Authors: Richard Kluger
By the end of 1953, filters had claimed only 3 percent of the cigarette market, and American Tobacco’s Hahn nonchalantly told the press, “If we can find a good filter—a really good one that doesn’t take all the taste and flavor out of the cigarette—we’ll have a serious look at it.” But as AT’s two closest runners-up for market share pushed actively into the filter arena in 1954 and little Brown & Williamson, on the strength of its hot Viceroy, moved past Philip Morris into the No. 4 slot in the industry, the filter market share reached 10 percent by the end of that year. To hedge his bet ever so slightly, Hahn ordered a filtered version of his king-size Herbert Tareyton brought out; its unfiltered version, though, outsold it heavily. American Tobacco commanded one-third of the cigarette market, while RJR, even with its new Winston off and running, held no more than a quarter of the market. Hahn saw no point in dashing into the increasingly crowded filter market. It was a major blunder.
VI
AT
the annual clinical congress of the American College of Surgeons, held in October of 1953 and attended by 11,000 of the nation’s most eminent sawbones, heightening suspicion was voiced about the possible deleterious effects of smoking. Among those most concerned was a University of Pennsylvania surgical professor named L. S. Raydin, one of the organizers of the convention program, who used the platform to assert that “in a matter which involves public health to such a degree, the tobacco industry has a moral obligation to pay for the research necessary to prove or disprove the suspected relationship … . Who else should be more interested?”
If the cigarette makers scoffed privately at the notion that they should underwrite serious studies of the health issue which might prove their undoing, a report appearing in the journal
Cancer Research
not long after the surgeons’ 1953 conclave forced tobacco manufacturers to rethink the proposal. The lead investigator was once again Dr. Ernst Wynder, the young epidemiologist whose 1950 retrospective study with St. Louis master surgeon Evarts Graham had initiated the new age of scientific scrutiny of smoking. Wynder, by now the one-man staff of the new division of clinical investigation at the Sloan-Kettering Institute of New York’s Memorial Hospital for Cancer, took the question to a level beyond the statistical association of smoking and disease with a two-year experiment on the susceptibility of mice to the growth of cancerous tumors on their backs after being painted with a distillate of cigarette tobacco smoke. The reported results were not good news for the industry.
Since a definitive experiment with human subjects was ethically impermissible, the next best thing would have been to try to replicate man’s smoking with large animals of comparable anatomy. But large animals were hard to handle and expensive to maintain, and their use would have required long years to measure the effects of slowly developing cancer, even assuming the subjects could be taught to puff and inhale like humans. The problem with inhalation experiments was compounded by the fact that most species in the animal kingdom, and especially small creatures, are, unlike man, obligatory nose-breathers, with a highly articulated filter system built in to block the passage of foreign matter or to absorb it, thus preventing whole smoke from ever reaching their lungs in a quantity remotely approximating the human experience. Heavy doses of smoke would be needed to compensate, but since oxygen-robbing carbon monoxide makes up about 3 percent of tobacco smoke, heavy repeated doses were found to asphyxiate small animals early in the experimental procedure or otherwise induce fatal diseases. Some animals buried their snouts in their own fur or that of their cage-mates to avoid the smoke
bath, thereby limiting their intake and foiling the point of the exercise. The most practical method, therefore, was to swab the shaved backs of mice—the preferred experimental model because they were small, short-lived, and comparable enough anatomically to humans—with tobacco tars derived from whole smoke and see how many developed tumors. While the results could not be readily extrapolated to humans, they might indicate to investigators if they were on the right track—“to give you some insight into the mechanisms of how cancers occurred in people,” as Wynder later put it.
Perhaps the most questionable element in the mouse-skin experiments was not the obvious differences between animal and human species or between the chemical constituents of whole inhaled fresh smoke and a tarry distillate of it, but the nature of the cells to which the tar was applied. Wynder of course understood that the cells composing the tough epidermis of a mouse’s back skin were different from the complex, thinly layered, and more sensitive epithelial cells lining the lungs. But Wynder would later contend that in the unfolding stages of the cancer-forming process, the lung epithelium under assault by tobacco smoke first underwent hyperplasia and metaplasia, thickening phases in which the cells revert to a more primitive form in a kind of protective reaction against the repeated insults by the irritants in the smoke; in that altered state, said Wynder, the lung cells did in fact resemble those of the mouse skin, thus allowing the cancer-forming process to be legitimately analogized.
Wynder’s experiment, performed in league with Adele Croninger, a coworker from Toronto who handled the animals, and Dr. Graham, the reformed smoker who continued to lend his prestige to the research of the nervy Wynder, used a tarry condensate derived from fifty-carton batches of Lucky Strike, smoked by specially designed rhythmic suction machines attached to a series of glass beakers and approximating the human smoking cycle. The dark brown viscous fluid that resulted after condensation was mixed, in order to avoid inducing toxic shock when applied to the mice, one part to three of acetone, a naturally occurring, noncarcinogenic dilutant; after a while, the solution was changed to equal parts of tar and acetone. The subjects were 156 mice purchased from Clarence Little’s Jackson Memorial Laboratory in Maine, 86 of which had their backs electrically shaved from the nape of the neck to tail and painted three times a week with a saturated camel’s-hair brush. The dose was 40 milligrams of the tar distillate, the rough equivalent of the tar and nicotine yield of a pack of cigarettes.
Of the 62 mice in the painted group surviving after a year into the experiment, 58 percent developed tumors proven to be cancerous upon microscopic examination; 44 percent of the overall painted group got cancerous tumors, suggesting that some animals had died from other causes before the carcinogenic effects might have taken hold. Only 10 percent of the painted animals survived the full twenty months of the experiment, compared with 58 percent
of the unpainted controls. As Wynder reported in
Cancer Research
, the “suspected human carcinogen has thus been proven to be a carcinogen for a laboratory animal”—an outcome that provided “a working tool which may enable us to identify and isolate the carcinogenic agent(s) within the tars.” Dr. Graham, in remarks to a news magazine after the results were published, was less cautious; the experiment, he said, “shows conclusively that there is something in cigarette smoke which can produce cancer Our experiments have proven it beyond a doubt.” Such certitude was spoofed in
Forbes
magazine, which noted, “Graham gave no estimate of the number of smokers who distill the tar from their tobacco and paint it on their backs in concentrated form.”
More substantive skeptics of the skin-painting technique would remain unpersuaded by Wynder’s contention that the mouse epidermal cells exposed to the tar were comparable to the bronchial cells after undergoing metaplasia. The more fervent criticism was that the dosages applied to the mice were so massive, in order to obtain speedy results, that the organism’s defense mechanisms naturally collapsed. But defenders of the skin-painting said it was far from clear that the mice had been overdosed to produce cancers, since in human smoking the surface of lung tissue exposed to the small but continuous assaults by carcinogenic irritants was exponentially greater, due to the highly ramified structure of the bronchi, than the back area on the mice from which the tar could be readily sloughed off—in short, the whole smoke was much more efficiently absorbed by the human lung cells than the tar could be by the mouse epidermal cells. Tobacco industry defenders would further claim a few years after the wynder mouse-skin report that 25 percent of all known chemical compounds, including simple sugars occurring in nature, caused some type of carcinogenic reaction in certain laboratory animals. For decades, industry scientists argued that while the tarry condensate was prepared by collecting the smoke in a freezing cold trap, then extracting it with acetone and concentrating the extract by evaporation before applying it to the back of the test mouse, all kinds of chemical reactions were taking place among the smoke components and confounding factors that further undercut the validity of the skin-painting study.
Yet for all of these objections, cigarette manufacturers and government and private laboratories would long afterward persist in using the skin-painting technique, because however flawed it might have been, however imperfect the analogy to human smoking, and however inconclusive, no more reliable form of bioassay has ever been developed. Wynder’s experiment may not have indicted cigarette smoke as a killer, but it surely served to place the substance at the scene of the crime. Publication of his findings at the end of 1953 was enough to persuade the American Medical Association to stop accepting cigarette ads in its main scientific journal,
JAMA
, and nine smaller medical magazines.
VII
THE
wide media attention paid to Wynder’s graphic findings in the mouse-skin experiment stirred a petulant response by the tobacco industry, so long coddled in the news and editorial columns of the press, whose owners had grown flush with revenues from cigarette advertising. Paul Hahn, head of the industry pacesetter, American Tobacco, offered a rare public statement, lamenting “much loose talk on the subject [of smoking and health] in the press during recent months” and contending that nothing had been proven either statistically or in laboratories. A company statement added that “[u]nwarranted attacks on tobacco products are as old as the industry itself … .”
Such a reflexive denial and the counterargument that a suspected major menace to the public health must remain at liberty until proven guilty were no doubt understandable. But in their facile reassurance to a customer base composed of nearly half the nation’s adult population, industry officials displayed insensitivity to the public’s dawning comprehension of the enormity of the issue. Their dismissive tone was even more pointedly sounded by Reynolds’s president, Edward Darr, who revealed his own and his company’s combativeness with the remark, “One of the best ways of getting publicity is for a doctor to make some startling claim relative to people’s health regardless of whether such a statement is based on fact or theory.” Here was a blatant instance of the pot calling the kettle black; it ill behooved a company lately chastised by the Federal Trade Commission for fraudulent health claims in its advertising to malign the studies of highly qualified scientific investigators.
Behind the bravado, the cigarette barons had become increasingly uneasy if not yet panicky. Their mood was suggested by a memorandum circulated among them on December 14, 1953, by Brown & Williamson’s president, Timothy V. Hartnett, a veteran of more than forty years in the tobacco business. Hartnett was of the opinion that the investigators on the smoking and health issue included men of “unquestioned integrity and sincere in their belief,” so he urged “[e]xcessive caution … in the methods we adopt to counteract these claims.” With remarkable candor he added that it was important for the industry leaders to recognize “we may be embarrassed by page after page of
pseudo-scientific
selling over the years. As one manufacturer says, we’ve perhaps too often been unlicensed medicine men rather than honest tobacco men” (italics in original text). Hartnett then called for a two-pronged counterattack “to get the industry out of this hole”: (1) “unstinted assistance to scientific research,” the most difficult part of this effort being “how to handle significantly negative research results if, as and when they develop,” and (2) “the best obtainable” public-relations counsel since none “has ever been
handed so real and yet so
delicate
a multimillon dollar problem” (italics in original).
Hartnett’s candid concern was heeded. Shortly thereafter, the tobacco industry retained the services of Hill & Knowlton, among the most highly regarded U.S. public-relations firms. In a December 24, 1953, memorandum to the cigarette makers’ chieftains, H&K offered sobering counsel: “It is important that the industry do nothing to appear in the light of being callous to considerations of health or belittling medical research which goes against cigarettes.” Moreover, the tobacco moguls ought to understand that their past advertising “may have created a degree of skepticism in the public mind” that could initially hamper a public-relations effort to mitigate the health scare. In endorsing an industrywide research initiative, the New York-based publicists called for the formation of a scientific advisory board of men “whose integrity is beyond question.”
Moving with a dispatch that reflected concern over a more than 3 percent decline in cigarette sales for 1953, fourteen leading tobacco manufacturers and allied groups among the growers and warehousemen opened the new year with a full-page announcement in 448 newspapers of the establishment of the Tobacco Industry Research Committee. Headlined “A Frank Statement to Cigarette Smokers,” the ad’s text began, “Recent reports on experiments with mice have given wide publicity to a theory that cigarette smoking is in some way linked with lung cancer in human beings,” and while not regarded as conclusive, the results ought not to be “lightly dismissed”—a decided change of tune by the industry within a few weeks of the Wynder revelations. After this opening concession, the ad dispensed a torrent of reassurances: Research had shown there were many likely causes of lung cancer, there was no proof that smoking was one, and—in an unjustifiable claim—the biostatistics so far produced “could apply with equal force to any one of many other aspects of modern life. We believe the products we make are not injurious to health,” the signers of the ad continued—a statement of creed, not a warranty, they would later contend, suggesting that the text had been meticulously lawyered. And then, revealing the fine hand of master publicists, the “frank statement” sought to blur any distinction between earlier denunciations of smoking without empirical evidence and the troubling new scientific data. For more than 300 years, tobacco “has given solace, relaxation and enjoyment to mankind” even while being held “responsible for practically every disease of the human body. One by one these charges have been abandoned for lack of evidence … .” Nevertheless, the industry was undertaking joint financial aid to research efforts on the smoking and health question and would soon appoint “a scientist of unimpeachable integrity and national repute” to oversee the program.