Insulin aspart (NovoLog / NovoRapid)

Insulin aspart is the workhorse rapid-acting mealtime analog. It's made by Novo Nordisk under the brand names NovoLog (US) and NovoRapid (EU), approved in the EU in September 1999 and in the US around 2000.²¹ A single amino-acid swap (proline → aspartic acid at position B28) means it is absorbed faster than older "regular" human insulin after a subcutaneous injection.¹

Type and role. This is a rapid (mealtime/bolus) insulin — taken to cover the carbohydrate in a meal, not to provide background coverage. It is scored here against other rapid analogs, so its numbers are the genre's reference point rather than a basal-style flat curve.

PK/PD — the real numbers. In clamp studies comparing faster aspart with aspart, onset of action for the aspart arm is around 25 minutes and the insulin first appears in the blood at about 7 minutes.³ Plasma levels peak at a median of 40–50 minutes — about twice as fast as regular human insulin (80–120 minutes) — while the maximum glucose-lowering effect lands 1–3 hours after the dose.¹ Total duration is 3–5 hours, with an apparent half-life of about 81 minutes.¹ The roughly 80% bioavailability figure sometimes cited comes from faster-aspart studies and should not be presented as a conventional-aspart product estimate.⁴ The practical takeaway: even a "rapid" analog still lags a healthy pancreas, which is why post-meal spikes and the need to pre-bolus persist.⁶

Absorption variability. Onset and rate of absorption shift with the injection site, blood flow, temperature, and physical activity — the same dose can behave differently day to day.¹ Across head-to-head trials, mealtime aspart lowers post-meal glucose more than regular insulin (about −22 mg/dL in type 1 diabetes) and modestly improves HbA1c.⁵⁶ That HbA1c advantage, however, looks smaller under the most rigorous synthesis to date — see below.

A sobering synthesis. In June 2026 a Cochrane network meta-analysis pooled 15 randomised trials and 6,335 adults with type 1 diabetes on multiple daily injections, comparing aspart, lispro, glulisine, ultra-rapid lispro and faster aspart against plain regular human insulin. Its conclusion is deflating: on low-certainty evidence, all five analogues may produce little to no difference in HbA1c versus regular human insulin (aspart −0.14%, 95% CI −0.21 to −0.06). Some agents may cut severe hypoglycemia and diabetic ketoacidosis — ultra-rapid lispro scored best on severe hypoglycemia (RR 0.80) — but every one of those confidence intervals crosses 1, the certainty is rated low to very low, and there were no long-term follow-up data at all. Aspart did show a possible reduction in severe nocturnal hypoglycemia (RR 0.63, 95% CI 0.46 to 0.88), though on very-low-certainty evidence. Thirteen of the fifteen trials were industry-funded.⁹ None of this makes analogues a bad choice; their advantages in dose timing, flexibility and convenience are real. But it is an honest reminder that on the single number most people are judged by, the analogue era bought less than the marketing implied — and it is exactly why this site ranks insulins on speed, tail and flexibility rather than on HbA1c alone.

Exercise. Because the 3–5 hour tail keeps working during activity, circulating aspart can contribute to exercise lows; consensus guidance covers cutting the pre-exercise meal bolus (and/or adding carbohydrate) for moderate activity.⁷

Delivery and approvals. Aspart is delivered by pen, syringe, insulin pump (it is cleared for continuous subcutaneous infusion), and intravenously in clinical settings.¹ It is approved for type 1 and type 2 diabetes in adults and children — from age 1 in the EU.²

Access, cost, biosimilars. Aspart is a widely available analog, sold in the US and EU. Biosimilars have now arrived: the FDA approved Merilog (insulin aspart-szjj) in February 2025 as the first US aspart biosimilar, and Kirsty (insulin aspart-xjhz) followed as the first interchangeable rapid-acting insulin — meaning a pharmacist may substitute it for NovoLog depending on state laws.⁸

What's coming. The aspart molecule itself is mature; the frontier is making it faster. Novo Nordisk's faster aspart (Fiasp) adds niacinamide and L-arginine to speed early absorption, and newer ultra-rapid formulations push onset earlier still — all aimed at closing the gap between injected insulin and the body's own faster response.⁶ We therefore rank aspart as the reference point, not the frontrunner, on speed: it's the right zero-line for reading the insulin-speed gap, even as its access score climbs with biosimilar competition.