Leonard Thompson and the First Sustained Insulin Treatment, 1922

Physicians had long recognized more than one clinical pattern under the name diabetes mellitus, but they did not yet use the modern type 1 and type 2 classification. For children and young people with severe, ketosis-prone disease, the most effective pre-insulin regimens used fasting and extreme restriction of carbohydrate and total calories. Such diets could reduce sugar and ketones and sometimes prolong life, but at the cost of hunger, emaciation, and no dependable long-term escape from the disease. Thompson weighed about 65 pounds and was receiving roughly 450 calories a day during his admission (University of Toronto Libraries).

The pancreatic hypothesis was also older than the Toronto work. In 1889, Oskar Minkowski and Joseph von Mering produced diabetes by removing dogs' pancreases. Researchers including Ernest Scott, Israel Kleiner, Nicolae Paulescu, and Zülzer later obtained antidiabetic effects from pancreatic preparations. Zülzer injected eight patients between 1906 and 1908; some showed reduced urinary sugar and ketones, but fever, vomiting, and other toxic effects prevented a usable continuing therapy. Paulescu's 1921 publications documented effects in pancreatectomized dogs, not successful clinical treatment (de Leiva-Hidalgo and de Leiva-Pérez).

The difference in Toronto was therefore not the sudden conception of a pancreatic “internal secretion.” It was the combination of laboratory physiology, repeated blood-sugar measurement, biochemical purification, hospital observation, and then a manufacturing system capable of supplying material of measured potency. That is why “first injection” and “discovery of insulin” are both too simple as descriptions of a process (Rostène and De Meyts).

1. 1889: at Strasbourg, Oskar Minkowski and Joseph von Mering show that complete removal of the pancreas produces severe diabetes in dogs, strengthening the case for a pancreatic role in sugar metabolism (Rostène and De Meyts).
2. 1906-1908: in Berlin, Georg Zülzer administers his pancreatic extract to eight patients. Some biochemical signs improve, but toxicity and inconsistent supply prevent sustained treatment (de Leiva-Hidalgo and de Leiva-Pérez).
3. 1921: in J.J.R. Macleod's University of Toronto physiology department, Banting and Best prepare extracts that lower blood sugar in pancreatectomized dogs. Macleod contributes experimental direction and physiological expertise; Collip joins late in the year to work on purification (Rostène and De Meyts).
4. December 1921: Thompson is admitted to Ward H at Toronto General Hospital. His chart identifies him as a thirteen-year-old with diabetes; the published report later describes a “boy aged 14 years” (Thompson patient record; Banting et al.).
5. 11 January 1922: Jeffrey gives the first formal human injection in the Toronto programme. The biochemical effect is limited and an abscess follows, so the clinicians discontinue that preparation (Bliss).
6. 23 January-4 February: Campbell administers Collip's improved extract to Thompson in a continuing course. Sugar and ketones fall, clinical condition improves, and the effect can be observed repeatedly (Banting et al.).
7. March 1922: Banting, Best, Collip, Campbell, and Fletcher publish seven hospital cases in the Canadian Medical Association Journal. Their preliminary report supplies contemporary clinical data, but its confident conclusion also served the authors' claim that the extract had established therapeutic value (Banting et al.).
8. May-August 1922: production shortages push the university and Connaught Laboratories toward collaboration with Eli Lilly. Connaught staff develop scalable methods, and a Toronto General diabetes clinic opens in August; access nevertheless remains scarce and controlled (Rutty; University of Toronto Libraries).
9. 1923: wider production and biological standardization make treatment available beyond a few experimental centres. The Nobel Prize goes to Banting and Macleod; Banting shares his award money with Best, and Macleod with Collip (Nobel Prize; Rostène and De Meyts).

Banting proposed a line of pancreatic research and worked with Best on the 1921 animal preparations. Macleod provided a laboratory, animals, methods, critical supervision, and later a larger research programme. Collip's biochemistry made the January clinical response convincing. At Toronto General, Duncan Graham authorized the hospital work; Campbell and Fletcher selected and managed patients; Jeffrey administered the first injection. Thompson and the other patients bore the risks and made clinical proof possible (Banting et al.; Bliss).

The authorship of the March paper—Banting, Best, Collip, Campbell, and Fletcher—captures more of this collective labour than the durable “Banting and Best” story, but it omits Macleod, who did not sign that paper. Later personal rivalries and commemorative retellings made priority an argument about names, even though laboratory preparation, clinical trial, purification, physiological interpretation, and manufacture were different achievements (Rostène and De Meyts).

The 1923 Nobel decision recognized Banting and Macleod “for the discovery of insulin.” The subsequent sharing of prize money acknowledged Best and Collip but did not settle the historical argument. Nor did the award erase earlier work by Zülzer, Paulescu, Kleiner, Scott, and others. The sound claim is not that Toronto invented the pancreatic hypothesis from nothing; it is that its linked laboratory and clinical teams produced the first repeatably effective treatment and a route toward supply (de Leiva-Hidalgo and de Leiva-Pérez; Nobel Prize).

Collip's method produced enough material for early trials but did not scale smoothly. When stocks ran short in spring 1922, critically ill people sought access to a treatment that Toronto could supply to only a few. Connaught Laboratories brought facilities and experience with biological products; chemists David Scott, Peter Moloney, and Donald Findlay helped improve extraction, concentration, and testing. The University of Toronto then entered a one-year manufacturing collaboration with Eli Lilly (Rutty).

Potency also had to be measured. Early “units” were biological: batches were tested by how far they lowered a rabbit's blood sugar. This standardization was essential because a volume of extract was not a dependable dose unless its strength was known. Purity, stability, sterile preparation, slaughterhouse access to animal pancreases, cold storage, assay animals, and distribution were therefore part of insulin therapy, not secondary details (de Leiva-Hidalgo and de Leiva-Pérez).

The university's patent policy is often reduced to the claim that insulin was “given away.” Banting, Best, and Collip did assign Canadian patent rights to the university for one dollar each, but the university actively patented and licensed processes to supervise quality, production, and royalties. That policy facilitated controlled manufacture; it did not by itself make insulin universally accessible (University of Toronto Libraries, archival highlights; Rutty).

Thompson continued to use insulin and lived until 1935, thirteen years after the trial. His survival is strong evidence of the therapy's historical importance, but it should not be turned into an uncomplicated rescue story. Early insulin demanded repeated injections, carefully coordinated food, urine and blood testing, recognition of dangerous low blood sugar, and an uninterrupted supply. Insulin replaced one fatal trajectory with the work and risks of chronic disease management (University of Toronto Libraries; de Leiva-Hidalgo and de Leiva-Pérez).

The event changed what clinicians could expect from treating severe diabetes. It also made the transformation from laboratory finding to therapy visibly dependent on patients, hospital wards, clinical chemistry, animal experimentation, purification, standardization, public institutions, and commercial manufacture. January 1922 matters less as an isolated flash of discovery than as the point when those systems briefly converged around a patient and produced an effect that could be repeated.