Insulin: 100 years of diabetes treatment

Tracing the history of this remarkable medical breakthrough.
07 January 2022

Interview with 

Kersten Hall, University of Leeds & Charlotte Boughton, University of Cambridge


Chris invites history of science expert Kersten Hall to discuss the first insulin treatment in 1922 before welcoming Charlotte Boughton, a Cambridge University endocrinologist, to tell us what future medical interventions for the condition could be...

Chris - For people who live with type 1 diabetes insulin is a lifeline because, as we know, people with the condition can't make any of the crucial hormone themselves. Insulin tells our cells to pick up the sugar glucose in the bloodstream. Now, if this doesn't happen, our cells run out of energy and levels of sugar in the bloodstream climb dangerously. Long term, this can damage blood vessels, nerves, our kidneys, and also our eyes. Historically, type 1 diabetes was a fatal condition until this week a century ago doctors began injecting the hormone and saving lives. The story of diabetes can actually be traced back much further than a century ago, though. In fact, to 1776 and a certain doctor, Matthew Dobson, who boiled the urine from someone with diabetes and noticed something peculiar:

"There remained, after the evaporation, a white cake, and this cake was granulated; broke easily between the fingers. It smelled like brown sugar. Neither could it, by taste, be distinguished from sugar except that the sweetness left a slight sense of coolness on the palate."

Chris - Luckily not something that we do in medical school these days. With us now is Kersten Hall who's written the book called 'Insulin - The Crooked Timber', which documents the history of diabetes. Bit of an unconventional technique these days, Kersten. Take us back to a hundred years or so ago and tell us a bit more about how diabetes was actually discovered and what insulin's role was found to be.

Kersten - Happy new year, Chris. Thanks very much for having me on the program. Actually, the oldest reference to diabetes is on an ancient Egyptian scroll from about 3000 years ago. I think the Ebers Papyrus talks about a medicine to drive away the making of too much urine, and that's a reference to what I think is probably the most common symptom of the onset of diabetes: frequent passing of water as the kidneys go into overdrive to clear all this excess sugar that's in the blood. By the 19th century, it was known that the pancreas was central to diabetes. Two researchers, Minkowski and Meren, noticed that when you surgically removed the pancreas from a dog the poor animal became diabetic. Researchers began to speculate that maybe the pancreas produces some hormone that controls blood sugar levels and so the hunt was on to find that hormone. The person who is credited with having found it was Canadian scientist Fred Banting. In the summer of 1921, Banting and a younger colleague, Charles Best, who was a final year honour student, had been removing the pancreas from a dog to make the animal diabetic. Then, they'd been making pancreatic extracts and injecting those into the diabetic dogs to see whether that would bring down their blood sugar levels. And then, of course, the time came to test this on a human patient. So, January the 11th, 1922, young Leonard Thompson had been brought into Toronto general hospital. He was 14 years old. He'd been diagnosed with type one diabetes a few years earlier, and when his father brought him into Toronto general hospital he was pretty much at death's door from diabetes because, prior to the discovery of insulin, a diagnosis of type 1 diabetes was pretty much a death sentence. There was nothing really that could be done for the patient other than to put them on a starvation diet to basically delay the inevitable. So, January the 11th, 1922, Leonard was injected with some of Banting and Best's extract.

Chris - Is that from a dog, Kersten?

Kirsten - No. By that time they were making the preparations from bovine pancreas.

Chris - They realised that you could go to another animal and the analogous organ would work? So it was a reasonable step to then say, "well, let's take this because it's a bigger animal, big pancreas, lots available", and try injecting that?

Kersten - That first injection on January the 11th was actually not considered to be a clinical success. The reason was, yes, sure, Leonard Thompson's blood sugar levels fell, but he was still producing ketones in his urine. These are toxic compounds that are produced in diabetic patients. But the most serious result of all was that Leonard had suffered an adverse reaction. Two weeks later, January the 23rd, 1922, Leonard was injected for a second time. This time it was a success. His blood sugars dropped and there was no toxic adverse reaction.

Chris - And how did they achieve the improvement? Did they filter it or clean it up in some way?

Kirsten - Well, what had changed was that second batch of preparation hadn't actually been made by Banting and Best: it had been made by their colleague, James Collin. What he'd been able to do was work out exactly how you could use alcohol to clean up the pancreatic preparation to remove those impurities that were causing the toxic reactions.

Chris - Did he continue to receive these things? Because there've been these sad stories in the history of science and medicine where people invent amazing remedies - and I'm thinking, of course, of Fleming and Florey's penicillin - where people's lives were initially saved but they couldn't sustain the production and people then died because they couldn't keep treating them. What happened here? Was the story a happy ending?

Kersten - Well, Leonard, I think, continues to be treated. He died very young. In folklore, he died in a motorbike accident: that's not actually true - I think he died from a pneumonic infection in the end. But what happened with insulin was, very quickly, word got out about this discovery. The thing is Banting and Best and the Canadian researchers had the media on this. They had two key factors: They had the media on their side, the newspapers, and they also had industrial clout because, of course, it's one thing to make a discovery like this at the lab bench, but to get it scaled up, to get it purified, to get it into patients, you need industrial backing and they had that in the form of Eli Lilly. Insulin is an interesting example in the history of medicine because, by early 1922, this stuff was being put into patients in clinical trials. The problem was production was the rate limiting step and there were stories about patients coming and camping out in order to try and get hold of this stuff. The issue really was how: "How do we make enough of this stuff and how do we make it fast enough to start getting it into patients?" The other issue, as well, was I think the clinicians at the time were concerned about the media trumpeting this as a miracle cure for diabetes because the clinicians at the time knew it was nothing of the sort. What it does is it transforms what would be an otherwise fatal condition into a long term chronic one that can be managed. There was a very eminent American diabetes specialist at the time, Dr. Elliot Joslin. He was concerned about patients suffering depression if their expectations were ramped up beyond what could realistically be expected.

Chris - This is the perfect opportunity to introduce the other person we have here today, Charlotte Boughton, and she's a diabetes and endocrinology researcher at the University of Cambridge. She's been part of a team who in the past have come up with the organ that you've been referring to, the pancreas, in an artificial form so that we can try to treat diabetes better. Charlotte, you better tell us what was in your mind. I mean, how does this work?

Charlotte - Thank you. An artificial pancreas is actually three bits of hardware: unlike the pancreas, which sits in the body, all the bits of hardware are on the body or nearby. It's all external and it comprises a glucose sensor which measures the glucose levels in the fluids surrounding the cells in real time, an insulin pump which delivers insulin just under the skin, and both of these components are commonly used by people with type one diabetes. The clever part of an artificial pancreas, or a closed loop system, is an algorithm which in the system that's been developed at the University of Cambridge is hosted as an app on a mobile phone. That receives information from the glucose sensor and calculates the right amount of insulin to be delivered by the pump. It does this automatically adjusting the levels every sort of 10 to 12 minutes and allows the person with diabetes to continue without having to do those adjustments manually.

Chris - Apart from superior convenience - not having to prick your finger, test your blood and so on - is it better to have a system continuously taking these measurements and topping up insulin? Is that because it's more similar to what your body would do itself? Does that translate into better health?

Charlotte - Yeah, exactly. It allows you to get more what we call physiological replacement of insulin and we've shown, and other groups internationally as well, that close loop systems allow people to get better glucose control than with a pump on its own or injections, and also can reduce the risk of dangerous hypoglycemia as well. So there's both biological benefits, but also psychological and burden benefits for somebody living with type 1 diabetes.

Chris - Is this the only show in town or are you working on version 2.0 which doesn't have any external features? How are you going to take this forward?

Charlotte - At the moment, the systems that are commercially available to people with type one diabetes still require people to count their carbohydrates and manually tell the insulin pump how much they're eating and therefore how much insulin to deliver. With the newer ultra rapid insulins that have been developed we are seeing at the moment whether those are quick enough to get rid of the need for people to interact with it. So, actually, people could literally just wear the devices and forget about it until the devices need changing. That would truly reduce the burden of type 1 diabetes management. There are also lots of other areas of research within this including smart insulins, which are designed to turn on when they're needed (when glucose levels are high) and turn off when they're not. I think these are quite far behind where we are with diabetes technologies such as closed loop. But I think these are certainly something to look out for in the future.

Chris - I wonder what Leonard would've made of this a hundred years ago, Kersten.

Kersten - I think, well, I think he would've been delighted. I just wanted to say to Charlotte, I actually have type 1 diabetes myself. In fact, as you've been speaking there I've just been doing a quick blood glucose check. I'm fitted with one of these cyborg devices. That's how the kids like to think of it as Dad being a cyborg - one of these implants. I just swipe my phone on it. I thought your work there on the artificial pancreas is so exciting. As somebody with type 1 diabetes, I'm really gonna be watching this with eager anticipation, and as somebody who's just written a book about the history of insulin I'm thinking, "oh my goodness, I better get in touch with my publishers and beg them to publish a second edition", so I can write another chapter to cover the research you've been doing! I just wanted to say thanks very much for all you're doing there and I look forward to it. Got big hopes. Fingers crossed.

Chris - Thank you very much Kersten and thanks also Charlotte. That's Charlotte Boughton and Kersten Hall. And you can get Kersten's book that's out from the 13th of January, appropriately enough for the anniversary; the centenary. It's called 'Insulin - The Crooked Timber: A History From Thick Brown Muck to Wall Street Gold.'


Add a comment