Miracle Cure

by William Rosen

  The battle lines had been drawn at the 1956 antibiotics symposium. On one side, Welch and Martí-Ibáñez argued, “The final verdict on the value of a new drug or a new therapy usually comes from one dependable source: the whole body of practicing physicians whose daily clinical experiences extend over many patients treated in actual conditions of practice over considerable periods of time. Medical practice itself provides the sole and ultimate verdict on the true value of a drug. . . .”

  On the other side were Max Finland, the infectious disease specialist from Harvard Medical School who had been one of the early skeptics about Aureomycin; and Harry Dowling, who had been Finland’s protégé at Harvard’s Thorndike Memorial Library and was, in 1956, chair of the Department of Preventive Medicine at the University of Illinois Medical School. Finland and Dowling, pretty good phrasemakers themselves, sniffed that a physician who chose an antibiotic based on testimonials, rather than peer-reviewed randomized clinical trials, was engaging in “therapeutics by vote.”* And, just so no one missed the point that such votes were very easy to rig, Harry Dowling addressed the 1957 annual meeting of the AMA with a much-reprinted speech entitled “Twixt the Cup and the Lip” in which he attacked the use of the techniques “that had been used so successfully in the advertising of soaps and tooth pastes and of cigarettes, automobiles, and whiskey” to market drugs to doctors.

  In some ways, the history of antibiotic discovery, from the days of Koch and Pasteur through Paul Ehrlich, Gerhard Domagk, Fleming, Florey, and Hodgkin, can be read as an exercise in epistemology. The great innovations, from Robert Burns Woodward’s elegant systems for elucidating chemical structure to Selman Waksman’s brute force technique for finding useful soil-dwelling bacteria, were as much about developing new methods for creating knowledge as they were about the knowledge itself. Likewise, no matter how much it looks like a battle over selling one’s soul to Pfizer or Merck, the conflict between Welch and Martí-Ibáñez on one side, and Finland and Dowling on the other, was really epistemological: What is the best way to know what actually cures disease?

  The significance of this battle to the practice of medicine can hardly be overstated. Several hundred thousand American physicians and at least as many overseas were, for the first time, pushed and pulled by two opposing forces: First was their virtually complete dependence on pharmaceutical companies for information about the relative effectiveness of the drugs they prescribed. Second, their no less complete insistence on autonomy in deciding which ones to use. The resulting gap wasn’t about information per se, but rather about credibility: Which claims are trustworthy, which not, and, especially, what cognitive tools can be used to decide between the two?

  It’s tempting to portray the conflict in simple terms. Doctors, after all, remain enormously respected in part because they are at least nominally obliged to place the patient’s good above any other consideration, while pharmaceutical companies, however large their contribution to human welfare, remain for-profit enterprises. There are complicating factors, of course. Physicians’ practices and hospitals are also businesses, after all; pharmaceutical company researchers are as motivated by the glory of discovery as by their stock options. Even more confounding was and remains the phenomenon that privileges personal knowledge over aggregated evidence, which manifests in doctoring in a particularly acute version. Physicians are famously confident in their clinical experience, inclined to trust the results from a dozen patients they’ve treated, even in the face of a study examining a thousand patients they’ve never seen. One possible solution was to evaluate drug effectiveness collectively, rather than individually. But the AMA had, in 1953, stopped issuing its “Seal of Approval” that permitted drugs to be advertised in JAMA, and dissolved its at least ostensibly regulatory Council on Pharmacy and Chemistry.

  In the face of these cognitive and political challenges, the key epistemological question of the antibiotic revolution—how can we know what medical interventions actually work?—remained unanswered. Because of the increasing complexity of medical decision making, and the cognitive biases that accompany virtually all human behavior, it sometimes seems to be unanswerable. One question that the last century of medicine has answered, though, is how much reliance ought to be placed on the counsel of a single clinician; to epidemiologists and public health researchers, the scariest thing said by a physician is any sentence that begins with, “In my experience.”

  —

  The hearings that began in Washington, DC, almost exactly eleven months after John Lear’s first article in Saturday Review were not called in order to solve an epistemological crisis about drug efficacy. As one might guess from the charter of the subcommittee that conducted the hearings—a branch of the Senate Judiciary Committee responsible for oversight on antitrust and monopoly issues—the original intent was to review how pharmaceutical companies priced and marketed their products.

  An investigation into drug pricing had been gestating for some time. In 1953, John Blair, a staff economist at the Federal Trade Commission, persuaded his bosses at the FTC to start an inquiry into the business of manufacturing and selling antibiotics. Though Blair was, to put it gently, no friend to large business organizations—in 1938, he had published a polemic entitled Seeds of Destruction: A Study in the Functional Weaknesses of Capitalism, a “none-too-happy picture of capitalism and its probable future”—his argument wasn’t especially ideological. His own physician and pharmacist had informed him that the price for each of the branded broad-spectrum antibiotics, whether Aureomycin, Terramycin, or Chloromycetin, was identical, and likely to stay that way. The reasons, Blair learned in his preliminary investigation, were the dizzying cross-licensing and cooperative marketing arrangements that followed the tetracycline peace treaty. Parke-Davis, for example, as Blair put it, “sold twenty of [the] fifty-one major drug products included in our study, but produced only one: Chloromycetin.”

  The business seemed ripe for a broad antitrust investigation, but Blair was unable to convince the commission to proceed. Instead, he took an oblique approach. In 1956, the FTC began work on an exhaustive Economic Report on Antibiotics Manufacture. When it was published in June 1958, it revealed that antibiotics had made the pharmaceutical industry the country’s most profitable business, with overall profit margins of nearly 11 percent after taxes—twice as much as the average U.S. corporation. For those firms lucky enough to produce broad-spectrum antibiotics, margins were as high as 27 percent. The seed that had been planted by the OSRD’s penicillin project fifteen years before had produced some extremely rich fruit.

  The day after the report was issued, the FTC issued a complaint alleging collusion in the marketing of broad-spectrum antibiotics, and questioning the whole structure of the tetracycline patent and cross-licensing agreements. Even so, it took another year, and John Lear’s series in the Saturday Review, before the subject attracted the attention of anyone outside the bureaucratic world of antitrust regulation.

  In September 1959, when Senator Estes Kefauver of Tennessee announced that the Antitrust and Monopoly Subcommittee would hold hearings on America’s drug business, he was already one of the country’s best-known politicians. He had become a television and newsreel star as the chairman of the Senate Special Committee to Investigate Organized Crime in Interstate Commerce, which riveted the country in 1950 and 1951: Every time a mobster like Frank Costello or Joey Adonis, among the more colorful figures ever to appear before a congressional investigation, invoked Fifth Amendment protections against self-incrimination, the man sitting opposite was the senior senator from the Volunteer State. Kefauver, a candidate for the Democratic presidential nomination in 1952 and its nominee for vice president in 1956, was a New Dealer with impeccable liberal credentials—in 1956, he was one of only three southern senators to refuse to sign the prosegregation Southern Manifesto (the others were Lyndon Johnson and Albert Gore, Sr.)—and a reflexive hostility toward big business of any variety. In short, he was th
e pharmaceutical industry’s worst nightmare: a combination of liberal populism and intellectual range, with an understanding of both old-fashioned political theater and the power of modern media.

  Even worse (or better): The first person Kefauver hired to join the subcommittee’s staff was John Blair.

  On December 7, 1959, in the Old Senate Office Building, the first witnesses were sworn in. For the next ten months, they would arrive, be interrogated in turns by friendly and unfriendly staff and committee members, and depart. As promised, the first subject on offer was pricing: what Kefauver and Blair saw as a shocking distance between the cost of manufacturing a particular drug and its price to consumers. It wasn’t just that the prices themselves were what Kefauver thought to be egregiously high. For the Antitrust and Monopoly Subcommittee, what mattered weren’t high prices as such, but the suspicion that the prices were being artificially inflated by a conspiracy in restraint of free trade. Since the demand for drugs was determined not by patients, but by a physician’s prescription pad, the place to look for such a conspiracy was in the unique marketing practices of the pharmaceutical industry. “The drug industry,” as Kefauver put it, “is unusual in that he who buys does not order, and he who orders does not buy.”

  By 1959, the pharmaceutical industry was developing and marketing considerably more than the antibiotics that had jump-started its growth the preceding decade. As a result, the first witnesses called by the subcommittee gave testimony on other, newer, wonder drugs. One of the first to be examined was the corticosteroid prednisone, an immunosuppressant used to treat diseases like colitis and multiple sclerosis, in which the symptoms are frequently caused by the immune system’s own inflammatory response. Francis C. Brown, the president of the Schering Corporation, which introduced the drug in 1955 under the name Meticorten, was blindsided by the initial line of questioning: Why, he was asked, was his company charging some seven hundred times more for a dose of prednisone than it cost to manufacture it? Despite attempts to explain that the price of a drug had to reflect its fixed development expenses as well as its marginal manufacturing cost, Brown had already lost the public relations battle. The front page of the New York Times for December 8 read, “SENATE PANEL CITES MARK-UP ON DRUGS RANGING TO 7,079%.”

  And so it went, for month after month. The subcommittee’s chief counsel, Rand Dixon, stated for the record that the Upjohn Company used only 14 cents worth of raw materials to make a drug it sold for $15, a markup of “about 10,000 percent.” Corticosteroids gave way to tranquilizers—the term had only recently been coined to describe Miltown, the brand name for the mild sedative meprobamate, and the world’s first blockbuster psychotropic drug—to arthritis medications, to antidiabetic drugs. Physicians and hospital directors went on record accusing pharmaceutical firms of “brainwashing” tactics and “perverted marketing attitudes.”

  Meanwhile, the subcommittee’s Republicans, led by Senator Everett Dirksen of Illinois,* counterpunched, asking why, if pharmaceutical companies were marking up prices several thousandfold, they were still only managing to achieve profit margins of less than 15 percent.

  By the spring of 1960, however, the subcommittee’s concerns had expanded from pricing and marketing strategies to patent and trademark reform, particularly regarding the ways in which branded drugs—especially the proprietary fixed-dose combination drugs like Sigmamycin—were being offered the same sort of intellectual property protections as compounds like streptomycin, even though they weren’t required to show that they were truly novel or even effective. The stage was set for the main event: antibiotics.

  The stars of the hearings’ climax ought to have been Henry Welch and Félix Martí-Ibáñez, both of whom had demanded an opportunity to appear and clear their names. Kefauver had taken the two up on their offers, notifying them that they would be given a chance to do so at the hearing scheduled for May 17, 1960. Neither appeared, pleading illnesses that stubbornly, and suspiciously, hung on until the hearings ended in September.

  Their absence had little impact on the hearings’ theatrics. At one point, Rand Dixon asked Dr. Perrin Long, a pioneer in the use of sulfa drugs, “Do you think a cost of $17 [for antibiotics] to the average mother or father every time their child has a cold is down to a point where it can be reached even by the needy?” The question made headlines; Dr. Long’s most relevant response—that an antibiotic would be useless at any price for treating a cold—went unsaid. Harry Loynd of Parke-Davis, called to account for both the marginal cost for the active ingredient in Chloromycetin and the battle over the way it had been promoted during the aplastic anemia scare, was a perfect foil for Kefauver. Loynd had never learned to hide his contempt for politicians, and came off not as a no-nonsense executive without the time to suffer fools gladly, but as an arrogant stuffed shirt. And, worse, an evasive one, who fought with Kefauver over every comma in every document, even to the point of arguing whether he had “seen” or simply “been aware” of an ad for Chloromycetin that seemed to underplay its risks.

  The twenty-one months that began with John Lear’s first article for the Saturday Review on January 3, 1959, and ended with the close of the Kefauver hearings on September 14, 1960, were as earthshaking, in their own way, as the two years following the transatlantic trip of Howard Florey and Norman Heatley in 1941. Though specific drugs and pharmaceutical companies had for years been the objects of criticism from academic physicians like Maxwell Finland and gadflies like Edgar Elfstrom, in 1959, the public at large remained almost uniformly optimistic about the era of the wonder drugs, in which miracles—from penicillin to the Salk vaccine—had appeared, it seemed, almost every day. By the end of 1960, they would never be quite so enthusiastic again. The most dispiriting news about the wonder drugs, it turned out, wasn’t that they were overpriced; it was that no one knew whether they were really effective. Haskell Weinstein, formerly the medical director of one of Pfizer’s subsidiaries, had revealed the “very prevalent misconception” that the FDA was required to document efficacy. “As a physician I blush with shame at the quality of some of the ‘studies’ done by some of my physician brethren.”

  When the Kefauver hearings closed in September 1960, the subcommittee had remarkably little to say about drug pricing or pharmaceutical monopolies. It recommended, instead, expanding the statutory authority of the FDA, which had largely been frozen since 1938. Henceforth, the agency should require proof of efficacy, as well as safety, of all new drugs, and “apply certification procedures to all antimicrobial agents for use in infectious diseases.” In April 1961, Kefauver introduced Senate Bill 1522 along essentially the same lines.

  Testimony on the bill would occupy the next seven months. The FDA was—no surprise—a big supporter. More surprising: So were many of the largest pharmaceutical firms, though not Parke-Davis; Harry Loynd was still smarting from his treatment by the senior senator from Tennessee. On the other hand, the American Medical Association opposed Kefauver’s bill most strenuously because their membership was violently hostile to the idea that drug efficacy could be known by anyone other than the individual physician. Unstated was the concern that the restrictions on pharmaceutical advertising that Kefauver had included in the bill would fall most heavily on JAMA.*

  Despite widespread public endorsements, the support of strong majorities in both houses of Congress, and even the pharmaceutical industry itself, SB1522 seemed doomed to die the death of a thousand cuts in committee. And so it might have, but for a scandal more gruesome, and more notorious, than the Elixir Sulfanilamide and aplastic anemia scares combined.

  —

  In 1962, Frances Oldham Kelsey had been balancing the costs and benefits of the antibiotic revolution for nearly twenty-five years. In 1938, with the ink barely dry on her doctorate in pharmacology from the University of Chicago, she performed the animal studies that revealed the extent of the damage caused by Massengill’s Elixir Sulfanilamide. Twelve years later, she became an MD as well as a P
hD; and, in August 1960, Dr. Kelsey joined the Food and Drug Administration as one of only seven full-time drug reviewers. Her first assignment was an application from Richardson-Merrell, Inc., a drug wholesaler in Cincinnati then best known for the menthol-infused petrolatum known as Vicks VapoRub. The drug for which the company had applied for U.S. marketing rights, to be called Kevadon, had achieved enormous popularity in western Europe as a sedative: one that was safer than barbiturates, and, in addition, was effective as an antinausea drug. The German pharmaceutical firm Chemie Grünenthal—a postwar company that got its start selling penicillin under the Allied occupation—had developed and sold it, first by prescription, then direct to consumers, as Contergan. In the United Kingdom, Distillers Limited marketed it as Distaval. Its generic name was thalidomide.

  In 1960, pharmaceutical companies were able to contact FDA reviewers directly as often as they pleased, and Richardson-Merrell pressed Dr. Kelsey for a quick approval to their application, which they expected in a matter of months, given the widespread use of the drug in Europe. In fact, the FDA automatically granted approval for new drugs if a reviewer failed to act on applications within sixty days . . . but action was understood to include any request for additional information. And Frances Kelsey definitely wanted more information. She asked for clinical and animal studies on toxicity. She requested data on the drug’s effect on pregnant women, given that it was being touted as a safe cure for morning sickness. Richardson-Merrell supplied testimonials. They sent more letters, and what purported to be the best research on the new product. Kelsey called it “an interesting collection of meaningless pseudoscientific jargon apparently intended to impress chemically unsophisticated readers.” Drug company representatives visited Kelsey in her office dozens of times. Richardson-Merrell’s executives went over Kelsey’s head, to George Larrick, the commissioner of Food and Drugs. Larrick—the inspector who had mobilized the entire FDA field force to track down virtually every drop of Elixir Sulfanilamide in 1937—backed his reviewer. Every sixty days, Frances Kelsey sent another letter informing Richardson-Merrell that their application still awaited approval.