Imagine this: an insulin that responds to blood sugar changes. An insulin that works when you need it and turns off when you don’t.

Drugmakers have been working on a so-called “smart insulin” for decades. A smart insulin could allow people with diabetes to more aggressively target normal blood sugar levels. It could reduce the risk of both hypoglycemia and hyperglycemia, leading to better health outcomes in both the short and long term.

“We’re asking for a lot, chemically, in the development of a ‘smart’ insulin,” says Matthias von Herrath, MD, the scientific director of the Diabetes Research Institute and a vice president and senior medical officer at Novo Nordisk.

Today, this innovation is still very far from human trials, let alone the shelves in your local pharmacy, but there has been a breakthrough. Dr. Von Herrath spoke to Diabetes Daily about Novo Nordisk’s recent successful experiments with its own glucose-responsive insulin. Here’s a closer look at the progress researchers have made and why it’s so darn complicated.

Smart Insulin Must Be Flawless

“Developing a glucose-responsive insulin means you have to build a molecule that, when it sees glucose, becomes active,” says von Herrath, who has not been directly involved in Novo Nordisk’s research into smart insulin.

“I know people with type 1 think of [smart insulin] like it might be a vacation,” says von Herrath, “you might simply inject your insulin and then you don’t have to think about your glucose levels. But, again, chemically, that is asking a lot from a molecule. Building that type of molecule to only become active when it sees glucose is not an easy feat.”

Von Herrath named a few of the most significant obstacles:

Smart insulin requires extreme accuracy: Glucose-responsive insulin must be very precise to ensure safety. It must know exactly how much insulin to release and when. It must also know when to stop releasing insulin. “What if you eat and you activate all of the glucose-sensing insulin versus just the right amount? Again, this is a very big ask!” says von Herrath.

Smart insulin has a risk of cross-reactivity: There are other things within the body that could look like glucose, explains von Herrath. “If these glucose-sensing molecules cross paths with other medications or hormones that could be confused with glucose, this raises a huge safety risk.” Von Herrath says this is a huge concern that will take extensive research to address.

Diabetes is more than just insulin: Within 20 seconds of eating, the islet cells in your pancreas begin communicating with a variety of cells in the body. This includes the beta cells that produce insulin, but there are many other cells and hormones that play a significant role in how your body manages food.

Smart insulin must be flawless: Imagine a day’s worth of “smart insulin” sitting in your system, waiting to be triggered by rising glucose levels. What if it were to all suddenly activate for no reason? Like an insulin pump dumping 50 units of insulin to your body all at once. This could easily be fatal. For a glucose-sensing insulin to truly be safe for human use, it must be essentially flawless. It cannot be easily confused or disturbed by other aspects of the body.

A New Molecule

Though the challenge is great, researchers have made important progress.

Novo Nordisk’s scientists have engineered a molecule they’ve named NNC2215. This molecule essentially has a “switch” that responds to rising levels of glucose in the bloodstream, allowing insulin to become more or less active. When blood glucose levels rise high enough, the insulin becomes more active. As blood glucose levels drop, the molecule’s action slows and closes the switch, making it no longer able to pick up glucose.

The first successful trial of NNC2215 took place recently. A team led by researcher Rita Slaaby used rat and pig models to test the molecule’s efficacy. As blood glucose levels rose from 50 mg/dL to 360 mg/dL, the glucose-responsive insulin became more responsive, taking up more glucose.

The results: the new insulin proved to be as effective as human insulin in lowering blood glucose levels. And just as hoped, the insulin was significantly less active when it was exposed to low blood sugar.

“The chemistry is advancing in an amazing way,” says von Herrath. “I’m surprised it’s even possible! When it was first brought up 20 years ago, I was thinking, ‘What a bunch of baloney! Nobody will be able to build something that works like that. That’s pie-in-the-sky impossible.’”

A Building Block

NNC2215 isn’t ready for humans yet, and it is years away from approval by the U.S. Food and Drug Administration (FDA), if it ever gets there. But its success has caused von Herrath to change his tune on the possibility of glucose-responsive insulin.

“When I first saw these advancements and the chemistry, I thought ‘Okay, this is amazing,’ but I would still be very cautious to say it’s there. It is advancing but it has a long way to go.”

Von Herrath also says the first types of this glucose-responsive insulin won’t be the magical solution to the challenges of dosing for meals like many people with type 1 might hope: “The first versions of these insulins will be smart basal insulins that aim to prevent hypoglycemia. I would temper my expectations that they would be a rapid-acting insulin that could handle food,” says von Herrath. “I don’t want to discourage anyone, but I think that will take much more time.”

The challenges, he says, come down to the severe precision a medication like this must demonstrate to be even reasonably safe.

“It would need to operate with extremely high specificity and sensitivity. It has to be completely fail-safe. Like a gas line, it cannot leak. It needs to be extremely precise.”

Regardless of the long path ahead, von Herrath says progress is significant and worthy of celebration.

“At the Diabetes Research Institute, we celebrate every small victory, and we should celebrate the progress, the little victories, with these glucose-sensing insulins, too,” he adds. “But this is a building block on a road that requires a fair amount of small victories.”

While he is optimistic and excited about the future of “smart” insulin, von Herrath knows his perspective is not the same as a person living with the daily burden of type 1 diabetes.

“I’m a glass-half-full kind of guy, but that’s easy to say when you don’t live with T1D yourself,” says von Herrath. “But patients should understand this research and why it is so challenging. Ten years ago, I would’ve never thought this was even possible, so we’ve come a long way.”