basic science had created a multitude of potential drug targets. And the chemotherapies that emerged from the government’s half-century hunt for drugs, licensed at no cost to industry, created a market where none existed before.
But selling drugs is not the same as curing disease. Any objective survey of the most recent statistics would show that cancer as a public health problem has not changed much over the years. Cancer—a catchall term for a class of more than 110 separate malignancies—claimed the lives of more than 555,000 Americans in 2002, leaving it the nation’s number-two killer, a position it has held since the early 1930s.
There were tentative signs of progress in the 1990s in the long-running campaign against the disease. The rate at which cancer killed fell gradually after decades of increases and by 2000 was 11 percent below its peak. However, the biggest factor behind the decline had nothing to do with medical intervention. The falling cancer mortality rate was largely driven by the public health campaign that convinced many people to stop smoking in the 1970s and 1980s, which led to a falling incidence of lung cancer in the 1990s. There were also small reductions in the death rates for colon, breast, and prostate cancers, which have been attributed to improved treatments and early screening techniques like mammograms, although that assertion has been challenged.1 But those improvements have been modest at best. After it had been adjusted for the age at which people contract cancer, the cancer death rate stood at 203 per 100,000 persons in 1999, slightly worse than 199 per 100,000 persons in 1975, which in turn was slightly worse than the 195 cancer deaths per 100,000 persons recorded in 1950.
The chances of surviving cancer are certainly better than they were a generation ago. Six in ten cancer patients survived at least five years after a diagnosis of cancer by the late 1990s, compared to a shade less than five in ten in the mid-1970s. But those gains were offset by cancer incidence rates that were still rising, especially among the elderly. As people lived longer by surviving infectious diseases, heart attacks, strokes, and the complications of diabetes, they increased their risk of contracting and eventually dying of cancer.
Cancer incidence among the middle-aged went up compared to three decades earlier (with slight declines in the past decade in some cancers). But it was among these younger cancer patients that oncologists registered their greatest treatment gains. Every age group under sixty-five years posted declines in cancer mortality rates between 1973 and 1999, with the biggest gains among the very young, as clinicians deployed improved treatments for childhood cancers like leukemia and some solid tumors.
The population that contracts cancer has also shifted. In the nineteenth and early twentieth centuries, cancer was considered a woman’s disease. Cancers of the breasts and uterus were observable, while many cancers of internal organs, whether they killed women or men, went undiagnosed. As late as 1950, the cancer mortality rate was higher among women than men. But as tobacco began taking its terrible toll on the American people, the demographics of cancer changed dramatically. Lung cancer rates among men soared in the 1950s and 1960s as the direct result of the marketing and glamorization of cigarette smoking that had begun three decades earlier. By midcentury, women’s smoking rates began to catch up to men’s with predictable results. Virginia Slim, meet the Marlboro Man. During the 1980s and 1990s, the rates at which women contracted lung cancer rose sharply. With lung cancer accounting for nearly one in three cancer deaths, the cancer gender gap began to shrink.2
One would think these unyielding statistics and the grim prognosis for nearly half the people diagnosed with cancer would dampen enthusiasm for cancer war stories in the popular press. But in their frequent updates from the frontlines of research, editors and reporters remained invariably upbeat, with their enthusiastic embrace of the latest laboratory discoveries often bordering on the deliberately misleading. A New York Times story splashed across the top of the paper’s Sunday front page in 1998 was typical of the genre. Gina Kolata, an experienced medical writer, set off a media frenzy when she published an article proclaiming, “Hope in the Lab: A Cautious Awe Greets Drugs That Eradicate Tumors in Mice.” Judah Folkman, an NIH-funded researcher at Children’s Hospital in Boston, had come up with experimental drugs that cut off blood flow to rodent tumors. No less an authority than James Watson, codiscoverer of the DNA double helix, anointed Folkman the next Charles Darwin. Richard Klausner, the head of NCI, called his work “the single most exciting thing on the horizon” and the data “very compelling,” although he did mention “the big ifs” when talking about a drug that had never been tried in humans. The stock of EntreMed, the company Folkman worked with, soared. Book contracts got signed. (A few days after the story broke and controversy about its overhyped nature emerged, editors at the New York Times advised Kolata to reject her multimillion-dollar offer.) Over the next few years, the drugs quietly disappeared from view as the company began the long slog toward proving their usefulness against several rare forms of cancer.3
Cancer therapy for most patients had changed over the decades because the government had poured enormous resources into understanding the basic biology of the disease and developing drugs for treating it. But most oncologists still prescribed some combination of surgery, radiation, and chemotherapy—“slash, burn, and poison,” in one deflating description.4 In recent years, the surgery had become less invasive, the radiation better targeted, and the newer chemotherapy agents slightly more effective. The new drugs also had fewer side effects, or other drugs were developed to offset their toxicities. But in the end, patients on chemotherapy were still left praying that the cancer hadn’t metastasized to some hidden corner of the body, which, alas, for many it had. The cure for nearly half of cancer patients turned out to be nothing more than a temporary respite from the ravages of the disease, therapy’s success measured by a few, often miserable, months or years of life.
To the extent that drug treatment options have improved, the government played the central role in bringing them about. The Food and Drug Administration (FDA) approved fifty-eight cancer drugs between 1955, when NCI launched its cancer drug development program, and 2001. NCI played the lead role in the development of fifty of those cases. It found the molecule, assisted in late-stage preclinical studies, or directly sponsored the clinical trials. In some cases, it did all three. Even in the post-1980 era, when the number of private firms involved in cancer therapy research expanded rapidly, more than 70 percent of all new cancer drugs approved by the FDA had at least some government involvement.5
Is it fair to say then, as many observers have, that the government’s war on cancer is a failure? Does its performance call into question all public sector attempts to solve pressing medical problems through targeted programs of applied research? Jerome Groopman, the respected AIDS oncologist and medical journalist, concluded the government would have been better off focusing exclusively on basic research during its long war on cancer rather than supporting research into new drugs. In a provocative essay published in New Yorker magazine on the thirtieth anniversary of the Cancer Act, the Harvard Medical School professor noted the recent round of ebullient press accounts praising the latest targeted cancer drugs. Groopman allowed that “the current atmosphere of hope is not without foundation.” But he went on to say that “it is not without precedent, either. Ever since 1971, when President Nixon declared war on cancer, oncologists and cancer patients have been caught in a cycle of euphoria and despair as the prospect of new treatments has given way to their sober realities. The war on cancer turned out to be profoundly misconceived—both in its rhetoric and in its execution.”6
A closer examination of the history and accomplishments of the war on cancer would suggest that a blanket condemnation of the government’s directed research program is far too harsh. The government never ignored basic scientific inquiry into the biochemical and biomolecular processes behind cancer. Far from it. From the cancer war’s outset (which actually dates back to the late 1930s), the leadership of academic medicine and nonprofit research institutes insisted that autonomous, peer-reviewed researchers and their proposals occupy a central role in the campaign. At the end of the century, basic research still consumed more than three-quarters of NCI’s resources. Indeed, as Groopman admitted, the most recent drugs emerging from industry’s labs owed their existence to the understanding of the basic biology of the disease derived from the previous thirty years of government-funded basic research.
However, the public, speaking through its representatives in Congress, would have never supported that vast expansion of basic science unless it was coupled with the simultaneous search for cures. In the late 1960s Mary Lasker, a key member of the American Cancer Society’s board and a longtime booster of NIH, recruited elite financiers to her Citizens Committee for the Conquest of Cancer with the promise that increased science funding would be linked to a “moon shot” to cure cancer. Liberal Howard Metzenbaum, a retired parking-lot magnate from Cleveland who would soon become a senator, served as Lasker’s cochair. He got his start in politics by championing the targeted research approach. Senator Edward Kennedy, new to the Senate and heir to his family’s liberal legacy, introduced the bill in the upper chamber.
The pharmaceutical industry also jumped on the applied research bandwagon in the person of Elmer H. Bobst, the retired president of U.S. operations for Hoffmann–La Roche. Bobst, a longtime member of the American Cancer Society board, used his close friendship with President Nixon to push for an autonomous NCI and “subject it to a carefully considered program of directed research in the most promising areas.”7 They got what they wanted, but not as a substitute for basic scientific inquiry into the biology of the disease. The war on cancer vastly expanded both realms.
If there was any misconception at the outset of Nixon’s war on cancer, it was in the academic- and institute-based scientists’ narrow definition of basic research. They focused resources almost exclusively on microbiology, biochemistry, and genetics, the fields they knew best. Great academic empires and eventually even large fortunes were built on the foundations laid by NCI-funded research. Unfortunately, these same scientists, whose administrators controlled the panels that oversaw grants, gave much shorter shrift to in-depth investigations into the environmental and social roots of the disease. Nor did they emphasize public health prevention strategies for curbing its incidence since to do so would have brought them into conflict with some of the same powerful economic interests whose representatives were championing a strictly clinical approach to combating the disease.
The cancer crusade did not begin with President Nixon’s proclamation in 1971. By the late 1920s, rising cancer rates had prompted Senator Matthew Neely of West Virginia to seek a special appropriation of one hundred thousand dollars to study the disease. Speaking on the floor of the Republican-controlled Senate, the Democrat Neely called cancer “humanity’s greatest scourge,” a “monster that is more insatiable than the guillotine,” and a force “more destructive to life and health than the mightiest army.” He warned his colleagues that “if the rapid increase in cancer fatalities should persist in the future, the cancer curse would in a few centuries depopulate the earth.”8
Neely’s overwrought rhetoric set off a debate whose parameters echo to this day. One senator wondered if the increasing cancer rates had more to do with better diagnostics than a rising tide of tumors. Others spoke for a large portion of the public who preferred to put their faith in religion, quacks, and patent medicines, which, as cancer historian James Patterson has noted, were part of “a very diffuse but stubbornly persistent cancer counterculture, one of the many constants in the modern social history of cancer in the United States.” Neely’s rhetorical flourishes won support from his colleagues in the Senate, but the legislation died in the House.9
By 1937, the nation’s mood had shifted. The Depression and Franklin Delano Roosevelt’s New Deal awakened Americans to the possibilities of government solving its most pressing economic and social problems, including the fight against disease. Surgeon General Thomas Parran, whose wife died of cancer in 1929, spearheaded a coalition that made curing cancer a central thrust of the U.S. Public Health Service’s new National Institute of Health.
The coalition was a classic example of the divergent interests that sometimes come together to grow a new public policy in the Washington petri dish. It was led by the American Society for the Control of Cancer (ASCC, later the American Cancer Society) and its Women’s Field Army, whose door-to-door fundraising and early detection public awareness campaigns had become a fixture in many middle-class communities. Senator Homer Bone, a Democrat from the state of Washington, championed their cause in the belief that cancer especially hurt the poor, who couldn’t afford the doctor visits that might enable them to catch cancer in its early stages, while it was still treatable. They were joined by a growing research community, led by a frustrated physician from San Antonio named Dudley Jackson. His anger grew out of the fact his cancer-related research proposals had been turned down by the fledgling NIH.
This disparate coalition was joined by a new and rising cancer establishment in which the ASCC played a leading but far from exclusive role. Shortly after the turn of the century, John D. Rockefeller had endowed the Rockefeller Institute for Medical Research in New York City. By the mid-1930s, its prestigious faculty had turned to cancer research as one of its major activities. In 1936, the oil tycoon gave another $4 million to build the world’s most modern facility dedicated to cancer treatment—Memorial Hospital. In the postwar years, the hospital would merge with the Sloan-Kettering Institute, which was endowed by General Motors industrialists Alfred P. Sloan and Charles F. Kettering. Also in 1936, the wife of Starling Childs, who made his millions in New York utilities, died of cancer. Childs gave $10 million to the Yale University to study the disease—at the time the largest ever bequest for medical research. These endorsements of cancer research by leading philanthropists and one of the nation’s top medical schools helped the field emerge from its quack-ridden roots and emboldened politicians to authorize the government to play a bigger role in combating the disease.10
By the late 1930s, Congress was well along the path toward giving government a role in medical research. In 1930, responding to complaints that the private sector’s investment in medical science was far from adequate, it created NIH from the Public Health Service’s Hygienic Laboratory, which had been founded in the 1880s to study infectious diseases. But in expanding the agency, Congress limited it to basic research. The original authorizing legislation established its purpose as the “study, investigation, and research in the fundamental problems of the diseases of man.”11
The new NIH was swamped with proposals from outside researchers. Dudley Jackson, who would play the key role in getting the NCI legislation passed in the House, was a surgeon who saw how the disease ravaged his mostly poor clientele. Jackson asked NIH for money to study carbohydrate metabolism in dogs in order to gain insights into the fundamental causes of cancer. The agency repeatedly turned down his requests. If their goal was to drum up hinterland support for more funding for the agency, they picked the right man to ignore. Angered by what he saw as the tunnel vision of bureaucrats in Washington, Jackson convinced his powerful cousin, Rep. Maury Maverick of Texas, to sponsor a bill creating a separate cancer institute. He then traveled to the Capitol to drum up support.
Creation of the new agency was never in doubt. But the debate over the legislation sparked heated arguments about who should control the institute and what it should fund. Surgeon General Parran was convinced that a powerful central authority should play the lead role in tackling the disease. “Whatever path we take,” he said, “inevitably will conform to the governmental framework.” But Peyton Rous, the Rockefeller Institute researcher who had created a stir in 1911 when he discovered a virus that caused tumors in chickens (he would champion the mistaken viral transmission theory of the disease until his death in 1970), expressed the concerns of many scientists who feared government control of science. It would lead to “regimented direction,” he wrote to a friend. The American Medical Association also warned against government control of medical research but did not actively oppose the bill, preferring to save its political capital for what it considered to be a more important battle—the fight against national health insurance.12
The 1937 law passed easily. It gave the new institute broad authority to fund “researches, investigations, experiments, and studies relating to the cause, prevention, and methods of diagnosis and treatment of cancer.”13 In short, its job was to cure cancer, not just study it. But in its early years, virtually all of its limited funding went to basic research. While the blue ribbon advisory panel set up under the law was dominated by leading anticancer figures of the day, it included prominent academicians such as Harvard University president James Conant, for whom the autonomy of science was sacrosanct.
The panel left treatment to the doctors and public education about early detection to the ASCC. It rejected an emphasis on prevention. Instead, the new NCI set up a small in-house research staff to focus on basic biology and established a peer-review system for screening and approving outside grant proposals. The system would become the model for all of NIH when dozens of new institutes were added to its roster in the two decades after World War II.
The advisory panel was far less democratic than Dudley Jackson would have liked. It concentrated the majority of its grants in the leading research institutions, whose heads took turns sitting on the board. Jackson dealt with his new problem by lobbying the Texas legislature to create what is now the M. D. Anderson Cancer Center at the University of Texas in Houston, today the second largest cancer research institution in the United States.14 “The scientists, taking firm command of the anti-cancer alliance, then implemented the law in ways that advanced their own laboratory research and institutional interests and that mainly dismissed preventive approaches,” wrote Patterson. The cancer lobby was born.
The battle over the program’s direction was renewed in the immediate postwar years. Despite fealty by men at the top of NCI to Vannevar Bush’s vision of concentrating on basic research, powerful forces gathering outside the agency began demanding a more practical approach. Early detection and surgery had been the primary method of treating cancer since the late nineteenth century. Radiation took its place alongside surgery in the early 1900s in Europe and won a wide following among American physicians after Nobel Prize winner Marie Curie toured the United States to tout the process in 1921. In those early years of cancer treatment, people who peddled medicinal cures were considered quacks.15
The image began to change during World War II in the wake of a tragic wartime incident. In December 1943, German bombers launched a surprise attack on Allied shipping in Bari Harbor on the Italian coast. One vessel that sank, the USS John Harvey, was carrying mustard gas, a weapon banned after World War I. President Roosevelt had ordered it stockpiled just in case Hitler decided to use it. The contents leaked into the harbor, mixed with the burning oil, and vaporized. Hundreds of sailors, fishermen, and other local civilians died, their skin covered with welts and their lungs filled with blood. Autopsies later showed that the victims’ white blood cells had been wiped out by the gas.
Several physicians in the army’s Chemical Warfare Division, headed by Memorial Hospital chief Cornelius “Dusty” Rhoads, speculated that minute amounts of mustard derivatives might prove useful in treating leukemia and other fast-growing cancers of the blood. The government’s Wartime Office of Medical Research began testing the proposition at Yale, the University of Chicago, and other top-secret labs around the country. Rhoads continued that research when he returned to his post at the merged Memorial Sloan-Kettering Hospital after the war.16
Rhoads was driven by two central beliefs. First, he saw cancer as primarily a disease of cells. Instead of surgery or radiation, which eliminated the tumor mass, he wanted to focus on finding chemotherapy agents that would stop cancer cells from dividing. His second basic principle was based on advice from the industrialists who sat on his board. They wanted to model Sloan-Kettering after the leading industrial labs of the day, like Bell Labs, which married efficiency and high throughput to scientific inquiry.17 Rhoads implemented their vision. Between 1946 and 1950, scientists at Memorial Sloan-Kettering synthesized and tested fifteen hundred derivatives of nitrogen mustard gas for their cancer-fighting properties. In 1949, the FDA made mechlorethamine (Mustargen) the first government-approved cancer chemotherapy agent (in those days, one only had to prove it was safe, not effective, to get FDA approval).
Though the treatment was hardly curative, Time hailed the discovery. The magazine featured a confident, crew-cut Rhoads on its cover clad in a white lab coat with the symbol of the American Cancer Society—a sword smashing through a crab—in the background. “Some authorities think that we cannot solve the cancer problem until we have made a great, basic, unexpected discovery, perhaps in some apparently unrelated field. I disagree,” Rhoads was quoted as saying. “I think we know enough to go ahead now and make a frontal attack with all our forces.”18
He wasn’t the only clinician-scientist pushing NCI to pay closer attention to the treatment mandate in its authorizing legislation. Shortly after the war ended, Sidney Farber, the scientific director of the Children’s Cancer Research Foundation in Boston and a member of the faculty at Harvard Medical School, began experimenting with chemical blockers of folic acid, which is needed for DNA replication. His target was acute leukemia in children. A prodigious fundraiser with good connections to many leading politicians, Farber scored the first major success of the chemotherapy era when he came up with the antimetabolite compound methotrexate, which produced remissions in some leukemia patients. Farber began pushing for government funds to expand his experiments. Over the next two decades Farber would serve as the chief advocate for more money to develop new cancer chemotherapies.
Responding to pressure from key opinion leaders like Rhoads and Farber, NCI in 1955 launched the Cancer Chemotherapy National Service Center, a formal effort to develop new drugs. Many top scientists at the agency, including its director Kenneth Endicott, were appalled. “I thought it was inopportune, that we really didn’t have the necessary information to engineer a program, that it was premature, and well, it just had no intellectual appeal to me whatever,” he said a decade later.19 But researcher-activists like Farber ran roughshod over those objections. Within two years, nearly half the agency’s budget was going toward drug development.
By 1970, the year before the government officially declared war on cancer, the NCI drug development program had already screened more than four hundred thousand chemicals for their cancer-fighting potential. Some were obtained from academic investigators, others from industry, and still others were manufactured in-house. NCI scoured the world for exotic microbes, plants, and marine organisms to test in its labs. Taxol, a derivative of the Pacific Yew tree, whose chemical equivalent remains one of the more successful chemotherapy agents on the market to this day, was a product of those efforts.
The agency took on all the functions of a private drug company. It designed screening assays for testing compounds for their anticancer potential. It learned how to perform preclinical tests for determining toxicity and proper dosing. It recruited a nationwide network of academic oncologists capable of conducting clinical trials. It hired contractors that could mass produce the handful of drugs that slowed the progression of the disease in early clinical testing. “In many respects, the program operates in a fashion similar to that of a pharmaceutical company,” one NCI staffer noted in the mid-1980s, on the thirtieth anniversary of the drug development program. There was one big difference, however. “There is no consideration of profit, although cost-benefit considerations obviously play an important role in the choice of drugs to pursue or problems to investigate.”20
The other half of the budget went to basic scientists, who grouped themselves around two competing theories of cancer causation, although there was no grand design fashioning their academic proposals. The minority school followed in the footsteps of Percivall Potts, the eighteenth-century British physician and epidemiologist who had first noted the high incidence of testicular cancer among British chimney sweeps. The young Brits failed to bathe at the end of their daily labors and as a result developed cancer in the most sensitive area, where the soot and tar came to rest. Relying on similar epidemiological studies, the environmental school of thought believed cancer was caused by exposure to toxic substances. Limit those exposures, it argued, and you limit cancer.
Inside NCI, its cause was championed by Wilhelm C. Hueper, who ran the agency’s Environmental Cancer Section from its founding in 1948 to his retirement in 1964. Hueper, a German immigrant, was fired from his job as an industrial hygienist by DuPont de Niemours and Company in 1937 after concluding that the high incidence of bladder cancer among workers in its dye factories was caused by exposure to chemicals used in the manufacturing process. Over the next five years, while working for a small chemical company, he wrote an eight-hundred-page textbook entitled Occupational Tumors and Allied Diseases. After being turned down by Yale University Press, he raised three thousand dollars to self-publish the tome. His persistence was rewarded in 1962 when Rachel Carson drew heavily from his obscure work to produce her classic Silent Spring. Carson’s book inspired the movements that led to passage of the Clean Air and Clean Water acts and the creation of the Environmental Protection Agency and the Occupational Safety and Health Administration.21
The chemical carcinogenesis crowd’s cause was bolstered by the growing body of research connecting cigarette smoking to lung cancer. During the 1940s, lung cancer rates grew at five times the rate of other forms of cancer. To many hypothesizers, the obvious suspects behind the plague were cigarette smoke and the increasingly visible air pollution of major cities. The former was easier to document. On May 27, 1950, the Journal of the American Medical Association published the first articles tying tobacco to elevated incidence of the disease.
One article was written by Ernst L. Wynder, a German Jew who had fled Nazi Germany with his parents. During the war years, he studied medicine at Washington University Medical School in St. Louis and did summer hospital internships at New York University, where he had studied as an undergraduate. One wartime summer, he began questioning the widows of lung cancer victims, virtually all of whom told him their husbands had been heavy smokers. When he returned to St. Louis, he asked his anatomy professor, the well-regarded thoracic surgeon Evarts Graham, if he could systematically question lung cancer patients who entered his clinic for surgery. Graham was a heavy smoker and extremely skeptical about Wynder’s hypothesis. But his scientific curiosity prevailed and he gave Wynder permission to travel the country to scour medical records and conduct interviews with lung cancer patients or their survivors. The article that eventually appeared under both of their signatures, “Tobacco Smoking as a Possible Etiologic Factor in Bronchiogenic Carcinoma: A Study of 684 Proved Cases,” was the first epidemiological evidence tying heavy smoking to lung cancer. Four months later, a similar article by Richard Doll, a leading British epidemiologist, and his colleague, A. Bradford Hill, appeared in the British Medical Journal. A follow-up study by Wynder in 1953 convinced the editors of the Journal of the American Medical Association to stop taking cigarette ads. The war on smoking had begun.22
Over the next decade, the tobacco industry fought a dissembling campaign against the mounting evidence. Its chief spokesman was renegade researcher Clarence Cook “Pete” Little, who had once headed the American Society for the Control of Cancer. A direct descendant of Paul Revere, Little was a Harvard-trained biologist who harbored huge ambitions in both science and academic administration. In 1922, at age thirty-four, he became the youngest president of the University of Maine and three years later repeated the trick at the University of Michigan. He was fired from that post, though, after being accused of conducting an adulterous affair. But by 1929 he had already raised enough money from a wealthy auto executive to open a genetics research institute in Bar Harbor, Maine, which he would run for the next twenty-five years.
Shortly after moving back to his native state to run the new institute, he was asked by the physicians who ran the ASCC to head up its shoestring operation. He commuted two days a week to its New York offices to spread its main message that early detection was the best hope for beating cancer. (Celsus, Roman physician in the first century A.D., was the first to offer this enduring insight.) The organization stagnated under his command, and in 1944, Mary Lasker, whose husband Albert, an advertising man, had coined the slogan “Lucky Strikes Means Fine Tobacco,” engineered his ouster. A highly refined socialite whose close friends included many of the leading industrialists and financiers of New York City, she was upset, according to at least one account, by an attitude that she felt was both patronizing and predatory toward women.23
Little, a “self conscious big shot” without a wider stage, returned to Bar Harbor. In 1954, with retirement looming and his financial prospects dim, he accepted a position as scientific director of the newly established Tobacco Industry Research Committee, which had been set up by the tobacco industry to question reports linking smoking to cancer. For the next two decades he would be the industry’s main spokesman debunking the smoking-cancer connection. He consistently raised questions about the data behind the studies that showed they were linked. His efforts became the model for other industry trade groups that wanted to counter epidemiological studies showing the carcinogenic properties of their products.24
The Surgeon General’s report of 1964, warning that cigarette smoking was hazardous to health, coupled with the growing movement against industrial pollution, seemed to vindicate the epidemiologists’ approach to cancer. But to the growing ranks of geneticists, cell biologists, molecular biologists, and virologists studying the disease, the epidemiologist road led nowhere. How did those chemicals turn the body’s cells into out-of-control mutants? How did one account for the majority of cancers that seemingly had no environmental cause? One could spend an entire career giving mice cancer by exposing them to carcinogens. How would this knowledge help cure someone who already had the disease? “By the 1960s, the wider scientific community . . . became dismissive, viewing the research on chemical carcinogenesis as an intellectually bankrupt enterprise—nothing more than a mountain of facts with few good ideas propelling it forward,” Robert A. Weinberg, one of the nation’s leading research scientists, wrote in his chronicle of the scientific history of the war on cancer.25
Virologists, cellular biologists, and geneticists rushed in to fill the void. The viral theory of cancer, first enunciated by Peyton Rous in 1911, had virtually disappeared from scientific discussion by the 1930s. But in the late 1950s, it was resuscitated by Sarah Stewart and Bernice Eddy, longtime researchers at the NIH labs in Bethesda. They discovered a virus that induced tumors when injected in mice. They dubbed it the polyoma because the cancers included solid tumors as well as tumors of the blood. The discovery created a stir among the growing number of geneticists interested in the cancer problem, including genomics pioneer James Watson, who referred to the NIH scientists as “two old ladies” in a talk he gave at his Long Island laboratories to spread the news to a wider audience. Viruses were the smallest known organisms and had very short genetic sequences, Watson noted. If they caused cancer, the number of genes in the host cell that were affected by the virus also must be quite small and were therefore identifiable.26
Other scientists began replicating and expanding the Stewart-Eddy work. In 1962 virologists in Houston found that the human adenovirus, which causes common respiratory infections, induced tumors in hamsters. Two years later, British virologists discovered virus particles in cells of lymphoma patients in Africa. The cancer-virus connection also began gaining treatment adherents. In the late 1950s, scientists working in France had identified an antiviral protein given off by cells when they were attacked by viruses. They dubbed it interferon. It would be nearly two decades before interferon could be isolated and produced through biotechnology means. But in those early days, a number of researchers began to believe that if viruses caused cancer, then interferon might prevent or cure it through its antiviral activity.27
A 1964 New York Times Magazine article pulled together the straws in the wind. “A new stage in the struggle against cancer cannot be more than months or at most a year or two away,” it gushed. The attention prompted Congress to appropriate $10 million in 1965 for a Special Virus Cancer Program (SVCP). The government hunt for the viral causes of cancer was on. By the end of the decade, it had grown to a $30-million-a-year program.28
But how did these cancer viruses do their dirty work? Conventional viruses like the polio virus killed cells by entering them and destroying their reproductive machinery. But cancer cells weren’t dying. They were living long, hostile lives and proliferating wildly. Howard Temin of the University of Wisconsin and David Baltimore at the Massachusetts Institute of Technology solved the mystery. Their intellectual spadework prepared the ground for a rapid expansion of the SVCP once the 1971 National Cancer Act passed.
The Watson-Crick DNA double-helix model, sometimes referred to as the central dogma of cell biology, postulated that cell reproduction always proceeded through DNA replication and division. Most viral researchers therefore assumed that viruses like polio and Rous sarcoma, which were made up of RNA, or messenger genetic material, operated in a similar fashion. The only difference was that their RNA reproduced. Working separately in labs hundreds of miles apart, Temin and Baltimore showed that the Rous sarcoma virus replicated by producing a DNA template of itself after it infected a host cell, which then became part of the cell’s normal DNA machinery. They were called retroviruses. Temin and Baltimore won the Nobel Prize in 1975 for discovering the enzyme that facilitated the process, which they called reverse transcriptase. Not only did scientists now have a theory for how cancer evolved inside the body, they had a plausible model for how it happened.
Although these stirrings in the world of basic science encouraged advocates to push Congress for a broader war on cancer, they played a secondary role in the debate. When Lasker began lining up political support for bigger anticancer budgets, she turned primarily to the already well-established network active in the search for medicinal cures. She leaned heavily on Farber, the aging patriarch of cancer research. He told Congress that more money and a federally directed battle plan would get the job done. “Based on the new insights with which I am familiar there is no question in my mind that if we make this effort today, and if we plan it, organize it, and fund it correctly, we will in a relatively short period of time make vast inroads in the cancer problem,” Farber testified in support of the legislation. “Everything that we have said finally must be directed and pointed to clearly for the benefit of the patient with cancer and all of the basic research and the great expansion of clinical investigation which we have recommended will be the surest way of bringing that about.” The rhetoric of the day called for curing cancer by the nation’s bicentennial, a scant six years off.29
Champions of basic science fought back, led by Francis Moore, a professor of surgery at Harvard Medical School and chief surgeon at Boston’s Peter Bent Brigham Hospital. Moore argued that a government-run institute would never have funded German scientist Paul Langerhans, who in 1889 discovered the cells in the pancreas that, when damaged, led to diabetes. Nor would it have funded Frederick Grant Banting and Charles Herbert Best, who discovered insulin in 1921 at the University of Toronto (they patented their method for purifying it from livestock headed for slaughter and licensed it for one dollar to any drug company that wanted to produce it). Moore argued that discoveries almost inevitably came from “young people, often unheard of people” housed in universities. His testimony bolstered those in the House who wanted to keep the beefed up NCI underneath the NIH umbrella and maintain its system of peer-reviewed science.30
In the end, the heated arguments that pitted advocates of basic science against applied researchers didn’t matter much. NCI budgets quintupled over the next decade to nearly a billion dollars a year. It was enough money for the agency to support a massive expansion of every program in its purview. Basic research grants; a far-reaching hunt for new drugs; the nationwide network of academic clinicians willing to test them—everyone got a piece of the growing pie.
On the basic science front, tens of millions of dollars poured into the SVCP. Far larger amounts went for related basic science research at the nation’s universities. It eventually bore fruit. In 1976, two virus hunters, Michael Bishop at the University of California at San Francisco and his ambitious young postdoc, Harold Varmus, who would go on to become head of NIH and Memorial Sloan-Kettering, discovered something that suggested a very different theory of cancer causation. They found a gene inside cells for regulating cell growth that, when mutated, made those cells cancerous. They called the healthy version a proto-oncogene, and its mutated version an oncogene. Their Nobel Prize-winning discovery opened the floodgates for other researchers, who within a few years identified a number of proto-oncogenes—given names like erb, ras, myb, and src—whose mutations led to cancer.
Ironically, the discovery of oncogenes provided a temporary opening for the chemical carcinogenesis theorists. If a virus could transform a tiny stretch of cellular DNA and make a cell cancerous, then it stood to reason that a constant assault on cells by mutagenic external agents would do the same thing. Their cause got a major boost from Bruce Ames, chairman of the biochemistry department of the University of California at Berkeley. In the mid-1970s Ames developed a simple way to test the mutagenic properties of chemicals. Previously, chemicals had to be tested on rats or mice to see if they caused cancer, an expensive, time-consuming process. By combining human liver extract with fast-growing bacteria, Ames created an assay that rapidly revealed the mutagenic properties of any given chemical. He gained national attention for his test in 1977 when he showed that a popular flame retardant used in children’s pajamas was a carcinogen. Overnight, regulators began using the Ames test on industrially and commercially available chemicals to determine their carcinogenic properties.31
The following year, the anticarcinogen movement reached its apogee when NCI, the National Institute of Environmental Health Sciences, and the National Institute for Occupational Safety and Health published a paper that estimated that anywhere from 20 to 40 percent of all cancers were environmentally induced. The paper has been called the most radical environmental document ever produced by the U.S. government. Joseph Califano, secretary of Health, Education, and Welfare, gave a rousing speech to the nation’s labor leaders at an annual meeting of the AFL-CIO charging that as much as 40 percent of all cancers were caused by six commonly used industrial chemicals.
The study closely paralleled the work of physician and epidemiologist Samuel S. Epstein, who in 1978 published his encyclopedic The Politics of Cancer. Epstein accused industry of hiding data that showed many of its products and chemical intermediates caused cancer. He attacked NCI for focusing almost exclusively on the mechanisms of cancer while ignoring its causes. “Billions for cures, barely a cent to prevent,” Epstein summed up in an environmentalist newsletter.32
Rival epidemiologists immediately began poking holes in the Epstein thesis. In 1981, Oxford University’s Richard Doll, who had helped put cigarettes in the dock during the 1950s, and his protégé Richard Peto, conducted a study for the U.S. Office of Technology Assessment that suggested two-thirds of all cancers were caused either by smoking (30 percent) or diet. Cancers caused by either occupational exposures or exposure by the general population to industrially generated chemicals accounted for just 6 percent of cases, they said. Industrial trade groups began echoing those findings by funding a host of scientists willing to raise doubts about the chemical carcinogenesis thesis.
The election of Ronald Reagan in 1980 signaled a shift in the nation’s mood, and the antienvironmentalist message began falling on receptive ears. Even Ames switched sides. He used his test to show that many common foods and plants were also mutagens, a natural defense against animal predators developed over many eons in the wilds. By the late 1980s, the general consensus among scientists was that environmental exposures other than cigarette smoke were only minor contributors to the overall cancer burden, representing somewhere between 2 and 6 percent of all cases. A 1988 broadcast of ABC-TV’s 20/20 featured Ames and declared that “for two decades now, we of the media have brought story after story where experts warn of links between all kinds of pollutants and cancer. But tonight a distinguished research scientist makes a case that many of the warnings we hear are unnecessary, that all the concern about this toxin, that pesticide, is ‘Much Ado about Nothing.’ Wouldn’t it be nice if he was right?”33 By the early 1990s, journalistic explorations of the link between environmental toxins and cancer had largely disappeared from the mainstream media. The ranks of academic scientists interested in the topic began to shrink.
The backlash against the environmentalist lobby and the changing politics of the nation created an environment where NCI could largely ignore the links between toxics, the environment, diet, and cancer. Investigators who wanted to explore questions about the concentration of certain cancers among subgroups like blue-collar workers and minorities, or clustered in certain parts of the country, were rarely funded. The results of the few studies that did get done were poorly disseminated. Few researchers focused on the social support needed to get people to stop smoking or the economic and advertising forces that encouraged people to eat unhealthy diets. The billions spent on cancer research largely excluded social scientists who might have provided both the public and policy makers with road maps for reducing the incidence of cancer in American society.
NCI’s basic science budget after 1980 concentrated on delineating the biological mechanisms of various cancers. By the early 1980s, the cancer virus hunt had been shelved. A decade had turned up only two or three minor cancers that were related to viruses, leading some cancer researchers to brand the SVCP “a profligate waste, a major boondoggle.”34 But in fact, the SVCP had created a corps of scientists who thoroughly understood the anatomy of retroviruses. When HIV arrived on the U.S. scene in the early 1980s, those scientists were able to identify the retrovirus rapidly, understand its inner workings, and target its cellular weak spots (see chapter 4).
The SVCP also prompted scientists to begin looking inside the living cancer cell, work that exploded in the early 1980s. By the end of that decade, nearly 80 percent of NCI’s budget was going for basic research and the agency was funding more than half of all microbiology work in the world.35 Working from insights gained from the discovery of oncogenes, scientists began exploring the ways those genes worked. Edward Scolnick, who worked at NCI before going to Merck to run its research labs, conducted pioneering research into the on/off switch in cells that gets locked in the on position when oncogenes go bad. Other scientists focused on the signaling mechanisms on the surface and inside cells. These enzymes, called kinases, ran wild in cancer cells. There are hundreds of them in the body. When a particular kinase associated with a particular cancer was identified, researchers began searching for a drug or a biotechnology-derived monoclonal antibody that might block its action and slow and possibly stop the cancer’s growth.
In the late 1980s, scientists working under Bert Vogelstein at Johns Hopkins University identified the so-called tumor suppressor gene, the body’s natural mechanism for correcting the inevitable mistakes that occur during the ten thousand trillion (that’s ten with fifteen zeroes after it) cell divisions that take place during the average person’s lifetime. Using biotechnology techniques, which by the mid-1980s were readily available in both academic and industrial labs, scientists isolated these genes and their proteins and expressed them in bulk so they could be studied for potential clues as to how they might be manipulated to arrest cancer. “Most every significant basic research project we fund is seeking to characterize the genes that turn tumor growth on and off,” Richard Klausner, head of NCI from 1995 to 2001, said in the mid-1990s. “That represents a major shift in the direction of cancer research over the past few years. It’s becoming pretty clear that the 1990s will be seen as a time when scientists finally uncovered the strategies that will be the foundation for every future research effort to eventually conquer cancer.”36
Another strain of basic research projects focused on the genetic anomalies or family predispositions to developing cancer. Mary-Clair King began her work life as one of Ralph Nader’s raiders before gravitating to genetic research as an outlet for her social activism. In October 1990 she shocked the annual meeting of the American Society of Human Genetics when she reported that her small lab at the University of California at Berkeley (she later moved to the University of Washington) had found a gene responsible for breast cancers that ran in families.37 The news electrified the gene-hunting world, which was just beginning its geometric growth with the money pouring out of the federally funded Human Genome Project (see chapter 3). Within a few years, scientists had identified genes connected to heritable forms of colon cancer and prostate cancer and created tests that could screen people to see who carried the mutations.
However, the hoopla surrounding the gene hunters (Wall Street Journal reporter Michael Waldholz called his 1997 book documenting their exploits Curing Cancer) overshadowed the fact that inheritance—like toxic exposures—plays only a minor role in cancer causation. Women who carry the breast cancer gene account for just one in twenty cases of the disease. Just one in ten men with prostate cancer carries a gene that predisposes him to developing the condition. From an epidemiological point of view, neither is any more significant than environmental toxins as a cause of cancer. Yet gene seekers fervently believe that identifying the genetic flaw associated with some cancer cases will one day help scientists come up with drugs that will mitigate the effects of the flawed genes and perhaps shed light on other forms of the disease. “Do I think all this will some day lead to a cure?” King said. “You bet I do.”38
The drug developers inside NCI plugged away amid this explosion of knowledge about the biochemistry of cancer. In 1979, Tadatsugu Taniguchi of the nonprofit Japanese Cancer Research Institute cloned an interferon gene. Within a few years, more than twenty interferon genes had been identified, mostly by university scientists who immediately transferred the results of their research to biotechnology start-ups. During the 1970s and early 1980s, interferon research was pushed by the American Cancer Society and Mathilde Krim, a geneticist at Memorial Sloan-Kettering who was married to Arthur Krim, former chairman of United Artists who contributed heavily to the Democratic Party. At their urging, the government invested heavily in dozens of new companies’ work on interferon therapies.39 Little came of it. In the middle of the 1980s, similar attention was lavished on interleukin, an immune system booster. Again, this early fruit from the biotechnology revolution generated government grants, venture capital from Wall Street, hope for patients, and intense media scrutiny. But in the end, it had little impact on improving cancer therapy.
The success stories in NCI’s drug development program came from its massive screening efforts. Only after bicyclist Lance Armstrong won his first Tour de France in 1996 did the general public learn about the extraordinary advances that had been made against testicular cancer, which strikes about eight thousand young men between the ages of fifteen and thirty-five every year. By the time the twenty-seven-year-old cyclist sought treatment for one testicle swollen to twice its normal size (“I’m an athlete, I always have little aches and pains,” he told one reporter), the cancer had spread to his lungs and brain. He eventually wound up in the hands of Lawrence Einhorn, a physician at the Indiana University Medical Center who more than any individual had been responsible for the advances in testicular cancer therapy over the previous quarter-century.
Early chemotherapy regimens thrown against testicular cancer in the 1950s had achieved anywhere from a 10- to 20-percent remission rate, with about half those patients ultimately cured. NCI-funded tests at the M. D. Anderson Cancer Center in the 1960s used newer chemotherapy agents, upping the disease-free survival rate to 25 percent. In the mid-1970s, Einhorn, an inquisitive and innovative clinician, began adding cisplatin, a platinum-based compound, to the regimen.
Cisplatin had been discovered in the mid-1960s by Barnett Rosenberg, a biophysicist at Michigan State University studying the effects of electric currents on E. coli bacteria growth. To Rosenberg’s surprise, the bacteria stopped reproducing because of exposure to a compound generated on the platinum electrode. He published his findings in Nature in 1965.40 Alerted to a potential anticancer agent, NCI researchers began testing it against various cancers. But the heavy metal’s severe side effects (it causes horrible nausea and damages the kidneys) led most oncologists to dismiss platinum’s potential in cancer chemotherapy.
Those early experiments would bear fruit a few years later when Einhorn, a testicular cancer specialist, learned from the early literature on cisplatin that the drug had achieved remissions in a few patients with germ cell tumors. Since fast-growing germ cells are precursors to sperm cells, he decided to test it on his patients. He was amazed to see that their tumors “melted away.” Einhorn, thinking he had achieved the major breakthrough that the government’s cancer warriors so desperately desired in the bicentennial year, nervously reported his findings to the May 1976 meeting of the American Society of Clinical Oncology (ASCO). For a brief moment, optimism that the war on cancer had found its magic bullet flourished. But those hopes quickly dissolved. No one knew how cisplatin worked, and it had only limited effects on other tumors. Germ cells, it turns out, are particularly susceptible to chemotherapy, and the human immune system is more resilient in the face of testicular cancer than most other cancers.
Still, it was a near cure for one cancer and a major victory for the government program. Michigan State licensed its method-of-use patent on cisplatin (the molecule itself was a very old and well-known compound) to Bristol-Myers, which marketed the highly effective agent. By the time Armstrong had his bout with the disease, ongoing trials by Einhorn and others had raised the cure rate to 95 percent. Cisplatin was also found useful against several other cancers and for many years was the first line of defense against ovarian cancer until it was replaced by Taxol in 1998.41
Taxol was the most significant victory for the government’s thirty-year screening program. The long and convoluted tale of its development is worth reviewing in depth since the resolution of the controversies surrounding its eventual licensing to Bristol-Myers, its pricing, and its post-approval marketing helps explain the private sector’s growing interest in the 1990s and early twenty-first century in pursuing cancer therapies.
NCI began scouring the natural world for potential cancer drugs in the mid-1950s. The agency was inspired in part by Eli Lilly’s success in developing vinblastine and vincristine, alkaloid extracts of rosy periwinkle. The company’s botanists had been led to the low-growing tropical plant’s enticing pink blossoms by folk doctors in Madagascar. Every human society uses plants as medicine, and their use in cancer treatment is part of the folklore of virtually every culture. The Madagascar medicine men were onto something. By extracting the active ingredients from the blossoms, Eli Lilly’s scientists were able to trigger remissions in some patients with Hodgkin’s lymphoma and childhood leukemia.
Jonathan Hartwell, the overseer of the government’s new natural products drug screening program, would not have had much to do if he waited for leads provided by shamans and folklorists. So between 1958 and 1982 (when the program was temporarily suspended), the agency collected and tested more than 180,000 microbe-derived, 16,000 marine organism-derived, and 114,000 plant-derived extracts to test for their anticancer potential. It was a mammoth government undertaking, one that would have made industrialists Sloan and Kettering proud. One botanist estimated nearly 6 percent of the world’s plant species passed through NCI’s screens. But in the end, only 4 percent of those extracts displayed any activity against cancer in test tubes or mice, and only a half dozen would ever make it into advanced clinical trials.
In 1962, U.S. Department of Agriculture botanists, working on a contract from NCI, tested Taxus brevifolia, an extract from the bark of the Pacific Yew tree. It was one of those rare hits, active against the cancer cells in one of NCI’s early screens. It soon found its way to the laboratories of Monroe Wall at the Research Triangle Institute in North Carolina. Wall, a former Agriculture Department biochemist who had honed his fractionating skills in the government’s hunt for natural sources of cortisone in the 1950s, initially ignored the Yew extract, focusing instead on a rare Chinese tree called Camptotheca acuminata (derivatives of its active ingredient, camptothecin, were later developed at Johns Hopkins and Smith Kline on NCI grants and eventually led to three drugs, including Topotecan, which extends life briefly in lung cancer patients who have failed other therapies). A few years later, Wall finally turned his attention to the bark of Taxus brevifolia, and in June 1967 isolated its active ingredient, an alcohol he named Taxol. “It had a nice ring to it,” he later said. He spent the next four years elucidating its chemical structure and developing methods for its large-scale extraction.42
For half a decade, Taxol languished in NCI’s labs. It was only mildly active against leukemia, a major focus of researchers in those years. It wasn’t until it got into the hands of Susan Horwitz, a molecular pharmacologist at Albert Einstein College of Medicine in New York, that scientists began to recognize its potential. She studied microtubules, a kind of scaffolding that cells temporarily erect when they divide. Peering through her electron microscope, Horwitz saw that Taxol froze the microtubules in place, thus preventing cell division. After publishing her findings in Nature in 1979, interest in Taxol took off.43
Over the next few years, NCI scientists showed Taxol was active against colon and mammary cancers in mice (by the early 1980s, NCI’s skill at breeding mice with various cancers had become a high art and a subject of derision to outsiders who referred to the agency’s scientists as mouse doctors). It took several more years to figure out how to suspend the toxic chemical in a solution so that humans could tolerate it. It wasn’t until April 1984 that the first cancer patients received intravenous infusions of Taxol. It took another year for physicians at seven oncology centers to figure out proper dosing, and even then there was strong opposition to continuing with the experiments since nearly a fifth of patients had strong allergic reactions to the drug. More important, there was no evidence that it shrank tumors. But with NCI’s blessing, oncologists at Johns Hopkins, Albert Einstein, and teaching hospitals grouped into the Eastern Cooperative Oncology Group proceeded with a second round of clinical trials.
By the end of 1986, only the Johns Hopkins group, led by William McGuire, had anything positive to report. Two of his seven patients with ovarian cancer who had failed other therapies showed a partial response, and one showed a marginal response. Word began getting out there was a new treatment for women dying of ovarian cancer, and they flocked to McGuire’s clinic. In May 1988, he reported to ASCO that 30 percent of his refractory ovarian cancer patients (refractory means their cancer had returned or had been unresponsive to earlier treatments) had responded to Taxol therapy. Their tumors had shrunk by at least half and in some cases had disappeared entirely.44
Before proceeding to the third and final stage of clinical trials needed to gain FDA approval, NCI had to confront the major issue that had dogged it throughout its initial decade of experimentation—the problem of Taxol supply. The Pacific Yew is a slow-growing evergreen found in the Pacific Coast range from Northern California to Alaska. Even after 125 years, the tree grows to only thirty feet in height with a diameter of less than a foot. The bark from several trees that size produce enough Taxol to treat only a single patient. For decades, timber companies operating on public lands had treated the trees like trash, burning them after they had taken out the surrounding giants in clear cuts. Now they were a valuable resource.
During the 1980s, NCI hired contractors to harvest the trees and a chemical company to extract the bark’s Taxol. As experimental demand for the drug expanded, so did the number of foresters making a living from harvesting Pacific Yew bark. Some timber companies, such as Weyerhaeuser, became interested in developing the technology after making a strategic decision to add high-value products to their basic logging operations. In terms of value, nothing is higher on the industrial food chain than pharmaceuticals.
But environmentalists soon jumped into the fray. Some saw mass Taxol extraction as a small price to pay for developing an alternative economic engine for a region that depended on clear-cutting the forests of the Pacific Northwest. Others hugged every tree. “This is the ultimate confrontation between medicine and the environment,” said Bruce Chabner, chief of the investigational drugs branch at NCI. “It’s the spotted owl versus people. I love the spotted owl. But I love people more.”45
Only four companies responded to the request for proposals, which Bristol-Myers won over Rhône-Poulenc and two small companies.46 The lack of interest didn’t surprise officials at NCI. Previous drugs tested by the agency had either been licensed directly to private firms or developed in informal collaborations. No company benefited more from this arrangement than Bristol-Myers. It licensed its first NCI-developed drug in 1972 and by the late 1980s had more than a dozen oncology drugs in its portfolio. “They never invented anything themselves but they were great developers,” recalled Joseph Rubinfeld, who ran Bristol-Myers’s oncology division from 1968 to 1980. “They took the jobs that somebody else had done, especially NCI, and commercialized them.”47 The company bolstered its cancer expertise via the revolving door at NCI. Top agency officials who went to work for the firm included Stephen Carter, who was at NCI until 1975; John Douros, who had been chief of the natural products branch in the mid-1980s; and Robert Wittes, a drug development program specialist who would return to NCI in 1990 as its chief clinician after two years at Bristol-Myers.
Rep. Ron Wyden, a Democrat from Oregon, challenged the ethical implications of the government’s dealings with the firm during hearings in 1991 and 1992, after public interest groups branded the Taxol CRADA a giveaway of taxpayer-funded research. By that time, the drug’s success in the clinic suggested it might become cancer’s first blockbuster. Before Taxol, cancer was not a lucrative market, and few drug companies even bothered with the disease. At the time of the NCI and Bristol-Myers CRADA, anticancer drugs accounted for less than 3 percent of pharmaceutical sales.
But Taxol was about to transform the commercial landscape. During the first two years of the CRADA, NCI and Bristol-Myers cosponsored the major third-stage trial on women with advanced ovarian cancer. They also conducted smaller, early-stage trials that tested the drug against breast, colon, gastric, head and neck, prostate, cervical, and lung cancers. The government agency organized and paid for the trials while the company provided the drug. In July 1992, the company submitted data supporting a new drug application to the FDA. Patients given Taxol who had failed previous therapies lived on average almost a year longer than patients without the drug. The FDA approved the drug after just five months of deliberations. In 1994, the agency gave the green light for relapsed breast cancer patients to take Taxol. In that trial, one in four patients saw their tumors shrink more than 50 percent. Their life expectancy increased by nearly a year.
It had taken more than thirty years to bring Taxol from its initial discovery to a government-approved drug. Like most chemotherapy agents, Taxol was no cure. It depleted white blood cells and left patients prone to infection. Their hair fell out and they lost feeling in their fingers and toes. But it extended life in some patients. And in the world of cancer therapy, that was progress.
Bristol-Myers still had to deal with the long-term supply problem. In the late 1980s, NCI had recognized that bark processing would remain an expensive and controversial process. It began funneling millions of dollars to academic chemists to develop a synthetic method of producing Taxol. More than two dozen investigators eventually worked on the problem, including Robert Holton of Florida State University. In 1990 he patented a semisynthetic process that started with the renewable leaves and twigs of Yew trees. Bristol-Myers licensed his invention.
The stage was now set for a company bombshell. In January 1993, with Bill Clinton newly arrived in the White House and rising health care costs high on the Democratic Party’s agenda, Wyden held a hearing of his small business subcommittee to question Taxol’s price and the cozy relationship between NCI and the one pharmaceutical firm that had consistently shown interest in commercializing its products. Bristol-Myers had set its initial Taxol price eight times higher than the price NCI had paid to its contractors to produce the drug. A typical course of therapy would cost patients or their insurers more than ten thousand dollars. Ralph Nader and James Love of the Consumer Project on Technology demanded the government exercise its “march in” rights and establish a reasonable price for the drug. Wyden was listening.
At the public hearing, the Oregon representative, whose constituents had jobs because of this new use of its forest products, read a letter from NCI director Sam Broder. It reviewed the history of Taxol, which showed NCI had not only been instrumental in its development but had been alone nearly every step of the way. “There are limits to what Americans ought to pay for drugs developed through billions of dollars of federal research and federal tax credits,” Wyden complained. “Americans should not be held hostage to drug companies who threaten to walk away from cures if the Congress requires reasonable price justification.”48
Bruce Chabner, the director of NCI’s cancer treatment division, defended the company. The price was in the midrange of cancer drugs, he suggested, and the agency had ensured that poor and uninsured patients would get access to the drug. He then addressed the issue that was at the heart of NCI’s fear of getting involved in drug pricing. Alluding to his recent experiences with Abbott Laboratories (see chapter 5), he said several companies had recently stopped participating in a government program to develop a novel anti-AIDS drug because of the “reasonable pricing” clause. Other potential collaborators had rejected the pricing clause outright. “There is no doubt that companies will not accept the risk of investing large sums in the development of a government product if their freedom to realize a profit is severely compromised,” he said.49
Finally, Zola Horovitz, a vice president at Bristol-Myers, took the microphone. He ignored the price issue and made headlines by declaring the Taxol supply fears were over. The drug would be available to every woman in the United States who needed it. Moreover, the company would no longer harvest the Pacific Yew tree for its bark. By 1994, the company would rely exclusively on synthesized Taxol manufactured through the Holton process. He made only one oblique reference to the controversies swirling around the price of Taxol. “Substantial financial incentives are necessary to justify an enormous investment, sometimes measured in hundreds of millions of dollars, required for the rapid development of new pharmaceutical products,” he said. “Any attempt to regulate prices will destroy the financial incentives necessary to attract private companies to these important collaborative research projects.”50
Bristol-Myers, the major beneficiary of more than three decades of government-funded cancer drug development, had invested nowhere near “hundreds of millions of dollars” in bringing Taxol to market. Only after it received FDA approval did the company begin to invest substantial sums in Taxol—mostly on clinical trials aimed at expanding its use. In 2002, shortly after being sued by twenty-nine state attorneys general for delaying access to generic versions of the drug, chief executive officer Peter R. Dolan wrote a public letter to his employees claiming that Bristol-Myers had invested more than $1 billion since beginning its work on the drug’s development eleven years before. Most of that money had been spent in six hundred postmarketing trials involving more than forty thousand patients, which were aimed at refining and expanding its potential uses.51
None of that investment represented risk capital. It represented a small portion of the ongoing sales of the new product. In its first full year on the market, Taxol generated nearly $600 million in sales. By 2001, Bristol-Myers was selling more than $1 billion of Taxol a year and had reaped more than $8 billion in total sales since the drug’s approval. It was the first billion-dollar blockbuster in the history of cancer chemotherapy. Taxol had done more than extend lives. Bristol-Myers’s price on the government’s discovery turned cancer into a lucrative market.
In the wake of the Taxol discovery, numerous new players jumped into the hunt for anticancer drugs. By the mid-1990s, years of basic biological research into the molecular mechanics of cancer had turned up a range of potential therapeutic targets. The biotechnology revolution on Wall Street had simultaneously turned up hundreds of new companies, flush with venture capital funding, who were willing to explore the possibilities. Sometimes their lead compound—it could be a traditional small molecule drug, a monoclonal antibody that used the body’s immune system to hit its target, or a recombinant protein—was developed in academic labs with government funding. Sometimes the targeted therapy was rooted in a private firm’s own discoveries. But most often the potential therapy was the product of a collaboration between the two, facilitated by the technology transfer process established under the Bayh-Dole Act.
The shifting landscape forced NCI to reevaluate its own drug-hunting strategy. Richard Klausner, who became director of NCI in 1996, began placing bets on academic researchers who had developed potential targeted therapies but had failed for one reason or another to hook up with a drug company. “We were looking for the strength of the science. It could be anything: a peptide, a small molecule, a biological entity, a vector. We wanted to evaluate it through peer review that said it made sense as a cancer drug, based not on market consideration but only on the idea it was aimed at a target and there was data suggesting it might have an effect on cancer cells,” he said. Using its virtual drug company skills, the agency backed a drug candidate through the various development stages, knowing that some company would pick it up if it ever became viable. In the first five years of the new program, the agency backed over fifty compounds with several reaching clinical trials.52
Despite those efforts, the Pharmaceutical Research and Manufacturers Association 2001 survey of cancer medicines in development revealed how much the landscape had been transformed over the previous decade. The trade group counted 402 cancer drugs and vaccines under development at 170 pharmaceutical and biotechnology companies. NCI, either alone or in partnership with a private firm, accounted for just ninety-three of those potential therapies. Most of the major pharmaceutical companies had a few entries on the roster. But the vast majority of candidates were being developed by companies with names like ImmunoGen, Intracel, SuperGen, and NeoPharm—small biotechnology firms whose financial prospects hinged on hitting a home run with the one or two drug candidates in their portfolios.53
To some longtime participants in the cancer wars, the locus of innovation had permanently shifted. “Initiative and creativity have moved to the private sector,” Sam Broder, the former head of NCI, said. “There is just no way of getting around it, and anyone who tells you otherwise is on a different planet. What was done in the early seventies was necessary, even in retrospect, but that doesn’t mean we should do it that way now.”54 After leaving NCI in 1995, Broder worked for a time with a company seeking to bring a generic version of Taxol to market before joining Celera Genomics’ drug discovery team as its chief medical officer.
But Klausner, who left NCI in 2002 to lead the Bill and Melinda Gates Foundation’s program for developing drugs for the less developed world, believes the government still has a major role to play in the hunt for new drugs. “We could support the only laboratory big enough for the entire drug industry, which is everyone,” he said during 2001, the war on cancer’s thirtieth anniversary year. “That’s the right-size laboratory for discovering pharmaceuticals. That doesn’t mean there isn’t a role for a drug discovery in a drug company. But there’s no way for one company to keep track of all the biology that’s out there.”55
Recent experience suggests the financial incentives that drive private firms won’t be sufficient to overcome the hurdles that inevitably stand in the way of researchers seeking to bring new cancer drugs to market. The first targeted molecules that made it through the regulatory process required something as old as science itself—the passion of dedicated researchers who wouldn’t take no for an answer.
Dennis Slamon is one such researcher. Son of a disabled West Virginia truck driver, he attended the unheralded Washington and Jefferson College in Washington, Pennsylvania, before his good grades earned him a full scholarship to the University of Chicago Pritzker School of Medicine, where he simultaneously earned a doctorate in cell biology. After completing his residency, he gravitated to a junior faculty position at the University of California at Los Angeles, at the time considered a backwater in cancer research. Slamon, who preferred lab work to treating patients, began collecting tumor specimens and dreamed of finding oncogenes, which by the time he arrived in the late 1970s had become the hottest topic in cancer research.
In the mid-1980s, Slamon hooked up with Axel Ullrich of Genentech, an experienced gene cloner who had isolated several oncogenes believed to be associated with cancer. Ullrich visited UCLA to present his work in isolating and cloning epidermal growth factor (EGF), which regulates cell growth and overpopulates the surfaces of some cancers. The gene had been discovered by Massachusetts Institute of Technology scientists and cloned by three different labs. But Genentech, which in those days was not far removed from its entrepreneurial roots, gave Ullrich a free hand to pursue the target. He began sending his gene samples to Slamon to test against the DNA extracted from his collection of tumor cultures to see if they could find a match.
In 1986, Slamon hit pay dirt. Actually, it was an undergraduate research assistant in his lab named Wendy Levin who made the discovery since Slamon couldn’t afford his own graduate students or postdocs. Ullrich’s gene for the human-epidermal-growth-factor receptor-2 (Her-2) matched a protein on the surface of some breast and ovarian cancer cells. A normal breast cell has about sixty thousand Her-2 receptors on its surface. But on a cell with a Her-2 mutation, there were more than a million, each telling it to divide. The result was a particularly virulent form of the disease that hits about one in four breast cancer patients.56
While other cancer researchers were dismissing his results as nonreproducible, Slamon began a full-court press to develop monoclonal antibodies aimed at the receptor that might jam up its machinery. The immune system produces antibodies when an invader like a virus or bacteria enters the body. They latch onto a receptor on the invader and stop it from functioning. After isolating the antibody, scientists can use recombinant technology to manufacture it in bulk (the multiple expression of a single antibody is called a monoclonal antibody) so they can give it as a drug. Georges J. F. Köhler and César Milstein—joint winners of the Nobel Prize—developed that critical technology at Cambridge University in the mid-1970s with the support of the British government’s Medical Research Council. Within a decade it found widespread use in university and industry labs around the world. Slamon asked at least three different industrial labs and a similar number of academic labs to make a monoclonal antibody aimed at the Her-2 receptor. “We wouldn’t restrict ourselves to any one person or lab,” Slamon recalled in an interview more than a decade later, “including ones we developed ourselves.” Ullrich’s Genentech lab won the race.57
Slamon ran into his first roadblock. Genentech had been down the cancer road before. It had invested heavily in interferon in the early 1980s based on the media-inflated hype that it was the ultimate cure for cancer. But its alpha-interferon turned out to be effective only against hairy cell leukemia, an extremely rare form of the disease. It licensed the product to Roche, which soon showed it was effective against hepatitis B and Kaposi’s sarcoma (the skin lesions that were the first manifestations of full-blown AIDS) and reaped the rewards from the drug. Genentech’s other anticancer drugs—recombinant versions of gamma-interferon and tumor necrosis factor—failed to show results.
By the late 1980s, Genentech was also a company rapidly shedding its feisty start-up status. Kirk Raab, a former Abbott marketing executive, had taken over the company’s reins in the mid-1980s. The company finally had two FDA-approved drugs in its portfolio—human growth hormone and an anticlotting medicine—and the science-driven agenda of its early years was giving way to the Wall Street imperative for sales and profits. It farmed out most of its clinical trials to contractors, leaving it ill-suited to bringing a brand new drug to market.
A few years later the government began investigating the company for the unethical and possibly illegal promotion of its two FDA-approved products. A congressional subcommittee investigation found the company had granted stock options to outside scientists testing its anti-blood-clotting medicine. Genentech’s overly aggressive sales force used those tests to push sales of the wildly expensive biotech drug even though subsequent trials by nonaffiliated scientists showed it was no better than a much cheaper and older drug. To push human growth hormone, its other product, the company set up a program to measure millions of school children, hoping to convince parents with short kids to put them on the drug.58
It wasn’t until 1995, when Genentech’s board ousted Raab and installed Arthur Levinson, a first-rate scientist who had done his postdoctoral work under oncogene pioneer Michael Bishop at UCSF, that the company put Herceptin on its front burner. Commenting on the scandals surrounding the firm at the time of his takeover, Levinson said, “We haven’t always been proud to work here.”59
Slamon’s relationship with the company bridged that troubled history. Ullrich quit Genentech in 1988, in part because he was frustrated by the company’s refusal to dedicate resources to investigating the Her-2 cancer connection. Over the next several years, Slamon repeatedly traveled to the Bay Area to haunt the corridors of the biotechnology firm, looking for anyone interested in the drug. It limped along with support from Michael Shepard, who had joined Genentech right out of graduate school in 1980 and replaced Ullrich. Shepard complained bitterly about his lack of support inside the firm. “They were really very allergic to cancer,” he said.60 The program also had support from Levinson, who was moving up the ranks of Genentech’s research department, and a vice president of manufacturing, whose mother was dying of breast cancer. Frustrated by the company’s slow pace, Slamon looked to NCI for assistance. Still skeptical about targeted therapies, the agency’s drug development program said no. “Antibodies had been tried in the past and weren’t successful. They said [my proposal] was derivative, not very innovative, and not very new. It got the usual knocks,” he recalled.61
Slamon would not have made it through those rocky years without support from Hollywood, a story that in itself speaks volumes about the serendipitous road to therapeutic innovation.62 In the early 1980s, Brandon Tartikoff, the creator of Cheers and Hill Street Blues and a cancer survivor, felt a swollen gland in his neck. Fearing a recurrence of Hodgkin’s disease, he went to a top physician at UCLA who told him everything was fine. Unconvinced, Tartikoff called his former roommate at Yale, a physician who had been Slamon’s roommate in medical school. The referral got made, and after Slamon reviewed the slides, he correctly diagnosed the cancer’s recurrence. He put Tartikoff on a drug regimen that kept the disease in remission for well over a decade.
Over those years, the Tartikoff and Slamon families grew close. Their kids attended the same schools. Brandon’s wife, Lilly, an accomplished ballet dancer whose career was cut short by injury, decided to devote her life to raising money for Slamon’s lab. She worked her connections in Hollywood and eventually convinced Revlon chairman Ronald Perelman, who made billions peddling cosmetics, to do more for women beyond donating money for a dermatology clinic. “You give millions for zits but nothing for breast cancer,” she told him.63 Perelman’s foundation held its first Fire and Ice Ball in 1990, a black-tie five-hundred-dollars-a-plate affair that attracted more than a thousand of Hollywood’s glitterati. Between 1989 and the drug’s approval in 1998, Revlon would raise more than $13 million to support Slamon, who would later say that it would have taken another decade to bring the drug to market without the foundation’s help.64
The clinical-trials collaboration between a reluctant Genentech and an eager Slamon that took place over the 1990s was a fitful one. The earliest trials for determining a safe dose of a drug now called trastuzumab (trade name Herceptin) justified one of the more important claims for the superiority of targeted therapies. Other than a mild allergic reaction treatable with antihistamines, the drug had almost no side effects. As soon as word of the nontoxic therapy got out, the company found itself besieged by volunteers for its clinical trials and in a nasty confrontation with breast cancer treatment activists. Modeling themselves after the AIDS activists, groups like the National Breast Cancer Coalition, led by breast cancer survivor Frances Visco, demanded that Genentech make the drug available to desperately ill women on a compassionate use basis. The company at first refused. Genentech also resisted Slamon’s recommendations that they combine Herceptin with several different chemotherapy agents in critical third-stage trials in the hopes of finding the best regimen.
The bitterness spilled out when the results of the third-stage trial were announced at the annual meeting of the American Society for Clinical Oncology in Los Angeles in May 1998. Four hundred sixty-nine women whose metastatic breast cancer tumors overexpressed the Her-2 gene had received either traditional chemotherapy or chemotherapy combined with Herceptin. The response rate (the tumors shrank by at least half) increased from 29 percent of patients on chemotherapy alone to 45 percent in the Herceptin group. The disease did not resume progressing for 7.2 months on average in the Herceptin group compared to 4.5 months in the group on chemotherapy alone. The median survival time for patients in both groups was indistinguishable—about two years.65
The drug wasn’t a slam dunk, but it was better than existing therapies and bound to gain FDA approval. Slamon made only a brief presentation at the press conference and refused to attend the party given that night by Genentech to celebrate the trial’s results. “This drug could have come to market anywhere from four to six years earlier if people had believed in the data early on,” he said. “I thank my lucky stars for some people in Genentech like Shepard. But there were others who didn’t believe in it.”66 Subsequent trials using Herceptin alone showed the poisonous chemotherapy could be eliminated from the regimen entirely without increasing a patient’s risk. Moreover, using the drug in one of the regimens that Slamon initially proposed (but was rejected by Genentech) decreased the death rate by 30 percent.67
Herceptin, one of the first targeted therapies that had been derived from three decades of basic science, was no magic bullet. It was only useful for a fourth of breast and ovarian cancer patients, with no clear indication of how long its benefits last. But it was an improvement over previous therapies, and for that, tens of thousands of women every year could be thankful. Genentech later claimed it spent from $150 million to $200 million on developing Herceptin. Within three years of its approval in September 1998, it was generating twice that in annual sales.68
Dennis Slamon’s experience was far from unique. Many FDA-approved cancer therapies have had to overcome scientific skepticism and commercial reluctance, and in nearly every case a dedicated academic researcher made it happen. Tamoxifen, the earliest approved targeted therapy and the most widely prescribed medicine for women with breast cancer, would not exist were it not for Northwestern University’s V. Craig Jordan, whose indefatigable work with the antiestrogenic compound began in the mid-1960s.
Jordan first encountered tamoxifen as a student intern for England’s ICI Pharmaceuticals (now part of AstraZeneca). It was the swinging 1960s, and the estrogen blocker had been developed as a potential birth-control pill. It flopped when ICI clinicians discovered to their chagrin that tamoxifen actually enhanced the chances of pregnancy among sub-fertile women. However, one of its developers, Arthur Walpole, who was Jordan’s doctoral adviser, believed the drug might have anticancer properties. Cancer surgeons had discovered in the 1930s that removing the ovaries—the primary source of estrogen—induced tumor regression in about a third of advanced breast cancer patients. In the mid-1950s, Elwood V. Jensen of the University of Chicago helped identify the estrogen receptor on breast and uterine cells and subsequently theorized that breast cancer patients who responded to ovary removal had developed mutant cells with an overabundance of estrogen receptors.
Drawing on that theory a decade later, Walpole encouraged his prized pupil to pursue using estrogen-blocking tamoxifen as a cancer therapy. “Charged with producing birth-control agents, he could not pursue the notion himself,” Jordan recalled. “They weren’t a cancer company. They had no history in this area. It was not seen to be a big market, maybe a few hundred thousand dollars at the most.”69
Early clinical trials showed tamoxifen reduced tumors temporarily in about a third of patients with late-stage breast cancer. Since the drug had very few side effects, the British government allowed its use in 1973, with the U.S. government following suit five years later. NCI funded the U.S. trials; ICI provided the drug. Lois Trench, an NCI researcher who later went to work for ICI, was the main advocate for the drug within the agency.
Jordan, meanwhile, moved to the United States, first to the Worcester Foundation for Experimental Biology in Massachusetts and eventually to Northwestern University, where he continued working with the drug. Early on, he and his coworkers discovered that it was actually a byproduct of digested tamoxifen that blocked estrogen from latching onto the cancer cells (one variation of the metabolite would become raloxifene, sold by Eli Lilly under the trade name Evista, for treatment of osteoporosis).
Jordan next turned to using tamoxifen for adjuvant, or additional, therapy after breast cancer surgery. “The drug, we hoped, would destroy micrometastases—undetectable tumor cells that had already spread around a woman’s body and that, left unopposed, could evolve into fatal masses,” he said. But for how long? A year of adjuvant therapy wasn’t enough, he quickly discovered. It took years of clinical trials, many of them funded by NCI, to determine that five years of postoperative therapy worked best, reducing the recurrence of cancer by nearly 50 percent.
In 1993, NCI launched a massive Breast Cancer Prevention Trial among 13,388 postmenopausal women to assess tamoxifen’s impact among women who had never had breast cancer. Six years later, the FDA approved its prophylactic use based on the results of that trial, which showed tamoxifen cut the risk nearly in half.70 Once Eli Lilly’s raloxifene was approved for osteoporosis prevention, NCI launched a major comparison trial to see if it worked better than tamoxifen at breast cancer prevention, which as of this writing is still underway. “This is the strength of America,” said Jordan in a thick British accent that has survived his three decades in the United States. “It has made the commitment to find solutions to this problem. The government and taxpayers’ money has been well served in this case.”71
The poster child for targeted therapies is Gleevec, a small-molecule drug that originated in the Swiss labs of Ciba-Geigy, which is now part of Novartis Pharma AG. When the drug was first approved for chronic myeloid leukemia (CML) in May 2001, the press proclaimed a new era in cancer therapy, “the first fruits of some three decades of investment in research into the basic biology of cancer.”72 Yet once again, after stripping away the hyperbole, one finds that it took a dedicated independent researcher, aided by desperate patients, before a reluctant drug company delivered on the promise of its proprietary product.
CML, marked by an explosion of white blood cells up to fifty times greater than normal, strikes about eight thousand people in the United States a year, most of them in their fifties and sixties. People live on average about six years after developing the disease. In the late 1960s and early 1970s, researchers at the University of Pennsylvania discovered that the white blood cells in CML patients were marked by a chromosomal mutation—named the Philadelphia chromosome after the place of its discovery—where a small fragment of one chromosome had shifted onto its neighbor. It took another decade before researchers identified the gene disrupted by the transfer. It turned out to be a gene that triggered cell division. The mutated gene produced a cell surface receptor, one of the hundreds of tyrosine kinase signaling proteins produced in the body, whose switch was permanently turned on or at least was hyperactive. “We knew then that we had our molecular target for CML,” said Owen Witte of UCLA, one of the researchers credited with the discovery.73
Witte spent most of the 1980s trying to interest drug companies in developing inhibitors without success. Most firms weren’t interested in diseases with small patient populations. Companies like Novartis that were developing tyrosine kinase inhibitors focused their attention on receptors implicated in many cancers or those involved in heart disease. But in 1993, Brian Druker, newly recruited to Oregon Health Sciences University in Portland from Harvard Medical School, where he had studied CML, heard that one of the Novartis inhibitors that had been developed to block another receptor was active against the mutant kinase.
He called the company and convinced Nicholas Lyden, the drug’s synthesizer, to send him samples of STI-571, which would become known as Gleevec. It proved extraordinarily potent in the test tube against CML cultures. Over the next five years, Druker became a transcontinental lobbyist demanding that Novartis conduct the preclinical work needed to get the drug ready for human testing. Lyden was on his side, but the small patient population meant he had little clout inside the firm. Testing the drug for animal toxicity; determining how fast it cleared the body; developing a soluble derivative—at every stage of its early development, Druker and Lydon had to push the company to move the drug along.74
Four years into the process, Lydon left the company to start his own firm, leaving no one inside Novartis eager to champion STI-571’s cause. Though the drug was ready to test for safety and proper dosing in humans, Druker had to fight with Novartis to make enough drug to conduct the trial. Voices inside Novartis resisted, arguing the company would be better off focusing on diseases that affected larger patient populations. “I forced them to call the question,” Druker recalled. “I couched it in terms of ‘make a decision’: either get it into clinical trials or license it to me.” Druker, whose research throughout the 1990s was supported by NCI and the Leukemia and Lymphoma Society, convinced the company to produce enough drug to conduct at least one trial. “It was entirely my own lobbying that got the trial going,” he said.75
In December 1999, he reported the startling preliminary results to the nearly ten thousand physicians attending the American Society of Hematology meeting in New Orleans. Every one of thirty-one patients given the drug had their blood counts return to normal, and nine of twenty patients treated for five months or longer had cleared the cancerous white blood cells from their systems entirely.
Even before that meeting, word of the experimental drug’s remarkable success had begun rocketing through the CML patient community, largely via the Internet. The patients mounted a massive letter-writing campaign to Novartis headquarters demanding the company begin producing more drug. One patient submitted a petition to chief executive officer Daniel Vasella with more than four thousand signatures. NCI director Klausner, meanwhile, called Vasella to suggest Novartis collaborate with the government on trials testing STI-571 against gastrointestinal stromal tumors (GIST), a rare stomach cancer, since it also overexpressed the rogue receptor targeted by the molecule.
Stunned by the response, Vasella overrode his go-slow managers and ordered production of commercial quantities of the drug. It would be used for the next round of clinical trials and given free to desperately ill patients in a compassionate use program. “I told people not to worry about excess supplies of STI-571 that might never be sold,” Vasella told the Wall Street Journal. “People had been trying to manage the testing program in a controlled way. We want to get this drug available to patients quickly, and to do that you simply can’t stick to bureaucratic rules.” He also approved clinical trials to test the drug against GIST, but rejected NCI’s offer of help in its initial trials.76
A few months after the New Orleans meeting, STI-571, now known as Gleevec, became the best-known cancer drug in the world in the wake of publication of the completed studies in the New England Journal of Medicine. They showed fifty-three of fifty-four patients with CML responding to the drug. A companion study, had it received the same coverage in the press, might have dampened some of the enthusiasm. It showed that while the drug worked for about two-thirds of patients who had reached the crisis phase of the disease, virtually all of them within a year’s time saw their white blood cell counts begin to mount—a sure sign of resistance by cells with the Philadelphia chromosome. Still, the FDA granted its approval for the drug in near record time and a year later also approved it for GIST.77
The drug was clearly better than interferon, the previously approved first-line therapy against the disease. But would the fast-growing cancer cells figure out a way around it, even in the patients diagnosed early in the disease who responded well to the drug? “The relapse rate after ten years could be 25 percent or 85 percent. The problem is you can’t predict,” Druker said. “If you draw a straight line from our early experience, you’d get 15 to 25 percent after ten years. But the rate could accelerate.”78 He has turned his attention to understanding and characterizing the mutations that survive in those patients, so chemists can one day design a drug that when used in combination with Gleevec might halt the cancer entirely.
The hope that numerous targeted cancer therapies would soon come cascading out of industry’s labs suffered a series of setbacks shortly after Gleevec’s approval. Two companies developing drugs aimed at EGF receptors, which proliferate in a number of major cancers, failed spectacularly during 2002. The FDA rejected AstraZeneca’s Iressa, which had shown early promise in patients with lung cancer, when longer clinical trials showed no effect on survival. An advisory committee would later recommend approval based on a single study that showed it shrank tumors in fourteen of 139 patients who had already failed two other therapies.
One skeptic on the committee who voted against approval, Thomas Fleming, chairman of the biostatistics department at the University of Washington, complained that the committee had been unduly influenced by a well-orchestrated parade of patient testimonials at the meeting. “When you have conducted an expanded access program involving more than twelve thousand patients, as the sponsor for Iressa had done, wouldn’t you expect you could run out such a show, even when an intervention has at best very trivial effects?” he told The Cancer Letter, a newsletter that closely monitors developments in the cancer therapeutics industry. “I remember hearing many such testimonials for laetrile after tens of thousands of U.S. patients had traveled to Mexico to receive this agent in the late 1970s, until scientific trials were conducted that established laetrile provided no benefit.”79
The second fiasco involved ImClone System Inc.’s Erbitux, which also targeted EGF. The company became a household word when insiders dumped its stock after receiving an FDA letter warning that its new drug application was inadequate. ImClone chairman Samuel Waksal downplayed the agency’s concerns when he finally made the FDA letter public. The previous fall, he had negotiated a $2-billion deal with Bristol-Myers that made the medical entrepreneur and Manhattan socialite a millionaire many times over.
For a few days after the FDA’s letter, Waksal’s public reassurances stabilized the company’s stock. But after The Cancer Letter published the details of the FDA letter, which showed the company’s clinical trials would have to be repeated, the resulting publicity and stock price collapse led both Congress and federal prosecutors to launch investigations. Waksal eventually resigned from ImClone and pleaded guilty to securities fraud and bank fraud, while Bristol-Myers took control of its troubled Erbitux project.80
The inventor of Erbitux was John Mendelsohn, director of the M. D. Anderson Cancer Center. His story paralleled the remarkable tales of Slamon, Jordan, and Druker—but only up to a point. It began in the 1970s at the University of California at San Diego, where he went to teach and conduct research after earning his undergraduate and medical degrees from Harvard. The oncogene revolution set off by the Bishop-Varmus discovery put him on the path of searching for uncontrolled growth signals in cancer cells. He focused on EGF since it proliferated wildly in many cancers. He eventually developed a monoclonal antibody that could latch onto the EGF receptor and stop it from functioning.
Though he started the NCI-funded cancer center at San Diego and did his most creative work there, the agency turned down his requests for further funding. Targeted therapies simply weren’t on the agency’s radar screen in those days. So, like Slamon, he turned to private foundations, and by the mid-1980s had used their money to develop a potential drug that worked well in cancerous mice. Once he published his mouse studies, NCI began supporting his work. By the late 1980s Mendelsohn had a version of the large protein molecule that wouldn’t provoke severe immune reactions when injected into humans.81
UCSD, which owned the patent, initially licensed the drug to a small biotechnology start-up called Hybritech. Eli Lilly bought the company in 1992, but rejected working on the promising drug. The license reverted to UCSD. Mendelsohn, meanwhile, moved on to become the chairman of the department of medicine at Memorial Sloan-Kettering in New York. In an effort to rejuvenate work in his potential therapy, he hooked up with Samuel and Harlan Waksal, ambitious brothers who had started ImClone Systems Inc. in 1985 with a goal, as Harlan later told Business Week, to “focus on infectious diseases, cancer, and diagnostics, make some products, get rich, and retire early.”82
Samuel Waksal had worked at NCI and held research posts at Stanford, Tufts, and Mt. Sinai Medical Center before starting ImClone. But he was, according to some of his contemporaries, nothing more than a slick scientific salesman, a fast talker who could dazzle his colleagues with his knowledge of the latest advances without actually conducting his own research. Mendelsohn didn’t learn until later that his partner was forced from each of those positions after being accused of falsifying data, and that his brother Harlan in the early 1980s was convicted of possessing cocaine with intent to distribute, although the conviction was overturned on appeal.83
What was apparent from the beginning was that the Waksal brothers had virtually no experience developing drugs.84 Yet just when the experimental cancer therapy was about to go into clinical trials, Mendelsohn essentially withdrew from day-to-day involvement. While he wanted to push his experimental cancer therapy, he also wanted to pursue his career within the cancer establishment. In 1996, he accepted the top job at M. D. Anderson Cancer Center in Houston. Erbitux became a drug whose scientific champion couldn’t play a hands-on role in its development. Physicians at the institution would eventually take part in the clinical trials of Erbitux. But as a member of ImClone’s board of directors, with a financial stake in the firm and a potential recipient of royalties, Mendelsohn could play no role in either designing or implementing the studies. To do so would have violated conflict-of-interest guidelines he put in place shortly after joining the world famous center.85
Six months later, the FDA weighed in with its evaluation. The data was inadequate and uncontrolled. The evaluators couldn’t tell if the benefits came from the traditional chemotherapy agent used in the trial or Erbitux. Even the choice of patients was suspect. Not only did the agency refuse to consider the new drug application, they noted in their nine-page letter that the company had been repeatedly told in the months leading up to the application that the trials would probably have to be repeated.86
“We screwed up,” Waksal told an investors’ conference a few days later. Within a few months, Bristol-Myers would bounce him from the floundering firm. Mendelsohn, who gave a keynote lecture to the ASCO meeting in Orlando a few months later and was called before Congress several times to explain his actions during the drug’s development, was more upbeat. Erbitux was still a good drug, he insisted. At ASCO, he cited one study showing that all six patients with head and neck cancer responded to a combination of Erbitux and cisplatin. Yet when the full study was released a few days later, it showed just one patient out of forty-four achieving a complete response and just five (11.4 percent of patients) had a partial response that lasted long enough to reach clinical significance.87 In the fall of 2002, Mendelsohn defended the Waksals and ImClone’s actions before an oversight subcommittee of the House Energy and Commerce Committee. “The protocol was developed with the advice of medical oncologists from some of the world’s greatest institutions, twenty-seven of which participated in carrying it out,” he testified. Praising ImClone and Bristol-Myers for launching a new round of trials, he said his greatest regret was that “Erbitux will not be available for patients who need it as soon as we had originally hoped.”88
Are targeted therapies just one more failure on the cancer crusade’s long and winding road? Or are they another incremental step toward more beneficial and more benign treatments that might one day turn many cancers into manageable chronic illnesses? And why is Gleevec so successful, while Erbitux or Iressa seem to have so little impact? To Brian Druker, Gleevec’s champion, the answers can be found in the science that preceded the drugs. More than thirty years of research went into understanding the Philadelphia chromosome and its impact on CML. “Just because you know what targets your drug hits, doesn’t mean you’ve got a good drug,” he said. “You’ve got to have a good target and I don’t think we have that many good targets. We have a huge amount of work to do to validate good targets. I think there will be good targets in each cancer. But the current list is extremely short in my book.”
Druker questions weather the EGF receptor will turn out to be a good target for cancer drugs. It may take years, perhaps decades of additional investigation before the process of angiogenesis in tumors is well understood. “We have pseudo-empiricism in most cancers,” he said. “It’s like taking your car to a mechanic. He looks under the hood and says, ‘I see you have this part here. If I replace it, your car will run better.’ You say, ‘Is that part broken?’ The mechanic says, ‘I don’t know, but it might be.’ ” Many of the new cancer drugs are targeting things in cancer cells that may or may not be driving that cancer. “I think the paradigm will work,” he said, “but we’re not ready in most cancers with the right targets.”89
Jordan, whose career is synonymous with tamoxifen, believes a successful recipe for developing drugs must include investigator passion. During our long discussion, I asked him how he has stayed motivated to work on the same drug for more than thirty years. “If you had five hundred people with this kind of passion, this is how you make progress. Just by throwing money at various things and buying mercenaries to work on that problem doesn’t solve the problem. The following week, they’ll work on something else if there is more money over there,” he said. “You have to want to do it without the money.”90
Ellen Vitetta has brought that kind of passion to her study of non-Hodgkin’s lymphoma, which strikes about forty thousand Americans annually. Over the past two decades, she has studied the cancer, worked on developing a monoclonal antibody that would seek it out, and developed a poison derived from castor beans to attach to the antibody that, hopefully, will kill the cancer. She calls it immunotoxin therapy. A graduate of the New York University School of Medicine, she worked briefly with Kohler and Milstein at Cambridge before moving on to the University of Texas Southwestern Medical Center in Dallas.
What sets her lab apart is that it does all the work of translating her basic science into potential therapies in-house. “We develop the antibodies, we test them, we take them in preclinical trials in primates, we manufacture them in the academic institutions, we do all the FDA testing in-house. We are an entire drug company within an academic institution, and we do it all on federal grants [of about $2 million a year],” she said. Her lab has about a half dozen drugs in its pipeline. “When they begin to emerge, the drug companies begin sniffing around. Some of them want a license; some want to share in the trials; some want to sponsor the research and let us do the work; some want the rights if it ever reaches the finish line. My own view is I don’t want them telling me what to do. I want to do this in a scientifically driven way.”91
The story thus far has focused on documenting the government’s role in fostering basic science and the early steps of the drug innovation process. It has shown how taxpayer-funded directed research played the leading role in the battle against some of the nation’s most pressing health care problems like AIDS and cancer. In almost every case, pharmaceutical and biotechnology companies also played critical roles in bringing new drugs to market. Sometimes the role began early in the process, in chemically synthesizing a new drug. Sometimes it began late in the process, when the new drug was well along in clinical trials. Sometimes the private firms invested large sums in the process. Sometimes participation cost the drug company next to nothing.
So a central question remains: Do the steps drug companies play in bringing an innovative drug to market demand the extraordinary sums industry pours (over $30 billion in 2001) into research and development? Does the cost of these steps justify the high cost the industry charges the public for drugs?
The next two chapters seek to answer those questions. But first we must go back to the earliest days of the modern pharmaceutical industry to trace the history of an industry practice that has nothing to do with bringing innovative medicines to market, a practice that at the dawn of the new century accounted for more than half of all industry research and development.