Newly designed 'smart' insulin could improve type 1 diabetes treatment

https://www.sciencealert.com/a-new-kind-of-smart-insulin-could-help-diabetics-avoid-the-risks-of-low-blood-sugar-levels

by lnyan • 5 years ago

113 49 5 years ago

AnthonBerg3• 5 years ago

This is the paper in question: https://www.pnas.org/content/118/30/e2103518118

Direct PDF link: https://www.pnas.org/content/pnas/118/30/e2103518118.full.pd...

The sciencealert page is a digest of the author’s university’s digest. The papers themselves are usually much better material. Quite understandable in my experience.

That said, as someone in a relationship with autoimmune diabetes, I hope this work comes to fruition. It’s obviously a good idea to try to research and develop this further.

cletus• 5 years ago

Cynical title: Change with questionable benefit extends patent life and thus prices of insulin even further.

To be clear: I don't know a lot about insulin. I'm fortunate enough not to be diabetic. So there may have been useful advancements in insulin. The price of insulin in the US does seem to be awfully high however.

It is a common practice for pharmaceutical companies to alter the active molecule in some way, claim the new version has benefits, reset the patent timeline and stop producing the old version.

EDIT: in case it wasn't clear, if this is actually good for diabetics, that's fantastic. Insulin in the US shouldn't be as expensive as it is. I'm just seen too much profiteering (eg Epipen) by that industry.

_Microft• 5 years ago

I am not a biologist but the proposed behavior of the molecule seems highly beneficial for the patient. It is supposed to only activate in presence of enough glucose instead of always. This is meant to prevent dangerously low blood sugar. They only conducted experiments with fructose so far.

Trasmatta2• 5 years ago

I am T1 diabetic, and this change, if it works, could absolutely be a game changer.

So there may have been useful advancements in insulin

There have been massive advancements in insulin over the past couple of decades. Today's long acting and short acting insulins are a massive improvement over the older kinds. The price of insulin in the US is a disaster (when not covered by insurance), but otherwise the quality of insulin treatment is better than it's ever been.

randcraw• 5 years ago

I'm a T2 diabetic, and though I don't (yet) take insulin, I'm pretty sure this could be a very big win for other T2s who do. Dosing insulin is clumsy with an insulin pump, and always hit-or-miss without one (insulin arrives too early or too late or in the wrong amount). The adaptive release of insulin only when glucose levels rise could provide a near-ideal therapy for millions. If it's affordable...

jcims• 5 years ago

I'm not an expert, but have a lot of exposure to this through my kiddo who is a type-1 (insulin deficient) diabetic. This could represent a step function change in her therapy.

Only for folks using the latest and greatest in closed-loop pump therapy is there even an approximation of the pancreas function to elevate insulin levels in response to elevated blood glucose levels, and even then its too slow to be effective. Diabetics that want to closely manage their blood glucose levels have to carefully plan how they eat and how active they are, modulo sickness, periods/etc.

If it was possible to have a latent insulin bolus floating around the bloodstream that could react to gluecose directly, it could provide an important buffer and possibly even a wonderfully passive approach to therapy vs. the ridiculous dance that patients undergo today.

Someone• 5 years ago

If patents were a significant factor in the high prices for insulin in the USA, you would expect prices to be similarly high in all countries that recognize those patents.

They aren’t. https://www.rand.org/pubs/research_reports/RRA788-1.html:

“Although the ratio of U.S. to other-country prices varied depending on the comparison country and insulin category, U.S. prices were always higher (often five to ten times higher) than those in other countries.“

“Other countries” includes Canada and Germany, so they didn’t look at particularly poor countries. The linked PDF says

“The average U.S. manufacturer price per standard unit across all insulins was $98.70, compared with $6.94 in Australia, $12.00 in Canada, $7.52 in the United Kingdom, and $8.81 across all non-U.S. OECD countries combined”

jellicle2• 5 years ago

All of those countries set how much they'll pay for insulin. They have price controls. The USA does the opposite; requires governmental medical insurance providers NOT to bargain with pharma companies and instead explicitly requires them to pay whatever price the pharma company sets.

Low prices could come from either a) a robust, competitive market in producing and selling insulin or b) governmental price controls. The USA explicitly chooses not to have either. Other countries typically choose b) backed up by a threat of a) (if you're a pain in the ass over our negotiated price controls, we'll go source a large supply of insulin from somewhere else).

RHSeeger3• 5 years ago

Add to that that the insulins used by a lot of people are over 20 years old and off patent. However, they are biologics and, as such, can't be simply made as a generic. Any "generic" needs to go through testing and approval, which means the generics (like Basaglar is for Lantus) aren't that much cheaper (80% of the cost, I believe).

hooande• 5 years ago

This comment is just wrong. This isn't a 'questionable benefit', it's a major change to the fundamental technology. This has nothing to do with price or profits at all. Comparing this theoretical advance to existing insulin is like comparing electric vehicles to gas powered cars. It's a difference in kind, not degree.

To explain, currently your pancreas is measuring your blood sugar and producing more or less insulin to keep it at a safe level (let's call that level "100"). A type 1 diabetic has to do that process manually, measuring their blood sugar with a device and taking enough insulin via needle to keep it at 100. It's possible to overshoot and have blood sugar go far below 100, which can do real damage to cells that need sugar as an energy source. Obviously if you have to measure and adjust several times a day, every day, even a 1% human error rate is going to come up often.

This theoretical insulin would stop working if there wasn't sufficient sugar already in the blood. As another comment said, it would basically stop working if your blood sugar was below 75 or 80. Right now diabetics are at risk from blood sugar being too high OR too low. This would cut out half of that risk, and be life changing for millions of people.

Has nothing to do with price, and as always I must mention that generic insulin is abundant and readily available at low prices to anyone. I don't expect everyone to follow the details of insulin and having diabetes. But it would be nice if medical advancements could be discussed without people conflating the topics with their pet political issues.

ruined• 5 years ago

is the old insulin free yet

boublepop2• 5 years ago

It costs 25$ in wallmart, or if you are uninsured and low income there are patient care programs that will provide it to you for free.

Trasmatta2• 5 years ago

Wal-Mart insulin will do in a pinch, but it's pretty garbage compared to the better formulations

code_duck8• 5 years ago

We have been hearing of this in the type 1 diabetes community for some time and it sounds very promising.

As the article explains, the main risk of insulin use is hypoglycemia, which is essentially the result of an insulin overdose. When you dose insulin, you have to carefully match it to the carbohydrates you subsequently consume and the glucose your liver secretes. Too little insulin and you will have high blood glucose, too much and you will suffer low glucose. Highs are uncomfortable and risky long term, but lows are very uncomfortable and immediately dangerous.

For insulin to stop working once your glucose got to 75-80 would be perfect. If the insulin also stayed there and activated as soon as your glucose went up, that might be even better than having a fully functional pancreas.

cperciva4• 5 years ago

For insulin to stop working once your glucose got to 75-80 would be perfect.

I'm not so sure about that. Insulin does more than just regulate blood glucose levels; it also affects fat metabolism, potassium levels, and many other things. If the insulin "turned off" completely we could see patients who went into ketoacidosis as a result of fasting and/or exercise.

code_duck8• 5 years ago

Yes, your body cannot have zero insulin. You do not need/want more insulin once your glucose is 70-75, though. I'm quite familiar with both hypoglycemia and ketoacidosis, having nearly died of undiagnosed type 1 last Spring.

Ketoacidosis is typically associated with hyperglycemia. It happens when your body has no insulin. There is what's called euglycemic ketoacidosis but it's rare. In typical ketoacidosis, your blood is very high in glucose, but lacking insulin, your cells have no way to obtain it. Since your body cannot get energy out of your blood in the normal fashion it starts to digest ketones, which acidifies the blood and creates many harmful effects.

In the situation we are discussing, when your glucose rises above 75 from food or liver secretion, the novel insulin would start working, preventing ketoacidosis.

code_duck8• 5 years ago

I wanted to add that one also must know how insulin is dosed to understand this.

Type 1 diabetics typically take two different insulins. One is a 'long acting' that slowly absorbs over 16-24 hours, like what would be called time-release for oral medication. This insulin provides for metabolic function and counteracts the glucose your liver unpredictably and uncontrollably releases at certain times. People with pumps drip regular insulin at a slow rate for the same effect. I assume nothing would change there.

The other type is called 'fast acting' or bolus insulin. This is taken usually in one large dose prior to meals to handle the large rise in blood glucose that follows carbohydrate consumption. The insulin we are discussing here is a replacement for fast acting insulin.

The -ideal- situation for fast acting is for it all to be consumed at the end of digestion. You want fast acting to match 1:1 with the carbohydrates you ingest, every time, with none left to cause hypoglycemia (overdose). other metabolic needs are already handle by the pump drip/long-acting. This innovation would remove the danger of overshooting the insulin dosage and causing hypoglycemia, which is huge.

AnthonBerg3• 5 years ago

It would be cool!, but I’d go for the pancreas: As far as I know, the pancreas releases insulin in pulses. Each pulse 3-5 minutes after the other? The pulses seem to matter. Supposedly the loading-unloading cycle on the insulin receptor seems to help prevent insulin resistance. (It does somehow make intuitive sense to “shake” cells’ sensors a bit to help them stay on target.)

Another thing is C-peptide. It’s a “byproduct” of the insulin protein folding in the pancreatic cell. I believe C-peptide is now known to be a biologically active control molecule with antiinflammatory properties.

Source: The Wikipedia page on insulin iirc. I an only a layman and may be severely mistaken about all of this.

code_duck8• 5 years ago

Of course you need insulin. Just not when your glucose is 75 (below 70 is hypoglycemia).

Sure, it's complex.

People with type 1 don't have any insulin production at all, and typically get it in large doses (syringe) that absorb slowly or dripped out from a pump. With no endogenous production, we do not make c-peptide (it's a protein or something cleaved off of pro-insulin in the body's synthesis of insulin. Levels are measured in blood tests to determine your level of insulin production - after honeymoon period, type 1s produce no insulin and therefore no c-peptide).

gorb314• 5 years ago

I think this is an interesting research direction.

From my perspective though, being a T1, what I want is an even faster acting insulin than we already have, with a shorter "on" time.

To explain: the shortest acting insulins we have right now take effect about 30 minutes after injection. Then the dose stays active for about 4 hours. So for me to react to a high blood sugar, and until I can be certain that I did not inject too much (nor too little), can take up to 4 hours.

That is almost unbearably long, especially when my blood sugar climbs through the roof, and I am not sure whether it is because my pump line was blocked. Overloading on insulin (called "stacking") to compensate for a crazy high blood sugar can be very dangerous - plus I can't think straight when my sugar is that high. Nothing sucks more than having to sit and wait for a high blood sugar to come down...

Contrast that lagged response time to eating food, especially high GI sugars such as a soft drink, orange juice, or even bread. This can spike my insulin in a matter of minutes.

All this makes matching sugar spikes with insulin response curves very hard; it is a one-side-overdamped-and-other-side-underdamped dynamic system. Any diabetic will tell you how crazy a "roller coaster ride" can be, which is what happens when you overcompensate on either end.

To get back to the topic of the 'smart' insulin: if this research can lead to an insulin that can always be in my bloodstream, ready to act when my blood sugar rises, and stop when my blood sugar falls, and if it can do this quick enough, then that would be a game changer.

zippergz2• 5 years ago

Have you tried the inhaled insulin? It takes effect MUCH faster than any injected insulin, does it’s job, and is out of your body quickly so also reduces the risk of stacking. In some ways it is not as convenient as a pump, but it works much better in terms of timing.

zby2• 5 years ago

How do you take it? I imagine it would be ideal to quickly beat hyperglycemia when you underestimate a meal bolus. But I don't see it replacing a pump (or pens), because it would be difficult to: measure it exactly and keep supplying it continuously.

11235813212• 5 years ago

Fellow T1 here--when you're in that situation you shouldn't sit while your bolus barely keeps up with fast carbs, you should do something physically active to help get your numbers down more quickly.

I share your wish in your last paragraph very much!

pixl97• 5 years ago

If only physical activity paid as much as holding a chair down.

FigurativeVoid2• 5 years ago

That's a really great point. I am T1 as well. The improvements I would love is an insulin that is denser and more stable.

If we had insulin that was say.. 1000 units/mL (10x stronger) and didn't degrade as fast, then we could have some really cool and slim devices. You could also have a longer lasting pump site.

ninjin• 5 years ago

Not so sure about the pump sites. My experience is that the reason for changing pump sites is not just about your insulin cartridge running low, but also about tissue damage at the pump site. Based my own two years of pump usage, changing every three days (as recommended) rarely leads to visible damage, but pushing it to say four days frequently does. Sample size of one though.

zby2• 5 years ago

What the article speculates is a way to load up insuline in the blood that would not act if the blood glucose is below let's say 120 - but when it is above - then the insuline opens and acts. This is the fastest possible way of it to act. Even if it degrades and you could not keep much of it constantly in the blood stream - then you could inject it much before the meal and it would just start acting when the blood sugar increases and also you could inject more than enough and not worry about injecting too much of it - because it would never lead to hypo.

This would be the ideal - in a way it would be better than functioning pancreas - because pancreas needs a few minutes before releasing insuline and this would act instantly.

RHSeeger3• 5 years ago

It would be nice to have something like this with a several hour window. I currently suffer from the dawn phenomenon, and my blood sugar spikes by 100+ most mornings (starting anywhere between 5am and 7am). It's super frustrating to have to get up every day at 630-7am to check my blood sugar and, if necessary (most days), take some insulin.

gukov• 5 years ago

Have you looked into Fiasp? I found it to be faster than Apidra.

ryneandal• 5 years ago

I'm a fan of anything to make management of T1/LADA simpler. Didn't realize how much of a chore management is until I was diagnosed with LADA in October of 2019. I feel for all of you that have been dealing with T1 for years/decades. It's exhausting.

TheBlight• 5 years ago

Will they give it away for free in parking lots to anyone who shows up?

Newly designed 'smart' insulin could improve type 1 diabetes treatment

Related Stories