Ultra-Rapid Lispro: What is it and why is it interesting?

In the second article in this series looking at ultra-rapid insulin, we talk about Lilly’s new kid on the block:

Ultra-Rapid Lispro (URLi)

Ultra-Rapid Lispro is Lilly’s new formulation of Lispro to counter Novo’s Fiasp. In a similar fashion to Novo, they’ve added excipients to the Lispro formulation in order to speed the absorption. Specifically:

So what we have is a new, injectable insulin that will arrive around three years after Fiasp, using different excipients that in theory speed it up, significantly. 

But by how much? Well the clinical trials have proven very interesting.

Clinical Trials data

URLi has been undertaking phase III trials, and the data presented at EASD 2019 was from the injectable trials – the pump trials are currently ongoing. The data? Well, it’s certainly interesting…

The PRonto-T1D trial was a 52-week study of MDI users and threw up some excellent results. But first, the numbers quoted by Lily are, in comparison to Lispro (Humalog):

All this adds up to something significantly faster than previous insulins. But how much?

Referring to the slides from a second presentation, we can see that early absorption is significantly faster than Fiasp:

50% of max concentration was reached in two thirds of the time of Fiasp, while exposure increase in the first 15 minutes versus Fiasp was 1.5 times. Compared to the older analogues, it was obviously significantly faster. 

The other finding of note was the tail:

It appears that we now have an insulin that really does have a tail that is all but done after five hours, instead of the 6-8 that all the others have.

It’s also really noticeable how much less insulin there is in the tail after two hours. Okay, it’s not approaching the levels of Afrezza, but this is notably less than there has been in previous formulations of subcutaneous insulin.

Between them, the data from these trials shows something that is potential much more useful for use in a closed loop system. Okay, it’s not clearing at the rate of endogenous insulin, but it is quicker. 

What about site reactions?

In the Pronto-T1D study, there were reports of injection site reactions (a number of 2.7% was given) but alongside that was the statement that these were mostly mild and none led to treatment being discontinued. The question on every pumper’s lips is how this translates into the CSII trial.

Will there be significantly more irritation and will it affect the efficacy of the insulin? We’ll have to wait for those outcomes to find out.

Why do you care about this?

It’s those “on and off” numbers. If you’ve ever seen Dana Lewis speak, you’ll have heard her talk about not bolusing for 2 years, using Fiasp with OpenAPS. For those of us who have issues with Fiasp, an alternative is welcome, and this seems to be an alternative. 

What’s perhaps even more impressive is that the time this takes to have an effect is notably faster than that of Fiasp. In using OpenAPS/oref1 terms, this means that if you simply eat, the unannounced meal detection and SMB functions will be more effective in managing post prandial glucose levels than they have ever been. The additional speed of onset suggests that even with high carb meals, we may be able to simply eat and not bolus.

With this insulin, it’s possible that there may be a new bar for enabling true closed loop systems that require no meal announcement, and I for one would love to give it a try as soon as it’s available. 

That, of course, is the other question. According to press reports, Lilly has submitted to the FDA for approval. I’ve not seen anything about the EMA though. 

Suffice to say that right now, I’m excited about the prospects for URLi in closed loops, both DIY and commercial. Right now, the commercial ones won’t be able to update the models for insulin absorption to keep up. DIY however, will be able to. Maybe there won’t be too much irritation at sites, either…

With this update, maybe the dream of never bolusing or counting carbs again, for any food, will at last become a reality for DIYAPS users. And we all can dream…


  1. While exciting, doesn’t this have two very similar concerns to Fiasp? Similar mechanism for increasing absorption (vasodilation), and no study on pump users. As someone who had 2 wonderful months with Fiasp, then two ok ones, than one month of “Fiasp as water”, I’m very wary.

    • Which is commented on in the post. The issue with Fiasp doesn’t seem to be the Vasodilation, rather the mechanism by which this is achieved, namely Nicotinamide. This seems to cause a reaction in some pump users, either topically at the pump site, or, in my view, on the liver.

      URLi uses a different excipient that may not have an effect, but as I also mentioned there is an ongoing longer term trial with pump users that should answer these questions for us, which wasn’t undertaken for Fiasp.

      • Ah, now I see! This is very exciting. My first two months on Fiasp were pretty great, with one exception of a very bothersome persistent low one day which I blamed on the long tail (which URLi seems to improve a lot as well). Looking forward to the pump study and your thoughts on it.

      • What effect does it have on the liver? I had similar problems to you pumping it, but I’ve found it much more reliable on MDI.

    • Hey just wanted to give you a heads up that the ultra-rapid lispro has already been approved in japan and will be available from tomorrow on June 17th!! I’ll probably post a report sometime next month on its use with loop, in the cgm in the cloud off topic group in Facebook.

  2. Not true for Tandem Diabetes’ software update capability: “the commercial ones won’t be able to update the models for insulin absorption to keep up. DIY however, will be able to.”

    • Good point. However to introduce a new insuline profile into the Control-IQ software, I would assume that it would require clinical trials and an FDA sign off before being distributed, so it would coome at some cost.

  3. Lilly can thank Adocia’s work for this product. To know the subject, it is primarily citrate that makes lispro faster, the excipients used to stabilize the entire formulation to meet the regulatory requirements for the stability of the product. In the future, when you use this product, remember that this product was born of intellectual property theft. In any case, it’s a good news for T1D, I hope it will be useful for you

  4. According to this article (https://www.pharmazeutische-zeitung.de/ultraschnelles-insulin-zugelassen-116915/ ) the new ultra rapid lispro is now approved by the EMA and will be marketed under the brand name “Liumjev“.

    There has been a trial testing the new insulin vs the old lispro in pumps (https://www.clinicaltrials.gov/ct2/show/results/NCT03433677). If I interpret the results correctly then there were way more events of infusion site reactions and of infusion site pain under the treatment with the faster insulin.

      LY900014   Insulin Lispro (Humalog@)  
      Affected / at Risk (%) # Events Affected / at Risk (%) # Events
    Total 18/49 (36.73%) 6/48 (12.50%)  
    Infusion site pain 9/49 (18.37%) 11 2/48 (4.17%) 2
    Infusion site reaction 9/49 (18.37%) 36 4/48 (8.33%) 5

4 Trackbacks / Pingbacks

  1. Revisiting Fiasp
  2. Lilly-jev novi superbrzi inzulin na korak do tržišta EU! - NaInzulinu.com
  3. Long-term CGM analysis
  4. Full closed loop – Lyumjev + oref1 | Diabettech – Diabetes Wisdom

Leave a Reply

Your email address will not be published.