Issue: May 2020
Disclosures: Buse reports he receives research support from or serves as an adviser for Adocia, AstraZeneca, Boehringer Ingelheim, Cirius Therapeutics Inc., CSL Behring, Dance Biopharm, Dexcom, Elcelyx Therapeutics, Eli Lilly, Intarcia, Johnson & Johnson, Lexicon, MannKind, Metavention, NovaTarg, Novo Nordisk, Neurimmune AG, Pendulum Therapeutics, Sanofi, Senseonics, Stability Health, Theracos, Tolerion, vTv Therapeutics and Zafgen and holds stock options in Mellitus Health, Pendulum Therapeutics, PhaseBio and Stability Health. Cornell reports she serves on an advisory board for Novo Nordisk. Hirsch reports he has received research support from Medtronic Diabetes and is a consultant for Abbott Diabetes Care, Bigfoot and Roche. Isaacs reports she has received consultant fees from Dexcom and Novo Nordisk. Grunberger and Nathan report no relevant financial disclosures.
May 18, 2020
12 min read

The future of insulin: Pills, patches, weekly formulation could change diabetes management

Issue: May 2020
Disclosures: Buse reports he receives research support from or serves as an adviser for Adocia, AstraZeneca, Boehringer Ingelheim, Cirius Therapeutics Inc., CSL Behring, Dance Biopharm, Dexcom, Elcelyx Therapeutics, Eli Lilly, Intarcia, Johnson & Johnson, Lexicon, MannKind, Metavention, NovaTarg, Novo Nordisk, Neurimmune AG, Pendulum Therapeutics, Sanofi, Senseonics, Stability Health, Theracos, Tolerion, vTv Therapeutics and Zafgen and holds stock options in Mellitus Health, Pendulum Therapeutics, PhaseBio and Stability Health. Cornell reports she serves on an advisory board for Novo Nordisk. Hirsch reports he has received research support from Medtronic Diabetes and is a consultant for Abbott Diabetes Care, Bigfoot and Roche. Isaacs reports she has received consultant fees from Dexcom and Novo Nordisk. Grunberger and Nathan report no relevant financial disclosures.
You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact

Next year will mark the centennial of the discovery of insulin by the Canadian surgeon Frederick Banting and medical student Charles Best, who famously isolated the secretions from islet cells in 1921. Since then, insulin for treating diabetes has undergone multiple evolutions, from the earliest beef and pork formulations, to the introduction of the first synthetic “human” insulin in 1978 to the rapid-acting and long-acting human insulin analogues introduced during the mid-1990s.

Today, there are six main types of insulin produced by the three insulin manufacturers serving the U.S. market, each varying by onset, peak and duration of action: rapid-acting, short-acting or “regular” insulin, intermediate-acting, long-acting, ultra-long acting, and “premixed” insulin, a combination of intermediate and short-acting formulations. A fast-acting insulin aspart injection (Fiasp, Novo Nordisk), described as the only mealtime insulin without a premeal dosing recommendation, was approved for children with diabetes as young as age 2 years in January.

“We’re seeing faster insulins that better mimic our insulin production, so it is easier to manage glucose with formulations that last longer and don’t have the peaks,” Diana Isaacs, PharmD, BCPS, BC-ADM, CDCES, clinical pharmacy specialist and continuous glucose monitoring program coordinator in the department of endocrinology, diabetes and metabolism at the Cleveland Clinic Diabetes Center, told Endocrine Today. “In addition to that, diabetes technology has improved dramatically. With better pumps, more accurate CGMs and smart pens that record insulin dosing, we have come a long way, even in the last 10 years.”

Still, no currently available insulin perfectly mimics the body’s physiologic production of the hormone, and researchers continue to seek better formulations that do not involve subcutaneous injection.

George Grunberger, MD, FACP, FACE
The next decade of advancements in diabetes therapy may include oral insulins, according to George Grunberger, MD, FACP, FACE.
Photo by Igor Kupriyanchik, Grunberger Diabetes Institute. Printed with permission.

“If you give insulin subcutaneously, even the fastest formulations do not start rising for 10 to 15 minutes and don’t hit their peak for about an hour and then fall off and are gone by about 2 to 4 hours,” David M. Nathan, MD, director of the Massachusetts General Hospital Diabetes Center and professor of medicine at Harvard Medical School, told Endocrine Today. “The limitation of all of these drugs is that we are still injecting them subcutaneously. That is the major cause of the delay of action. They have to be absorbed from the fatty tissue into the vasculature. No matter how ‘fast’ they make them, no matter what modifications they make, there is still that barrier.”


Several cutting-edge advancements in noninjectable insulin delivery methods could offer new ways for people with diabetes to manage the disease and better control glucose response. New research from a phase 2b study suggests that an oral insulin formulation is now closer than ever before to becoming reality, and a coin-sized “smart” insulin patch has shown promise in recent animal studies.

A novel, super-long-acting basal insulin — dosed as a once-weekly injection — is also in early development.

“Now researchers are trying to tweak around the edges with insulin, because what else are you going to do?” George Grunberger, MD, FACP, FACE, chairman of the Grunberger Diabetes Institute in Bloomfield Hills, Michigan, told Endocrine Today. “Once you have a preparation like insulin degludec [Tresiba, Novo Nordisk], proven to last up to 42 hours, how can you do better? The question I ask is, do we still have to inject to manage diabetes? My answer is, not necessarily. As far as the future of insulin delivery, we just don’t know yet. We don’t know what is going to make it to market.”

Promise of insulin pills

Oral delivery is the simplest and least invasive way to deliver many pharmaceuticals, but many agents, including insulin, cannot survive passage through the stomach or the gastrointestinal (GI) tract, Grunberger said.

“Mother nature does not want your GI tract to absorb large molecules,” Grunberger said. “There are lots of things out there that serve as barriers to oral injections of peptides and proteins. These obstacles are built in to protect us ... to make sure nothing gets through that does not belong in the GI tract. For decades, people have tried to come up with other ways to deal with that, and there are some fascinating approaches that are trying to remove or decrease those barriers.”

One possible solution was inspired by the leopard tortoise’s ability to passively reorient. NIH-funded researchers developed an ingestible, self-orienting, millimeter-scale applicator, or SOMA, that autonomously positions itself to engage with GI tissue and inject insulin, Grunberger said. In an article published in February 2019 in Science, the researchers wrote that the oral capsule deploys milliposts fabricated from active pharmaceutical ingredients directly through the gastric mucosa while avoiding perforation. In vivo studies conducted in rats and swine support the applicator’s safety and, using insulin as a model drug, demonstrated that the SOMA device delivers active pharmaceutical ingredient plasma levels comparable to those achieved with subcutaneous millipost administration, the researchers wrote.


“You have this capsule, and inside there is a spring-loaded needle, and on it is compressed, freeze-dried insulin,” Grunberger said. “When you swallow it, it lands the right way and the needle ends up on the gastric mucosa, and then insulin gets injected into the body and the outer covering is dissolved. We’ll see where this device winds up.”

Diana Isaacs

An oral insulin capsule is in development by Oramed Pharmaceuticals. In February, the company reported safety and efficacy results from the final cohort of its phase 2b trial, showing that the lead oral insulin candidate, ORMD-0801, met its primary endpoint demonstrating that participants with type 2 diabetes who received once-daily and twice-daily 8 mg doses achieved statistically significant reductions from baseline in HbA1c. In announcing the findings, the company noted there were no serious drug-related adverse events observed and no hypoglycemia.

The Oramed data pave the way for FDA discussions regarding a phase 3 trial. The drug has the potential to be the first commercial oral insulin capsule for the treatment of diabetes.

“It works, at least if you judge the top-line results,” Grunberger said. “They claim that, compared with placebo, the oral insulin capsule lowered HbA1c by a clinically meaningful 0.6%. This is one of the efforts, and we will see if it winds up in the clinic.”

An oral insulin tablet has been pursued by Novo Nordisk. The company reported results from a phase 2, 8-week randomized, double-blind, double-dummy controlled trial with oral insulin 338 (I338), a long-acting, basal insulin analogue in tablet form with the absorption-enhancer sodium caprate. In a study published in March 2019 in The Lancet Diabetes & Endocrinology, researchers reported that I338 safely improved glycemic response for 25 insulin-naive adults with type 2 diabetes, with no evidence of a difference when compared with subcutaneously administered insulin glargine.

Further development was discontinued because I338 insulin doses were high and production of the required quantities of I338 was deemed not commercially viable.

“Cost is another barrier here,” Isaacs said. “Because so much is insulin broken down in the GI tract, you need to have higher amounts. For what might be 10 U, you may need 100 U of insulin in an oral formulation. It is similar to what we have seen with oral semaglutide [Ozempic, Novo Nordisk], the first oral GLP-1 receptor agonist. It is a higher dose, up to 14 mg, vs. the 1 mg in the injectable version, and it has to be taken a special way, 30 minutes before one eats. There are tricks with getting oral formulations to work.”


Other issues also must be considered with oral formulations, according to Susan Cornell, PharmD, CDCES, an Endocrine Today Editorial Board Member and associate director of experiential education and associate professor of pharmacy practice at Midwestern University Chicago College of Pharmacy.

David M. Nathan

“There are companies out there working on this, but now you have this pill — really, a delivery device — inside your body,” Cornell told Endocrine Today. “The first question would be, is it going to do harm? Then, when the body eliminates the device, what is that doing to the environment? Some of these devices contain metal. That’s getting into the water system. Here we are campaigning about bisphenol A (BPA) and water bottles, yet something like this could be just as bad or worse. We also have to keep safety in mind.”

Inhaled insulin

An inhaled rapid-acting mealtime insulin (Afrezza, MannKind), approved by the FDA in 2015 to improve glycemic response in people with type 1 and type 2 diabetes, is known for its ability to address postprandial hyperglycemia, Grunberger said. The drug, which starts working 12 to 15 minutes after inhalation, was approved nearly a decade after the first FDA-approved powdered native human insulin (Exubera, Pfizer) was withdrawn from the market in 2007 due to low sales.

“MannKind uses its Technosphere technology to deliver insulin directly into the lungs,” Grunberger said. “The idea here is a more patient-friendly device. When you get insulin into the lungs, there is a one-cell layer between the air sac and the circulation, so it works very quickly. It can reduce fear of any postprandial hyperglycemia.”

Inhaled insulin, however, has some of the same limitations as an oral formulation, according to Irl B. Hirsch, MD, professor of medicine at the University of Washington School of Medicine in Seattle.

“There are so many difficulties with getting enough insulin in,” Hirsch told Endocrine Today. “Even then, once you have something, you’re building factories for the amount of insulin you need, because you’re going to need a very large amount. Even when we use Afrezza now, it’s a lot of insulin powder required to get a little bit of insulin into the system.”

Oral-lyn, a buccal insulin spray (Generex), also remains in development. In 2015, the University of Toronto’s Center for Molecular Design and Preformulations enhanced the Oral-lyn formulation to make it more attractive for people with diabetes and prospective commercialization partners, Grunberger said. Changes included an increase in the bioavailability of insulin in the product and a reduction in the number of sprays required to achieve effective prandial glucose control.

Irl B. Hirsch

This year, the NuGenerex Diabetes Research Center accelerated development of the reformulated Oral-lyn-2 for type 2 diabetes and Altsulin for the treatment of type 1 diabetes, Grunberger said, adding that programs were expected to begin in the first quarter of 2020.

Despite the advancements, uptake of inhaled insulin has been low, Isaacs said.

“Inhaled insulin is probably underutilized,” Isaacs said. “We can’t use them in insulin pumps, so that is one barrier. People who smoke or have asthma can’t use it. The dosing is different too, so it could have something to do with a person’s comfort level with it. Some people just don’t know about it, and also, it does have to be dosed frequently. We may use it in addition to other insulins; it’s great for correcting high glucose. It’s not for everyone.”

‘Smart’ insulin patch

In February, researchers from UCLA, the University of North Carolina School of Medicine and MIT announced the successful test of a smart insulin-delivery patch that could one day monitor and manage glucose levels for people with diabetes and deliver the necessary insulin dose.

Susan Cornel

In a proof-of-concept study published in Nature Biomedical Engineering in February, Jicheng Yu, a doctoral student in the joint department of biomedical engineering at the University of North Carolina at Chapel Hill and North Carolina State University in Raleigh, and colleagues reported that the removable, transdermal patch, bearing microneedles loaded with insulin and a nondegradable glucose-responsive polymeric matrix, was shown to regulate glucose in insulin-deficient mice and minipigs.

“Under hyperglycemic conditions, phenylboronic acid units within the polymeric matrix reversibly form glucose-boronate complexes that — owing to their increased negative charge — induce the swelling of the polymeric matrix and weaken the electrostatic interactions between the negatively charged insulin and polymers, promoting the rapid release of insulin,” the researchers wrote.

“The smart insulin patch leverages chemistry to release insulin in parallel with fluctuations of glucose,” John Buse, MD, PhD, director of the Diabetes Center, director of the North Carolina Translational and Clinical Sciences Institute, executive associate dean for clinical research at the University of North Carolina School of Medicine in Chapel Hill and a principal investigator for the study, told Endocrine Today. “The results in mouse and pig models of diabetes are impressive.”

The researchers are now applying for FDA approval to conduct clinical trials in humans.

“If successfully translated to human use, it could dramatically reduce the burden of insulin treatment,” Buse said. “That effort will require tweaking the formulation to tune it to application in people with type 1 and type 2 diabetes, and then demonstrating safety and effectiveness. We are optimistic that this effort will be successful, but it will take time.”


Weekly formulation

PhaseBio is developing PE0139, a so-called super-long-acting basal insulin for people with diabetes dosed as a once-weekly injection. PE0139 is a fully mature, native insulin molecule (B and A chains) genetically fused to an elastin-like polypeptide (ELP) biopolymer. The compound is manufactured using PhaseBio’s ELP-based Escherichia coli expression system, where refolding of the drug occurs naturally in the cytosol, according to the company.

“This is essentially like a hexamer on steroids that will release slowly during the course of a week,” Cornell said. “We offer fixed-dose combinations of insulin with GLP-1 receptor agonists now, but those are daily. If we have a once-weekly insulin and you can combine that with a once-weekly GLP-1 receptor agonist, imagine that. To me, that is the exciting end product coming, in terms of the molecule.”

John Buse

The company has stated that the simple formulation facilitates coformulation with GLP-1 receptor agonists, including PhaseBio’s proprietary GLP-1, PB1023, and other GLP-1 receptor agonists.

“When you hear about weekly insulin, there are concerns about dosing, since it lasts so long in the body, so there are concerns about titrating it,” Isaacs said. “We have seen with insulin degludec, which has a long half-life, that has actually reduced incidence of hypoglycemia and allows for more flexibility with the dosing, where one does not have to worry about missed doses. Weekly insulin could present a good opportunity, especially when you look at the improved adherence we see with weekly GLP-1 receptor agonists. I’m excited about it.”

Optimizing diabetes devices

As novel insulin advances progress through the development pipeline, it is optimized technology — pumps, CGMs, hybrid closed-loop insulin delivery systems — that can offer the biggest improvements for the person with diabetes right now, Nathan said.

“At this point, it’s a refinement in the way we deliver insulin that is needed,” Nathan said. “Would faster insulins be better? Yes, but I believe the improvement would be marginal. We must complete the work that has already been started — putting more effort into delivery to make type 1 diabetes more safely treated and make those therapies more accessible for everyone.”

Hirsch agreed, noting that advances in closed-loop insulin delivery will continue to make all diabetes technology more automated.

“What we want to see in type 1 diabetes is everything automated, even automated for people who decide not to wear pumps,” Hirsch said. “The sensor tells the patient how much insulin to take, and the patient doesn’t have to be so precise with their carbohydrate counting. If you tell a smart pen you’re going to have a big meal, the pen knows how much insulin to give. Do I think that an ultra-ultra-fast acting insulin — faster than what we have on the market now — will make that even better? It might, but our current insulins are probably going to be good enough for most people.”


Hirsch said goals for the coming decade should focus on improving quality of life for people with diabetes while also working to improve insulin accessibility and cost.

“We don’t have to make a ‘perfect’ insulin,” Hirsch said. “If we have, as a target, a blood glucose that is same as that for a person without diabetes, we will never get there. On the other hand, getting a person with diabetes to have extremely good control with almost no hypoglycemia, a time-in-range of 80% to 90%, and an HbA1c almost always below 7%, those are much more doable goals for the next 10 years.” – by Regina Schaffer