Insulin is a protein. Like all protein (or peptide) drugs it needs to be injected, you can’t take it in pill form. This is because proteins are digested in the GI tract, are generally large and difficult to absorb intact, and if they make it that far they then have to deal with the liver before they can get to their intended target. So protein or peptide drugs are simply not able to be delivered orally.

The need to give daily injections is literally a pain, however. There has therefore been extensive research in the pharmaceutical industry into methods to allow for oral, transdermal, or mucous membrane administration of so-called biologicals. As this review states:

These problems can be overcome by adopting techniques such as chemical transformation of protein structures, enzyme inhibitors, mucoadhesive polymers and permeation enhancers.

A new study which is getting some media attention takes a completely different approach – they have created an autoinjector that you swallow, which then injects insulin into the stomach from the inside. The research is the result of collaboration between academics and industry:

Giovanni Traverso, an assistant professor at Brigham and Women’s Hospital, Harvard Medical School, and a visiting scientist in MIT’s Department of Mechanical Engineering, where he is starting as a faculty member in 2019, is also a senior author of the study. The first author of the paper, which appears in the Feb. 7 issue of Science, is MIT graduate student Alex Abramson. The research team also includes scientists from the pharmaceutical company Novo Nordisk.

This is a great example of exactly why academics benefit from their collaboration with industry, and should not be shamed for it as if it is necessarily nefarious.

In any case, the technology is quite impressive. The researchers previously made pills lined with multiple needles that were themselves made of compressed drug. The problem was that the drugs desolved in the stomach and most of it was destroyed before it could get into the stomach lining. In their new design they have a single needle, in this case made from compressed freeze-dried insulin. The shaft is made of a separate biodegradable material. The needle is attached to a spring held in place by a sugar-based material that dissolves in the stomach acid, releasing the needle which is supposed to shoot into the stomach lining where the insulin can dissolve and be absorbed into the bloodstream.

This is a neat idea, but presented some technical challenges. For example, they needed to make sure that the needle was oriented properly, so that it would shoot into the stomach wall. They were apparently inspired by the design of a leopard tortoise’s shell, which can be self-orienting. The shape of the capsule was designed to that it always turns upright in the same orientation, allowing for some control over the direction of the needle.

They have been testing this system in pigs, and have been able to inject 5mg of insulin in this manner, with the needle dissolving over one hour (the rate can apparently be controlled by how the insulin is compressed). This is in the therapeutic range. The system has to undergo human trials, which means it is still at least several years away. So far they have detected no harm from this delivery system, with the unused components passing through the GI tract without problems.

If this system pans out, which seems likely given what they have achieved so far, this could be a huge benefit for people who are dependent on protein or peptide based drugs (for quick background, a peptide is a chain of amino acids, while a protein is a folded peptide). Not having to stick yourself with a needle on a regular basis, causing chronic irritation of the skin and discomfort, would be a benefit to quality of life, and may be safer. Of course we need to see if there are any downsides to sticking small needles into the stomach every day over years, but remember the stomach is evolved to hold powerful acid, and is therefore quite robust.

There are also many drugs other than insulin that are protein or peptide based. Any hormone, enzyme, or monoclonal antibody is protein based. The system could also theoretically be used to deliver chemical drugs in order to get more rapid absorption. Regular GI absorption can be slow, taking 40 minutes or more from the time you take a pill. This is a long time if you have a debilitating migraine, for example, which is why nasal spray and subcutaneous versions of some migraine drugs are available. The pill injector could be another option, if the needles can be made to contain the desired drug and dissolve very quickly.

The major downside I predict for this system is expense. I suspect that this delivery system will be more expensive than traditional needles or injectors. Pharmaceutical companies will love it, because when their patent is expiring they can release this version of the drug and essentially get an extension on their patent for the new version. There will also actually be an advantage to the new system – but will insurance companies be willing to pay simply for patient convenience? So drug companies will search for some medical justification for the new delivery system, and doctors will be fighting with insurance companies to get coverage for their patients. This is a small price to pay for improved patient quality of life, but it’s unfortunate the system has to work this way. The end result is not bad, as everyone’s interests are eventually balanced out, but the system is not optimally efficient.

This is also a small window into why medical costs are increasing. Studies show the increase in health care costs is driven largely by technology. That technology does not always result in better hard outcomes (like survival), but may result in higher quality of life, which is harder to measure. We just have to decide as a society how health care quality of life we are willing to pay for.