Wednesday, May 4, 2022

NN1845 (a glucose sensitive insulin) Starts A Second Phase-1 Clinical Trail

One approach to creating a "practical cure" for T1D is sometimes called "smart insulin", "glucose sensitive insulin", or "glucose responsive insulin".  This is insulin which is inactive if blood glucose numbers are low, but becomes active when those numbers are high. A person with T1D would inject a fixed amount of this insulin every day (or maybe every week), and it would become active only as needed.

This would not truly "cure" T1D, but it would result in a disease which did not require measuring blood glucose, counting carbs, or changing insulin doses based on food, exercise, hormones, etc.  It would make treating T1D much more like treating high blood pressure.  Some people (myself included) would consider this a practical cure.

Several different companies have attempted to create glucose responsive insulins, and several university researchers are also working on it.  However, most are still in animal testing.  Only one is being tested on people.  That one is being developed by Novo Nordisk and is called NN1845  (it used to be called NNC0363-0845).

The situation here is a little unusual in that Novo Nordsk has already run one phase-I clinical trial aimed very narrowly at safety and tolerability, and has recently started a second phase-I clinical trial, aimed at how well it works.

Results from the Previous Phase-I Clinical Trial

This is the summary of research results as published in a Novo Nordisk update:

During the third quarter of 2021, Novo Nordisk completed a phase 1 trial investigating safety, tolerability, pharmacokinetics and pharmacodynamics of subcutaneously administrated glucose-sensitive insulin (NN1845). In the trial, NN1845 appeared to have a safe and well-tolerated profile and demonstrated proof of principle of glucose-sensitive properties. Further development of glucose-sensitive insulin to optimize pharmacokinetic properties is now being evaluated.

First Phase-I Clinical Trial Registry:

I want to stress that this is not a publication in scientific literature, and it does not include any actual data, just English phrases.  However, they are starting a follow-on clinical trial, which shows that they are optimistic about the results, and want to move forward with the research.

Plans for the Next Phase-I Clinical Trial

They will recruit 30 adults who have had T1D for more than a year.  Participants will receive the "smart insulin" once every 4 hours up to 6 times daily for 3 days.  The study is blinded and has a cross over design.  Each person will be treated with the experimental insulin and Levemir® (at different times) and the results compared.

They will be followed about 6 weeks, up to a maximum of 14 weeks.  The primary end point is blood glucose levels in the hours after a simulated meal.  Secondary outcomes include adverse side effects and several additional measures of blood glucose levels during different time periods after a simulated meal.

This study is recruiting in Graz, Austria:
Contact: Novo Nordisk     (+1) 866-867-7178     

Second Phase-I Clinical Trial Registry:

Discussion and Opinions

I like the fact that this study should be quick.  Classic cure-focused trials usually gather data for 2 years, but these researchers will gather data for 14 week at most.  Also, since they are recruiting people with established T1D, it should be easier and quicker to recruit the 30 people they need.  Together, these should lead to quick results.

Two things I don't like about this trial are the use of Levemir® and the lack of any background or fasting data.  Levemir is long acting insulin.  But is that the right comparison for effects after a meal?  It depends on Novo Nordsk's goal with this insulin.  If their goal is a better, safer long acting insulin, then this is a good comparison.  However, that is not a practical cure.  A practical cure would require the glucose sensitive insulin to cover a meal much like fast acting insulin.  I'm worried that this comparison telegraphs Novo Nordsk's opinion that this is not a practical cure, but just a better long acting insulin. 

Of course, the optimist in me thinks that even if this particular "smart insulin" is not fast enough to be a practical cure, maybe the next one will be.  Or maybe an improvement to this one will make it faster in the future.

Both of these studies are phase-I so an obvious question is: how are they different?  Let me summarize the major differences:  

  • The biggest difference is in end points.  In the first study people were only monitored for adverse effects and to see how the experimental insulin is absorbed by the body.  In the second study most of the monitoring is aimed at blood glucose results.
  • The first study includes 68 people in three parts, and each part is quite different.  The second study is 30 people all together, which is a much more common design.   Phase-I studies in T1D often have 10-20 people, so both of these studies are big for phase-I.
  • In the first study, people get treated once with the experimental insulin and once with the control.  In the second study, people will be treated for 3 days and given the control for 3 days.

Obviously, I'm excited to see this research progressing.  Unfortunately, because the research is run by a commercial company, there is no requirement that they publish results quickly, and so far, they have not published at all.  Therefore there is no way to know the results from the first phase-I study.  They think there is a chance of a product there (commercial success) and the evidence is strong enough for them to put more money into the research, to fund a second phase-I trial.  So that suggests good results, but there is no way to independently confirm it.

The Real Publication Requirement

A lot of people seem to think there is some rule that says you must publish the results of clinical trails.  Twenty years ago, the answer was simple: no.  There was no law saying anyone had to publish anything.  Sure, academics built their reputations by publishing, so there was pressure for them to publish.  But for commercial companies, there was no reason to publish anything that would not increase their profits.  Then the USA passed a law saying that any clinical trial that was part of a new drug application must be listed on the FDA's clinical trial registry.  That was enforced as part of drug approvals, so suddenly there was a trial registration requirement, but no requirement to publish results.  (Companies had to submit results for FDA approval, but not publish them in the scientific literature or make them available to anyone else.)  

Years later the law was updated to require results be added to the clinical trial registry.  However, there was no enforcement, and therefore few researchers did.  Even researchers who published results rarely added them to the clinical trial registry.  Years later, a news service published a series of articles describing the legal requirement, the fact that it was widely ignored, that there was no enforcement, and that most of this research was funded by US taxpayers via government agencies.  Finally, there was some movement.  Over the next few years many old clinical trial registries were updated with results, and now more results are posted when the studies are completed.  But the number is still low.

So the summery right now, in the US, is that there is a paper rule that researchers must make results of clinical trials public, but, in fact, this data is not available for many clinical trials, including many focused on T1D.  I think the situation is worse in the rest of the world, but I've never investigated in depth.

For example, in 2015 none of the research results previously funded by JDRF were available in the FDA's clinical trial registry.  The publication of this information by Statnews resulted in several JDRF funded clinical trials posting results to the clinical trial registry, in some cases years after the research finished.  I'm sure JDRF funded researchers are doing better now, but I don't have any more recent data.

As of 2019, the two local (to me) universities who do the most T1D research were University of California San Francisco (reporting less than 50% of results) and Stanford (reporting less than 75%).  These numbers are for all the clinical trials done at those universities, not just their T1D trials.

More to read:

Joshua Levy
publicjoshualevy at gmail dot com
All the views expressed here are those of Joshua Levy, and nothing here is official JDRF or JDCA news, views, policies or opinions. My daughter has type-1 diabetes and participates in clinical trials, which might be discussed here. My blog contains a more complete non-conflict of interest statement. Thanks to everyone who helps with the blog.