The Discovery of Insulin

by Michael Bliss

  V

  The rediscovery of a way to make insulin made it possible to consider resuming clinical tests. Banting, as we have seen, had played little part in the clinical work at Toronto General Hospital, for he had been denied an appointment to the hospital’s staff. In February and March it had seemed to Banting as though he had no further role to play in the development of the discovery. As he pulled himself together that spring, however, probably relying heavily on such friends as Velyien Henderson for advice, Banting must have realized that he had an unchallengeable claim to use the extract. A very large number of people, including Banting himself, believed that he had discovered it. It would be unthinkable to deny Dr. Frederick Banting, a licensed physician in good standing, priority in the clinical use of insulin. If Banting did not get his way the amount of trouble and bad publicity he and his friends could cause was practically unlimited.

  The fact that Duncan Graham would not give Banting an appointment at Toronto General Hospital was not the barrier it had first seemed. Why should TGH have a monopoly of the clinical tests of insulin simply by virtue of being the university’s chief teaching hospital? In the spring of 1922, Dr. F.G. Banting established an office at 160 Bloor Street West in Toronto and began the private practice of medicine. This one step instantly gave him the right to use the facilities of TGH’s private patients’ pavilion for his private patients. Then, early in April, Banting was interviewed about the discovery by the Director of Medical Services for the Canadian Department of Soldiers Civil Re-Establishment, which handled the affairs of war veterans; several weeks later Banting was appointed head of a new diabetes clinic at Toronto’s Christie Street Military Hospital, where he had worked briefly in 1918–1919. Now he had all the facilities he needed.19 About the same time, an agreement was reached with the Connaught Laboratories on the distribution of insulin for clinical use. One-third of the production was to go to Banting for his private practice, one-third was to be used in Banting’s Christie Street clinic, and one-third would be available for work at Toronto General and the Hospital for Sick Children.20 Macleod began referring all the inquiries he received from diabetics to Dr. Banting, “my clinical associate.”

  By mid-May enough insulin was being produced by the new method to permit resumption of limited clinical testing. Dr. Joe Gilchrist received his second injection on May 15. Gilchrist had agreed to work at the Christie Street clinic under Banting, and so served as both physician and patient. In the early months of sporadic production and frequent impurities, Gilchrist became Toronto’s self-proclaimed “human rabbit,” testing each new batch on himself after it had been tried on the rabbits.21

  There was also enough insulin in mid-May to allow Banting to meet the urgent request of Dr. John R. Williams, who had come to Toronto from Rochester, New York, some miles away on the other side of Lake Ontario, to see if he could get some insulin to try on his most desperately ill patient. Jim Havens, son of a vice-president of Eastman Kodak, had been diagnosed as diabetic seven years earlier at age fifteen. He did fairly well on an Allen diet until 1920 when his capacity began a sharp decline. The boy was treated by Allen, and his father supported Allen’s “heroic efforts” to get research going at the Physiatric Institute. By early 1921, however, James Havens, Sr., had given up hope that Allen or anyone could help young Jim.22 When news came of the discovery in Toronto a year later, James Havens, Jr., was a 73½-pound skeleton, living on 820 calories a day, barely able to lift his head from his pillow, crying most of the time from pain, hunger, and despair. According to Williams, he was

  a most pitiable spectacle. Blood sugar 450 mgs. Plasma bicarbonate [a measure of acidosis] 24.9 volumes per cent. For weeks the patient had suffered severely from pains in his legs, which made the constant use of codeine necessary. The edema and profound weakness confined him to bed and he was rapidly approaching death, when through the great kindness of Doctors Banting and Macleod, extract was supplied for his treatment.23

  Havens got his first insulin on the evening of May 21, 1922. He was the first person treated with it in the United States.24 The first injections of one or two cubic centimetres (throughout this period one rabbit “unit” of insulin, roughly defined as the amount necessary to send a rabbit into hypoglycemic convulsions, was usually about one cubic centimetre in volume) were very painful and had no effect, confirming, it is said, Williams’ hesitancy about trying the new cure in the first place.25 On May 26 Banting went to Rochester to examine Havens. He advised doubling and then tripling the dosage. Within a day or so Havens’ urine was sugar-free, his blood sugar was down to normal, and his clinical condition greatly improved. Banting agreed to have fresh supplies of insulin sent by train from Toronto.* Two weeks after first receiving insulin, Jim Havens was able to rise from his bed and walk. “The patient is very much better,” Williams wrote Banting. “His appearance indicates much greater improvement than the laboratory studies suggest. Dr. Joslin made the statement to me one time, that of all the ways of measuring the condition of a diabetic, he thought the clinical appearance to be one of the best and that is my experience…The greatest advance noted is in his state of mind.”27

  Williams had come to Toronto personally to plead for insulin. Others, alerted by the May 3 paper in Washington, were beginning to do the same or to write Banting or Macleod asking when the new treatment would be available. “I have some really heart-breaking cases under my care at the present time,” the chief pediatrician from the Johns Hopkins Hospital in Baltimore wrote in a typical appeal, “two of them lone children of different families whose carbohydrate tolerance is gradually going down. They know of your work and are pestering me to get some of the material if I can. I do not wish to pester you, but only to let you know how anxious I am to use some of the ‘insulin’ if I can get it.”28 But Toronto had no extra insulin to give to him or anyone else. For several weeks Jim Havens was the only diabetic outside of Toronto who was being treated with insulin.

  VI

  The pressure to produce more insulin was mounting daily. All attempts to produce the hormone in large quantities continued to fail. Toronto’s recovery of a method in mid-May was accompanied by a realization that the group had to have help. The ubiquitous George Clowes had spoken to Macleod again at the Washington meeting at the beginning of May, had written offering advice on the American patent situation, and continued to urge Toronto to collaborate with his firm. The Torontonians finally came around and invited Clowes to come to Toronto on May 22. (They also invited Rollin Woodyatt of Chicago, who was offering an informal collaboration that would put his expert staff, financial backing, and the pancreas resources of the Chicago stockyards at Toronto’s disposal. Woodyatt, however, was unable to come to Toronto at the time specified.)29 Clowes brought with him a chemist, a patent attorney, and the vice-president of Eli Lilly and Company, Mr. Eli Lilly. In two or three days of meetings at the King Edward Hotel, the Americans and Canadians worked out an agreement for the development of insulin.

  The proposed collaboration was explained in another formal letter, written on May 25 to Falconer from the research team plus Fitzgerald of the Connaught. They now recommended that the University of Toronto Board of Governors accept from Collip and Best a United States patent on the process, for which they were applying. The Board would eventually license North American firms making insulin for sale and collect royalties from them to support research in the university. For now, however, the group recommended that a temporary exclusive licence be given to Eli Lilly and Company of Indianapolis. The explanation of the recommendation was as follows:

  Experience on production of “insulin” on a moderately large scale in the Connaught Laboratories has shown that this is fraught with many difficulties not encountered in the small laboratory scale, and we do not believe that production in amounts that are adequate to supply the demands for it can be accomplished without further experimentation in its preparation on a much larger scale than is possible here. To make this further step possible it will be necessary for
us to collaborate with some well equipped and properly staffed commercial house engaged in this work. After careful consideration we have decided that it is much better to arrange to deal with one firm rather than several, partly because concentrated effort is likely to be more efficient than divided effort, and partly because we could not act as consultants to several establishments at the same time….

  We have chosen to collaborate with [Eli Lilly and Company]. We recognize this firm will be placed at an advantage over its competitors through this collaboration with us, but we believe that it is much less serious than there should be further delay in proceeding intensively with production on a large scale, and, moreover, we propose to give other firms, as well as hospitals and other non-commercial concerns, every chance to do the best they can by publishing the details of the method as at present used by us in the Connaught Laboratories in full at an early date (within three months). By this step, our proposed co-operation with the Lilly Co. cannot be criticized as unethical or unfair or as in any way prejudicial to the free manufacture of “insulin.”30

  Toronto’s decision to collaborate with his company was a triumph for and testimony to the persistence of G.H.A. Clowes. It also reflected a vote of confidence in both Clowes and his company by the Toronto researchers, who had not acted impulsively or without consideration of their several alternatives.

  In 1922 Eli Lilly and Company had been making and selling pharmaceuticals for forty-six years from their base in Indianapolis, Indiana. It was a family-owned “ethical” drug company (no patent medicines, no extravagant claims, and advertising and sales to doctors and pharmacists only), which had grown to become a major, though not dominant factor in the industry. In 1921 Lilly employed about eleven hundred people and did just over $5 million worth of business. The firm was managed, according to the founder’s son and president, J.K. Lilly (whose own son, Eli, had been the family man in Toronto), with the aim of being “conservatively progressive.” Part of the house’s progressiveness in the early 1900s had been the creation of a substantial research facility. At the end of the First World War the Lilly family had decided to strengthen further the firm’s links with the scientific community, even though the short-term returns from such ventures might be minimal. As part of this continuing policy G.H.A. Clowes was hired as a special research chemist in 1919 and appointed director of research the next year.

  George Clowes (pronounced clews) was an Englishman in his mid-forties, a minister’s son who had taken a Ph.D. in chemistry in Germany, done post-doctoral studies in England and France, and then emigrated to America as a land of greater opportunities than Britain. Before moving to Lilly, Clowes had spent many years at a state research institute in Buffalo, New York, ninety miles from Toronto. He was a serious researcher in his own right, most interested in problems relating to cancer, but with a restless curiosity about all kinds of knowledge. His job with Lilly was almost unprecedented in a commercial organization, Clowes wrote, in giving him virtually free reign. During the summers, for example, he worked at the Marine Biological Laboratories at Woods Hole, Massachusetts, where he had his own lab, assistants, and no commençai responsibilities. His research reports emphasized the absolute necessity of continuing to strengthen the firm’s links with the scientific community, including the universities, and raising its prestige and authority among researchers. Few broad corporate strategies have ever paid off as quickly and magnificently as this did for Eli Lilly and Company.31

  Macleod had known Clowes for some years, was impressed by his stature as a scientist, and by his company’s enlightened support of research. He and the other Torontonians were probably also impressed by the plans Clowes outlined to them for the development of insulin. The firm had recently been very active in work on glandular products, and had a good team of chemists ready to work on insulin. It wanted an exclusive licence for an “experimental period” of one year, during which there would be a complete pooling of knowledge between Toronto and Indianapolis. There would be a several-stage development of the product involving large-scale clinical tests in Toronto and the United States, with Lilly supplying extract free of charge in the initial stages and then selling it at cost. Lilly would share any improvements it made in the manufacturing process with Toronto, and if any improvements were patentable would pool the patent rights for all territory outside of the United States. At the end of the experimental period, Lilly wanted a licence to manufacture insulin on the same terms as Toronto would license other manufacturers. As Clowes had proposed in earlier letters, the firm thought it would be appropriate for insulin licensees to pay Toronto royalties on all insulin sold.

  The collaboration was formally established in an “Indenture,” dated May 30, 1922, between the Board of Governors of the University of Toronto and Eli Lilly and Company. It was intended to be a close, but not necessarily exclusive, relationship. The Lilly company was prohibited from divulging details of the process to other parties, for example, but Toronto was not. As they had told Falconer, the Toronto team still intended to publish their method in order to make sure others would know how to make insulin when Lilly’s exclusive rights expired and to protect themselves from charges of unethical secrecy. As well, the Lilly agreement limited the company’s territorial rights to the United States, Central and South America. To handle insulin in Britain and the rest of the Empire, perhaps Europe too, the Torontonians had decided to offer the patent rights to the British Medical Research Council, for administration in a way parallel to the University of Toronto’s handling of the Americas. At the end of May, Macleod wrote the Medical Research Council conveying the offer. On its part, the Board of Governors appointed a small committee to work with the discovery group in carrying out the licensing and development arrangements. These bodies soon evolved into Toronto’s Insulin Committee.32

  The Toronto group was anxious to get policies for developing insulin in place during May, not only because of the demand from doctors and diabetics, but also because of the imminent break-up of the group. Collip’s appointment at Toronto expired on May 31. Such negotiations as there may have been about his staying on seem to have dissolved in the quarrels with Banting and then the difficulties making insulin. Whether or not Collip wanted to stay on but was not wanted, or was wanted but was fed up with the fighting, is not known.

  At the end of May the Toronto researchers read six short papers on their work at a session of the annual meeting of the Royal Society of Canada. The first two of these, scheduled to be published several months later, contained methods for making insulin as developed by Banting, Best, and Collip. There were two recipes – “The Preparation of the Earlier Extracts,” by Banting and Best, and “The Preparation of the Extracts as used in the first Clinical Cases,” by Collip – an indication for the record of who had done what by members of a team that had fallen apart.33 Best and Collip then travelled to Indianapolis. On June 2 and 3 they told the Lilly chemists all they knew about making insulin and helped with the first attempt to extract it. The process worked.34 His time in Toronto over, J.B. Collip went back to his job at the University of Alberta.

  VII

  Beginning work immediately, the Lilly company poured men and money into insulin production. But they were not the first to make insulin in the United States. Dr. W.D. Sansum of the Potter Metabolic Clinic in Santa Barbara, California, had noticed Banting and Best’s first publication and in April had written Banting to ask about progress. When Banting told him of the delays, Sansum decided to try making pancreatic extract himself. He and his associates tried various methods. As soon as they learned to use alcohol as an extractive and normal rabbits for testing (from Banting and Best’s May article, combined with a letter Joslin had published immediately after the May 3 meeting), the Potter group found they could make potent extracts. On May 31 they began administering insulin to an adult male patient, and soon succeeded in making him sugar free. They tried to increase their supply of the extract early in June, collecting the pancreases from sixteen hundred sheep.
Just as had happened in Toronto, they found that the attempt to scale up production failed completely. Macleod learned of the California work in mid-June when Sansum wrote to him asking for advice.35

  Macleod had expected some such development. It was only through professional courtesy that Rollin Woodyatt was delaying trying to make extracts and other diabetologists, including Allen, were starting to become impatient. They all had dying diabetics on their hands, patients they had encouraged to carry on in the faint hope that some treatment would be discovered. Now the announcement of the treatment had come out of Toronto, but no treatment. Some of the patients’ life expectancy was a matter of weeks.36

  To meet the clinicians’ demand, while at the same time usefully spreading out the research job, Toronto and Lilly had agreed that a select group of physicians and institutions would be given the extract for testing purposes as soon as it became available. Until then, the Torontonians saw no reason why other researchers should not be able to make insulin. Macleod sent both Sansum and Woodyatt details of the method. To honour the Lilly agreement, he required them not to divulge the method to anyone likely to produce the extract commercially.37