Like most cool biotechnology tools like this, the hangups will probably come from ironing out safety, dosing, and delivery issues in the clinic, rather than som

Like most cool biotechnology tools like this, the hangups will probably come from ironing out safety, dosing, and delivery issues in the clinic, rather than some mismatch in the biochemistry. They mention one of them in the paper: In the long term, a balance is assumed to be established between the amount of injected PLL-FPBA and cleared PLL-FPBA. Thus, the long-term toxicity of the complex formulation needs to be thoroughly evaluated for clinical translation. Proving that PLL-FPBA only responds to glucose fluctuation (and no other physiological changes, with a large enough sample size to cover the range of possible physiological changes that could be experienced by patients), proving that it is biocompatible with all patients, proving that the dose-response is uniform and can be correctly tailored in all patients, and then proving that all of these results hold over very long periods of time with material that can be cost-effectively manufactured at scale will probably be the tricky part. There probably aren't any fundamental reasons that the chemistry of the gluconic acid polymer wouldn't work in humans. Citations 13-40 provide a pretty good overview of the history of glucose-responsive insulins. These ideas are awesome and I hope they can be successful some day, but I also get the impression that engineering improvements (and cost reductions) in insulin pump technology will provide more immediate quality-of-life improvements for the broader Type-1 Diabetic population.