AT278 (U500 ultra-rapid insulin aspart)
Arecor Therapeutics
An investigational ultra-concentrated (500 U/mL), ultra-rapid mealtime insulin aspart from Arecor. In a Phase 1 type 1 diabetes clamp study it was absorbed faster than standard aspart despite being five times more concentrated, pointing to lower injection volumes for people with high insulin needs.
The scorecard
Mealtime convention (faster onset is better): in T1D, onset of glucose-lowering action was 10 min earlier than standard aspart and serum appearance 6 min earlier; faster than today's analogs but no head-to-head vs Fiasp.[1]
Mealtime convention (faster/earlier exposure is better): 4.0-fold higher insulin exposure in the first 30 min and 8.9-fold higher early glucose-lowering effect vs aspart, so action is front-loaded — though still far from the sub-20-min ideal.[1]
Mealtime convention (shorter tail is better): front-loaded exposure implies a relatively shorter tail, but the clamp study did not report a clear duration advantage, so this is scored conservatively.[1]
Only single-dose Phase 1 clamp data exist in a small all-male T1D cohort; day-to-day and real-world absorption variability is unproven, especially at 500 U/mL.[1]
A faster, shorter-acting bolus could in principle reduce residual insulin during activity, but no exercise or hypoglycemia-flexibility data are available; scored neutral.[1]
Access convention (cheaper/more available is better): not approved anywhere, no price, no marketed product — effectively inaccessible today.[2]
Insulins are scored relative to their role peers (see tags: rapid, ultra-rapid, basal, inhaled). A basal insulin's onset score compares it to other basals, not to mealtime insulins.
The full picture
AT278 is an investigational mealtime (bolus) insulin from Arecor, built on its Arestat formulation technology. It is unusual in two ways at once: it is ultra-concentrated at 500 U/mL — five times the strength of standard insulin aspart — yet engineered to be absorbed faster, not slower. Normally, concentrating insulin slows absorption, so combining high strength with rapid onset is the technical achievement.
The key evidence comes from a Phase 1, double-blind, randomized crossover clamp study in 38 men with type 1 diabetes, comparing a single dose of AT278 against standard insulin aspart. AT278 reached the bloodstream about 6 minutes earlier and began lowering glucose about 10 minutes earlier. Early exposure was 4.0-fold higher in the first 30 minutes, and the early glucose-lowering effect was 8.9-fold higher in the first 30 minutes — a strongly front-loaded profile that more closely mimics a meal-time insulin burst.
The practical promise is twofold: faster post-meal coverage, and a much smaller injection volume for people with high insulin requirements, including pump users. The limitations are real — only single-dose Phase 1 data exist, in a small all-male cohort, and no Phase 2 trial is yet registered on ClinicalTrials.gov.
But AT278 is no longer an orphan asset. In September 2025 Arecor signed a co-development agreement with the US pump maker Sequel Med Tech, each company committing up to $1.3 million to the work needed to pair AT278 with Sequel's twiist automated insulin delivery system — FDA interactions, clinical trial batch manufacturing, and pump-compatibility testing. As of April 2026 Arecor reports positive feedback from a Type C meeting with the FDA on its Phase 2 study design, says it is targeting a Phase 2 start in the second half of 2026, and is negotiating a broader co-development and commercialisation deal.
Be precise about what that does and does not mean. Positive Type C feedback is not an approval and not an IND clearance — as of the April 2026 update the IND had not yet been filed. The planned study is a roughly six-week crossover against NovoLog in fewer than 100 people with high daily insulin requirements, and it is designed to enrol both type 1 and type 2 diabetes, so it is not a T1D-only trial. A concentrated, ultra-rapid insulin built specifically to run inside a closed loop is close to what the insulin-speed problem actually calls for — but it still has to survive a Phase 2 that has not started.
Coming soon
ETA · Phase 1 complete in type 1 diabetes (2023). Arecor signed a co-development agreement with pump maker Sequel Med Tech in September 2025 to combine AT278 with the twiist automated insulin delivery system, and reports positive FDA feedback on the Phase 2 design. It is targeting a Phase 2 start in 2H 2026 and negotiating a broader co-development and commercialisation partnership. No Phase 2 trial is registered on ClinicalTrials.gov as of 14 July 2026, and there is no approval anywhere.
- →Phase 2 targeted for 2H 2026, pairing AT278 with Sequel Med Tech's twiist automated insulin delivery system — not yet registered on ClinicalTrials.gov
- →Broader co-development and commercialisation partnership under negotiation as of April 2026
- →If developed, the 500 U/mL strength would cut injection volume ~5-fold for people on large doses or in pumps
Sources
- [1]Svehlikova E, et al. Pharmacokinetics and Pharmacodynamics of a Novel U500 Insulin Aspart Formulation: A Randomized, Double-Blind, Crossover Study in People With Type 1 Diabetes. Diabetes Care 2023;46(4):757-764 · peer-reviewed · 2023-04-01
- [2]Arecor Therapeutics — Diabetes portfolio (AT278 ultra-concentrated ultra-rapid insulin) · manufacturer
- [3]Arecor announces publication of Phase 1 data for AT278 in Diabetes Care · news · 2023-01-30
- [4]Arecor announces Co-development Agreement with US Insulin Pump Device Company (Sequel Med Tech) for AT278 · manufacturer · 2025-09-25 — Each company commits up to $1.3M to Phase 2 trial-enabling activities (FDA interactions, clinical trial manufacturing, twiist compatibility). An IND filing is anticipated on completion, with a Phase 2 possible in 2H 2026.
- [5]Arecor targets broader AT278 partnership as phase II trial nears · news · 2026-04-13 — Reiterates positive feedback from a September 2025 Type C meeting with the FDA on the Phase 2 study design. This is neither an approval nor IND clearance.