Miracle Cure

by William Rosen

  In 1945, they hit pay dirt—literally—closer to home, with the sample from Plot 23 at Sanborn Field. Living in Plot 23 was a yellow actinomycete, a relative of Selman Waksman’s Streptomyces griseus that they designated A-377. It took nearly three years of testing and experiment before Duggar announced his discovery to the world in an article in which he named his newly discovered organism Streptomyces aureofaciens: the “gold maker.”

  The name was almost certainly an attempt to describe the appearance of the bacterium, but Duggar’s employers at Lederle may have had a different meaning in mind. S. aureofaciens produced a chemical of unknown identity and structure, but one that checked the activity of an enormously wider variety of bacteria than either penicillin or streptomycin. The substance, which Duggar christened Aureomycin (aurum is Latin for gold), was effective against Gram-positive and Gram-negative bacteria, including the pathogens responsible for common ailments like urinary tract infections, and rare ones like bubonic plague. It even seemed to have a powerful effect against a number of viruses. The first broad-spectrum antibiotic had been discovered.

  In 1948, after a series of highly successful animal experiments, Aureomycin was ready for clinical investigation in humans. The facility chosen was Harlem Hospital, where Louis Tompkins Wright had spent years studying the treatment of diseases, such as the sexually transmitted lymphatic infection caused by the bacterium known as Chlamydia trachomitis. Wright—the most famous African American physician in the United States, the first admitted to the American College of Surgeons—succeeded brilliantly, not just against chlamydia, but also on the variety of pneumonia caused by a virus, rather than the pneumococci bacteria.

  Aureomycin looked like a true magic bullet: the hoped-for drug that would cure nearly everything. The Harlem Hospital results didn’t convince everyone; Maxwell Finland of Harvard Medical School, perhaps the country’s most respected expert on infectious disease, found that Wright’s reports were “tinged with enthusiasm,” and he didn’t mean it as a compliment. In 1948, this made him a voice in the wilderness.* Lederle promoted it as “the most versatile antibiotic yet discovered, with a wider range of activity than any other known remedy.” It wasn’t just that it was superior to existing antibacterial treatments in treating disease (though it was). Unlike Prontosil, or penicillin, or—thanks to George Merck—even streptomycin, Aureomycin was patentable, and, on September 13, 1949, it was granted U.S. patent number 2,482,055. Even before the patent had been approved, in 1948 the company spent a then-unheard-of $2 million shipping samples of their gold maker to 142,000 doctors. Lederle had the first blockbuster drug in history, but it wouldn’t have the stage to itself for long.

  In 1945, Brooklyn-based Pfizer had initiated its own global soil collection program with the same goal as Lederle: to find a patentable antibiotic. They went about it the same way that Waksman and Schatz had discovered streptomycin, by testing large numbers of soil samples—very large numbers. Within a few years, Pfizer had collected more than 135,000 of them. One of Pfizer’s chemists, Ben Sobin, later recalled, “We got soil samples from cemeteries; we had balloons up in the air [that] collected soil samples that were windborne; we got soil from the bottoms of mine shafts . . . from the bottom of the ocean.” By the beginning of 1949, its investigators had conducted more than twenty million tests on them at the company’s Terre Haute, Indiana, facility, a state-of-the-art microbiology laboratory.

  In a replay of Lederle’s experience, the payoff came not from an exotic location but from the land around one of Pfizer’s own Midwest plants. They had found a yellowish actinomycete that Alexander Finlay, Pfizer’s team leader, named Streptomyces rimosus.* The canary yellow crystals it produced were given a code name: PA-76, for the seventy-sixth culture of a Pfizer antibiotic.

  At first, PA-76 seemed to be virtually identical to Lederle’s Aureomycin, which made it interesting but commercially problematic. Pfizer nonetheless proceeded to invest, since PA-76 had enormous potential to become Pfizer’s own blockbuster. It killed or at least slowed down Gram-positive and Gram-negative bacteria as well as dozens of fungi and, seemingly, even viruses. The honor of naming it was given not to its discoverer, but Pfizer’s new president, John McKeen. McKeen had designed and converted Pfizer’s Brooklyn fermentation plant, a former ice factory, for the wartime penicillin project. He had risen through the ranks to succeed John L. Smith as Pfizer’s president, and named it Terramycin, because, as he later wrote, “I wanted a name connected with the earth, and one that could easily be recalled by doctors and scientists and people in general, because it came from the earth.”

  In 1949, McKeen assigned Gladys Hobby, the microbiologist who had been part of the Columbia University team that led the way in American experimentation on penicillin, now a Pfizer team member, to take charge of testing the new drug. She wasted little time. On December 31, 1950, Hobby conducted Terramycin’s first human trial at Harlem Hospital, which had demonstrated in the Aureomycin experiments that it had the personnel and structure in place (including carefully identified pathogens) to perform a high-quality clinical trial.

  Her employers couldn’t have been more enthusiastic about her results. Though Pfizer had produced huge quantities of penicillin and streptomycin—far more than any other pharmaceutical company in the world—they weren’t making much money selling it. In March 1950, McKeen famously gave a speech to the New York Society of Security Analysts, in which he told them, “If you want to lose your shirt in a hurry, start making penicillin and streptomycin.” The surest way out of the trap that had swallowed up the first antibiotics was finding a new drug that was superior to its competition, so the company that found it could profit from a de facto monopoly.

  Now Pfizer had their drug: Terramycin was just as good as hoped. However, no one at Pfizer or elsewhere thought it was demonstrably superior to Aureomycin, and the Lederle drug already had a head start in winning the hearts and minds of America’s physicians and pharmacists. By the beginning of 1950, Aureomycin accounted for 26 percent of the entire antibiotic market in the United States.

  Moreover, Pfizer wasn’t even really in the business of selling drugs. All of the penicillin and streptomycin they had produced to date had been sold under the label of other companies, ones that had consumer sales and marketing capabilities that Pfizer—which had, after all, been largely a manufacturer of citric acid before joining the penicillin project—lacked. McKeen was undaunted. Terramycin would be a company changer: Pfizer’s first branded drug.

  But first, the company needed to know more about both Terramycin’s mechanism and its structure, a task requiring the most sophisticated understanding of organic chemistry. Pfizer needed the best organic chemist in the world. They needed Robert Burns Woodward.

  Woodward was a rare bird: a prodigy—he entered the Massachusetts Institute of Technology in 1933 at age sixteen, and left four years later as a twenty-year-old PhD in chemistry—whose adult achievements were even more prodigious. During his forty-one years at Harvard’s Department of Chemistry, he authored or coauthored nearly two hundred peer-reviewed papers, received twenty-four honorary degrees and twenty-six medals and awards, including the National Medal of Science, the Royal Society’s Copley and Davy Medals, and the 1965 Nobel Prize in Chemistry for his “outstanding achievements in the art of organic synthesis.”

  Woodward was the first to synthesize cortisone, cholesterol, strychnine, and chlorophyll. In 1944, as an advisor to the War Production Board, he discovered how to synthesize the antimalarial compound quinine, which was utterly essential for the war effort in both southern Europe and throughout the campaign in the Pacific . . . especially so since, from 1941 forward, the world’s entire supply of the only natural source for quinine, cinchona tree bark, was under the control of the Imperial Japanese Army. And, seven years after winning the Nobel, Woodward succeeded in one of the most impressive tasks in the then-brief history of chemical synthesis, leading the international team that spent tw
elve years decoding and producing the notoriously complicated molecule known as vitamin B12.

  But calling Woodward a master of chemical synthesis, though true, understates his gifts. To his contemporaries, he was even more brilliant at describing complex organic chemicals—deciphering the incredibly complicated three-dimensional shapes adopted by the stuff of life—than in making them.* In January 1945, it was Woodward who demonstrated that the beta-lactam structure of penicillin, the one proposed by Ernst Chain and E. P. Abraham, had to be correct, thus anticipating Dorothy Crowfoot Hodgkin’s X-ray crystallography by five months. In the words of one of his biographers, he could integrate facts “both clear and misleading into a coherent whole better than any chemist who ever lived.”

  Credit: Getty Images/Keystone

  Robert Burns Woodward, 1917–1979

  Readers living in a time when technologies like nuclear magnetic resonance imaging are routinely able to determine molecular structure at a relatively low cost in a single afternoon may find it difficult to understand the value placed on Woodward’s talent during the 1940s and 1950s. Instead of simply examining a three-dimensional picture of a complicated organic molecule on a screen, chemists of his era could only derive structure by working as enormously sophisticated puzzle solvers: taking all the known facts about a molecule, such as whether, and how quickly, it reacted to heat or cold, to acids or bases, or to other molecules; and from that information, and a detailed knowledge of the laws of chemistry, figuring out which atoms connected to one another, through what sort of bonds, and in what configuration. This is a little like drawing the blueprint of an office building knowing only its floor-by-floor heating bills and the number of people who used its elevators daily.

  At this, at the manipulation of what chemistry students know as stereoisomers—alternative spatial configurations of three-dimensional molecules—Woodward was unmatched, not merely because of what might be called his architectural eye, but his profound understanding of the underlying physics of the molecules in question. So, when John McKeen went looking for the chemist best equipped to aid Pfizer in understanding Terramycin, he didn’t have to look far.

  Nor did he have to do much persuading. Any qualms about working for a commercial employer, which had earlier stymied A. N. Richards during his imbroglio with the American Society of Pharmacology, had disappeared, seemingly overnight. Woodward wasn’t ever going to be an industrial chemist per se, but he was more than happy to consult with industry. (One of his best friends was Edwin Land, the founder of Polaroid, who started paying Woodward as a consultant as early as 1942, and eventually made him the only nonemployee to be given options on Polaroid stock, and thereby made him a very wealthy man.) Pfizer had a puzzle that needed solving, and Woodward, who famously completed the New York Times crossword daily, needed puzzles to solve.

  After dozens of other chemists had tried and failed to figure out the compound’s molecular blueprint, Woodward, in legend anyway, took a large piece of cardboard, wrote on it every fact known about the compound, and “by thought alone, deduced the correct structure for Terramycin.”

  What he reported might have seemed, at first, problematic. Pfizer’s new drug was, indeed, not just functionally similar to Aureomycin. It was structurally similar as well. Both compounds were built around a four-ring structure, which gave them the generic name “tetracyclines.” But Aureomycin had a single chlorine atom—generically chlortetracycline—that Terramycin lacked. Meanwhile, Terramycin (or oxytetracycline) had an oxygen atom that was missing from Aureomycin. From a medical standpoint, the differences were negligible. But as intellectual property, the discovery was huge. Terramycin was novel enough to be patented. Pfizer filed a patent for Terramycin in November 1949; the FDA approved it five months later. Pfizer was ready to start its production line.

  And this time, they wouldn’t just be producing a drug for other companies. They decided to sell it themselves.

  This was, to put the nicest possible construction on it, ambitious. In 1950, Pfizer’s sales force, including the sales manager, numbered only eight people. Undaunted, McKeen sent telegrams to eight hundred drug wholesalers—essentially, every wholesaler in the United States—announcing Terramycin’s availability as soon as it was cleared for sale by the FDA. And, on March 23, 1950, within an hour of receiving the formal approval, each of Pfizer’s eight sales representatives was manning a switchboard, telephoning a hundred wholesalers each, offering a very steep introductory discount.

  Within a year, Pfizer was employing a hundred salesmen. In 1951, the company supplemented them with seventy third-year medical students, hired to work for the summer, and sent them to forty cities to sing the praises of Terramycin. By 1952, three hundred Pfizer reps were selling Terramycin as fast as the company could manufacture it.

  Pfizer’s efforts—and success—were matched by the tactics Lederle was simultaneously employing with Aureomycin. In the battle for America’s antibiotic dollar, featuring ever more elaborate sampling campaigns, twenty-four-hour telephone blitzes (and competition to see which sales reps could drain their respective employer’s travel and entertainment budget faster, by entertaining more lavishly), Pfizer and Lederle both emerged victorious, equipped with an arsenal of anti-infective drugs that exhibited a new, and far wider, range of effectiveness.

  The biggest reason was that the broad-spectrum antibiotics really were new and improved. Penicillin works by weakening the molecules that form the bacterial walls of Gram-positive pathogens; streptomycin disrupts the way that bacteria make protein, though with some nasty side effects, including kidney damage and deafness. Tetracyclines also fight pathogens by inhibiting protein synthesis, but far more effectively: Both Aureomycin and Terramycin hijack the system that bacteria use to accumulate needed molecules from their environment, and so concentrate the bactericidal molecules precisely where they can do the most good, and fast; tetracyclines can accumulate in concentrations more than fifty times greater inside a bacterium than outside.* This made the new drugs effective against virtually every sort of pathogenic bacterium, from the spirochetes that cause syphilis to the bacilli responsible for anthrax and bubonic plague. Moreover, because protein synthesis is such a universal requirement of life, the tetracyclines were also useful against the pathogens that cause malaria—not bacteria, but protozoans, single-celled organisms with nuclei, which evolved billions of years after the first bacteria appeared.

  Largely because they generated greater revenues, which supported ever more aggressive marketing, broad-spectrum antibiotics completely overtook penicillin and streptomycin in sales.* By 1952, Americans were spending more than $100 million annually for broad-spectrum antibiotics, more than three times as much as they were spending on penicillin. The profit margins for the former ranged from 35 to 50 percent, while the profit on penicillin and streptomycin barely topped 5 percent. Pfizer accounted for 26 percent of all antibiotic sales; Lederle a bit more than 23 percent. And the pie was growing larger every year. By the early 1950s, more money was being spent on antibiotics than on all the new and improved patent medicines, toothpastes, mouthwashes, vitamins, hormones, botanicals, and even sulfanilamides, combined.

  It took some time before the complicated effects of broad-spectrum antibiotics like the tetracyclines were understood; in some respects, they’re not fully grasped even today. At their introduction, the results seemed simple enough: They were true wonder drugs. Unfortunately, though Aureomycin and Terramycin were distinctive enough to have “exclusive” patents, they weren’t truly different in any other important way. By late 1952, in fact, Woodward had demonstrated that the molecule that was doing all the heavy lifting for Terramycin wasn’t oxytetracycline, but simple tetracycline; inferentially, this meant that Aureomycin, or chlortetracycline, didn’t need its chlorine atom, either. It could have been Prontosil all over again: a discovery that the active ingredient in a branded medication was far simpler than the one described in the patent that prot
ected it from competition.

  The difference was that sulfanilamide had been discovered so long before it revealed its antibacterial properties that Bayer couldn’t protect it as a piece of intellectual property. Simple tetracycline, on the other hand, was new. In October 1952, the Pfizer team filed a patent application for the four-ring molecule at the heart of Terramycin: tetracycline itself. A few months later, in March 1953, Lederle’s parent, American Cyanamid, filed its own application. And, just to keep things interesting, in September both Bristol Laboratories and Heyden Chemical—members of the penicillin consortium that had independently discovered how to make tetracycline without starting from either chlortetracycline or oxytetracycline—filed patent applications both for the tetracycline molecule and for their own distinctive methods of producing it.

  This complicated aspect of the patent system—two molecules that differed from one another by the placement of a single atom could be separately patented, even if the atom was unnecessary to the therapeutic activity of the molecule; two different processes by which identical molecules were produced could, likewise, receive separate patent protections—made for a monumental mess. Though Pfizer was attempting to secure patent protection for simple tetracycline, they had no method for making it that didn’t require starting with the chlortetracycline produced by S. aureofaciens. This meant that producing a Pfizer-branded tetracycline—they named it Tetracyn—could only be done by licensing (or buying) Aureomycin from Lederle. Lederle, on the other hand, could produce tetracycline only by getting a process license from Bristol or Heyden before they could sell their version of plain vanilla tetracycline, which they called Achromycin. Since Lederle planned to launch Achromycin with a $2.5 million campaign—$1 million for samples alone—that included promoting it at two hundred separate professional meetings and sending more than a hundred mailings to every physician (and seven more to every dentist) in the country, some resolution of the license impasse was urgently needed.