Insulin + pramlintide (amylin) closed loop

Most artificial-pancreas work uses one hormone — insulin — to push glucose down. This approach adds a second hormone, pramlintide, a lab-made copy of amylin, which a healthy pancreas releases from the same cells as insulin at every meal and which people with T1D lose almost entirely.¹ Pramlintide does not raise glucose the way glucagon does; instead it slows stomach emptying, blunts the after-meal glucagon surge, and increases fullness, so the meal-time spike is smaller and slower.¹ In a closed loop, that smoother rise is easier for the insulin algorithm to chase — the goal is steadier glucose with less reliance on perfect carb counting.²

What the trials show. A McGill randomized crossover trial found that a rapid-insulin-plus-pramlintide artificial pancreas raised time-in-range from 74% to 84% versus insulin alone, driven mostly by better daytime control (63% to 78%), without any increase in hypoglycemia.³ A follow-on McGill trial showed the pramlintide system could replace carb counting with a simple meal announcement while keeping glucose control as good as carb-counting alone.⁴⁵ A fully-closed-loop version with no meal input held a high time-in-range and produced fewer low-glucose events than the insulin-only loop (one third vs over half of participants had a low).⁶ The recurring cost is gut side effects — mild-to-moderate nausea or bloating in a meaningful minority.³⁴

The biggest test so far came back flat. At the American Diabetes Association's 86th Scientific Sessions in June 2026, McGill presented the largest and longest trial of this approach to date: 26 adults, randomized crossover, three-week arms, comparing a fully automated insulin-lispro-aabc (Lyumjev) plus pramlintide loop at fixed ratios of 8 and 10 µg per unit against a hybrid loop where users counted carbohydrates.⁷⁸ An earlier outpatient pilot of the same system had looked promising.⁹ Time-in-range was 71.1% and 71.2% on the pramlintide arms versus 72.1% on the carb-counting control — the pramlintide loop did not meet the trial's non-inferiority criteria.⁷ Time below 70 mg/dL (3.9 mmol/L) was similar across all three arms, so the amylin loop bought no hypoglycemia advantage either.⁷ What it did buy was freedom from carb counting for a time-in-range about one point lower — close, but the trial did not establish that the pramlintide loop is no worse — plus a lower daily insulin dose (about 42.8 and 41.3 units vs 47.4, p<0.001).⁷ Gastrointestinal side effects, all mild to moderate, were the price: 7 and 8 of the 26 participants versus none on the control loop, with two people stopping because of pramlintide nausea.⁷ The one bright spot: in a subgroup of 10 participants whose time-in-range was under 70% on the control arm, the pramlintide loop did help (67.9% vs 60.6%, p=0.011) — which is where the investigators now say the approach should be tested.⁷ Two caveats worth holding onto: this was a conference presentation, not a peer-reviewed publication, and 26 people is a small trial.

Glucagon vs amylin. This is the key contrast with glucagon-based dual-hormone systems. Glucagon actively raises glucose, giving the loop a brake against lows, but it is chemically unstable in a pump. Pramlintide doesn't defend against lows directly — it makes the whole post-meal curve gentler, so the loop has less to fight.¹

The formulation challenge. Pramlintide and insulin cannot share a reservoir: pramlintide is formulated at an acidic pH, insulin near neutral, and the two aggregate and lose activity when mixed, which is why Symlin must be a separate injection.¹⁰¹¹ Today's research systems therefore use two pumps or two pods — one for each hormone.² A major engineering effort aims to fix this: Stanford researchers built a stable insulin-pramlintide co-formulation that resisted aggregation for over 16 hours,¹² and the company Adocia has tested a fixed co-formulation (ADO09, "M1Pram") in people, improving post-meal glucose and time-in-range with less mealtime insulin.¹³ A single stable mixture is what would let this run from one pump.

That commercial path has now stalled. On 3 July 2026 Adocia announced that Sanofi's three-year exclusive right to negotiate a partnership on M1Pram — an option Sanofi paid for in 2023 but never exercised — had expired, and that Adocia, keeping global rights, has put M1Pram development on hold given its current priorities, while saying it will keep evaluating the path forward.¹⁴ Adocia pointed to a treatment landscape reshaped by GLP-1 drugs rather than to any failure of the formulation itself. On hold is not the same as abandoned — but the most clinically advanced insulin-pramlintide co-formulation now has no sponsor pushing it, and the remaining active co-formulation work is academic and preclinical.

Maturity and access. The drug itself is mature — pramlintide was FDA-approved as Symlin in 2005 as a subcutaneous adjunct to mealtime insulin in T1D, and carries a boxed warning for severe insulin-induced hypoglycemia.¹⁵ But the automated closed-loop co-delivery is investigational: there is no approved insulin-pramlintide artificial pancreas, and it is reachable today only through studies (for example OHSU's "iPancreas" program, testing insulin-plus-pramlintide on an Omnipod platform with a Dexcom sensor).¹⁶

What's coming — and what isn't. Both of the threads this page used to point to have changed direction in the past month. The McGill trial that was the one to watch has read out, and it missed;⁷⁸ the investigators' own conclusion is that the next study should be longer and enrolled in people whose time-in-range is below 70%, the group that did benefit — and no such trial is registered yet. McGill's fully-closed-loop effort has meanwhile pivoted away from amylin toward incretins, with a 105-person fully-closed-loop trial using tirzepatide (registry ID NCT07284511) starting in May 2026. On the second thread, the single-reservoir co-formulation — the thing that would shrink a two-pump rig to the footprint of a normal AID system — has lost its commercial sponsor with Adocia's July 2026 hold on M1Pram,¹⁴ leaving academic, preclinical work (Stanford's supramolecular co-formulation) as the live route.¹² What remains genuinely in progress is OHSU's "iPancreas" study, pairing pramlintide delivery with meal-start prediction software on an Omnipod platform.¹⁶ The honest read in mid-2026: amylin remains a scientifically sound idea for flattening meals and removing carb counting, but it has not yet shown it can beat a good insulin-only hybrid loop, and the engineering that would make it practical has slowed down.

References

1. Pullman J, Darsow T, Frias JP. Pramlintide in the management of insulin-using patients with type 2 and type 1 diabetes. Vascular Health and Risk Management (2006). According to PubMed, PMID 17326327. https://doi.org/10.2147/vhrm.2006.2.3.203 ↩ ↩² ↩³

2. Haidar A, Tsoukas MA, Bernier-Twardy S, et al. A Novel Dual-Hormone Insulin-and-Pramlintide Artificial Pancreas for Type 1 Diabetes: A Randomized Controlled Crossover Trial. Diabetes Care (2020). According to PubMed, PMID 31974099. https://doi.org/10.2337/dc19-1922 ↩ ↩²

3. Haidar A, Tsoukas MA, Bernier-Twardy S, et al. A Novel Dual-Hormone Insulin-and-Pramlintide Artificial Pancreas for Type 1 Diabetes: A Randomized Controlled Crossover Trial. Diabetes Care (2020). According to PubMed, PMID 31974099. https://doi.org/10.2337/dc19-1922 ↩ ↩²

4. Tsoukas MA, Cohen E, Legault L, et al. Alleviating carbohydrate counting with a FiASP-plus-pramlintide closed-loop delivery system (artificial pancreas): Feasibility and pilot studies. Diabetes, Obesity & Metabolism (2021). According to PubMed, PMID 34047449. https://doi.org/10.1111/dom.14447 ↩ ↩²

5. Cohen E, Tsoukas MA, Legault L, et al. Simple meal announcements and pramlintide delivery versus carbohydrate counting in type 1 diabetes with automated fast-acting insulin aspart delivery: a randomised crossover trial in Montreal, Canada. The Lancet Digital Health (2024). According to PubMed, PMID 38906614. https://doi.org/10.1016/S2589-7500(24)00092-X ↩

6. Tsoukas MA, Majdpour D, Yale JF, et al. A fully artificial pancreas versus a hybrid artificial pancreas for type 1 diabetes: a single-centre, open-label, randomised controlled, crossover, non-inferiority trial. The Lancet Digital Health (2021). According to PubMed, PMID 34580055. https://doi.org/10.1016/S2589-7500(21)00139-4 ↩

7. Doumat J, et al. (McGill University), presented at the American Diabetes Association 86th Scientific Sessions, New Orleans, 5–8 June 2026. As reported by Healio: "Fully closed-loop system with insulin lispro plus pramlintide may benefit type 1 diabetes" (6 July 2026). Conference presentation — not yet peer-reviewed or published. https://www.healio.com/news/endocrinology/20260706/fully-closedloop-system-with-insulin-lispro-plus-pramlintide-may-benefit-type-1-diabetes ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷

8. ClinicalTrials.gov. NCT06046417 — A Fully Automated Lyumjev and Pramlintide Delivery System for Adults With Type 1 Diabetes (McGill University Health Centre; 26 adults; three-week arms; pramlintide at 8 and 10 µg/unit vs a carb-counting hybrid loop; primary endpoint % time in 3.9–10.0 mmol/L). Registry record. https://clinicaltrials.gov/study/NCT06046417 ↩ ↩²

9. Odabassian M, Tsoukas MA, Cohen E, et al. A Pilot Outpatient Assessment of a Fully Closed-Loop Insulin and Pramlintide System. Journal of Diabetes Science and Technology (2025). According to PubMed, PMID 41174925. https://doi.org/10.1177/19322968251371046 ↩

10. Maikawa CL, Smith AAA, Zou L, et al. A co-formulation of supramolecularly stabilized insulin and pramlintide enhances mealtime glucagon suppression in diabetic pigs. Nature Biomedical Engineering (2020). According to PubMed, PMID 32393892. https://doi.org/10.1038/s41551-020-0555-4 ↩

11. Maikawa CL, Smith AAA, Zou L, et al. A co-formulation of supramolecularly stabilized insulin and pramlintide enhances mealtime glucagon suppression in diabetic pigs. Nature Biomedical Engineering (2020). According to PubMed, PMID 32393892. https://doi.org/10.1038/s41551-020-0555-4 ↩

12. Maikawa CL, Smith AAA, Zou L, et al. A co-formulation of supramolecularly stabilized insulin and pramlintide enhances mealtime glucagon suppression in diabetic pigs. Nature Biomedical Engineering (2020). According to PubMed, PMID 32393892. https://doi.org/10.1038/s41551-020-0555-4 ↩ ↩²

13. Andersen G, Eloy R, Famulla S, et al. A co-formulation of pramlintide and insulin A21G (ADO09) improves postprandial glucose and short-term control of mean glucose, time in range, and body weight versus insulin aspart in adults with type 1 diabetes. Diabetes, Obesity & Metabolism (2023). According to PubMed, PMID 36633505. https://doi.org/10.1111/dom.14972 ↩

14. Adocia. "ADOCIA Announces the Expiration of Exclusive Right to Negotiate Partnership for M1Pram with SANOFI" — press release stating Adocia retains global rights and has "decided to put on hold the development of M1Pram". Business Wire (3 July 2026). https://www.businesswire.com/news/home/20260703008079/en/ADOCIA-Announces-the-Expiration-of-Exclusive-Right-to-Negotiate-Partnership-for-M1Pram-with-SANOFI ↩ ↩²

15. U.S. Food and Drug Administration. SYMLIN (pramlintide acetate) injection — full prescribing information, including boxed warning for severe insulin-induced hypoglycemia (NDA 021332, original approval 2005). FDA Drugs@FDA (2005). https://www.accessdata.fda.gov/drugsatfda_docs/label/2005/021332lbl.pdf ↩

16. Oregon Health & Science University / Breakthrough T1D Pacific Northwest. T1D Insulin and Pramlintide Dual-Delivery Study at OHSU ("iPancreas"): a closed loop delivering insulin and pramlintide via Omnipod pods with a Dexcom CGM, paired with a meal-forecasting algorithm. Breakthrough T1D (16 September 2024). https://www.breakthrought1d.org/pacificnorthwest/2024/09/16/t1d-insulin-and-pramlintide-dual-delivery-study-at-ohsu/ ↩ ↩²