Many of you will have heard about the ‘smart’ pump newly released onto the market, with the technology to cut out insulin delivery to prevent hypoglycaemia. This ‘smart insulin’ breakthrough isn’t about delivering insulin differently, as has been the case with previous insulin developments – it’s about the insulin itself being modified – and glucose responsive.

One of the major difficulties in managing diabetes is in knowing when and how much insulin to give, factoring in food, exercise, and other variables. The risk of hyperglycaemia and hypoglycaemia is ever-present, making daily management difficult and quality of life, well, frankly less than great.

This new ‘smart’ insulin bears some remarkable similarities to an ‘artificial pancreas’ (perhaps more so than the somewhat controversially released ‘smart pump’)  in its effect of automatically lowering blood sugar levels, closely mimicking the response of a healthy pancreas.

“Right now, all the insulin that a patient takes is ‘dumb,’ in a way,” says Daniel Anderson, a molecular geneticist at the Harvard–MIT Division of Health Sciences and Technology and a co-author of the study published in Proceedings of the National Academy of Sciences. “Currently available insulins act independent of the sugar levels in the patient”, which can cause dangerous highs or lows in a patient’s blood sugar levels. This smart insulin only ‘switches on’ when blood sugar climbs.

How?  This new compound is a naturally occurring hormone, but with a molecular switch made of a chain of fatty molecules added, to make the insulin stick around in the blood for longer periods, so it’s not free to cling on to sugar.

With an effect of up to 14 hours, it can repeatedly and automatically lower blood sugar levels in mice after they’re fed sugar. The product works in two ways: with a long-term lowering effect (like Levimir or Lantus), while also having the ability to reduce blood sugar levels in the short term. Testing by continuous glucose monitoring in mice (if you’re a visual person like me – your mind may also be boggling trying to imagine the difficulties in finding enough real estate on a tiny mouse for a CGM…  ) has shown the insulin to respond to glucose as effectively as a healthy pancreas.

This would take away the need for extra shots or boluses of  insulin to cover meals – giving back greater certainty in glucose control, and definitely more freedom.

The technology is a way off, having only been tested in mice, but may be ready for Phase I clinical human trials in 2-5 years. Actual release onto the market, if clinical trials prove successful, would be another few years from there.

One injection a day and stable blood glucose?

I mentioned all of this to my little girl in the car the other day, and she thought it sounded awesome. ‘I would LOVE that so much. Imagine if I could have one of those mice that have diabetes, and we could do little finger pricks on its paws at the same time I do mine, and give it insulin and take care of it’.

Erm…… yep. That would be, very cute. And high maintenance. Fearing she’d missed the point somehow, I asked a couple of extra questions.

How about the part where she would only have to have one needle a day and not do anything at meal times except eat?

‘Oh! Really? Is that what that means? That would be amazing. So much better than now. Could I still have a mouse?’

Sounds like it’s worth the wait.

The likely benefits: Fewer injections and better glycaemic control. With more predictable levels, finger prick testing might also be reduced.  Happy fingers, happy bellies. MUCH better glycaemic control, if early tests on mice translate to similar efficacy in human trials. Yes please!

Ref: Choua D, Webbera M, Tanga B et al. ‘Glucose-responsive insulin activity by covalent modification with aliphatic phenylboronic acid conjugates’, 2015; 112(8); 2401-6; February 24, 2015 vol. 112 no. 8 2401-2406. Epub before print February 9, 2015, doi: 10.1073/pnas.1424684112