Research · Single-study breakdown
An oral GLP-1 added to insulin: what the orforglipron trial actually showed
What this means for you
If you are on insulin and your A1c is still above target, this is an early look at a future option that would be a daily pill rather than another shot. In the trial it lowered blood sugar meaningfully and helped with weight — and it did not raise the rate of clinically significant low blood sugar, which is usually the thing insulin users worry about most when adding a second medicine.
The honest caveat: in the US, orforglipron (brand name Foundayo) is FDA-approved only for weight management in adults with obesity, or overweight plus a weight-related condition. It is not FDA-approved for type 2 diabetes, and Lilly’s diabetes application is still under review. The UK approved it for type 2 diabetes in August 2026, and EU regulators recommended approval in October 2026. Adding it to insulin is what this trial tested; in the US that is not an approved use. This is trial evidence, not an approved label — a reason to ask your doctor what is coming, not a reason to change anything you are currently doing.
If you try it eventually, expect some nausea or other stomach upset early on, which is why the dose is raised slowly. Nothing here means you should adjust your insulin or any other medication on your own.
How the trial was run
ACHIEVE-5 was a phase 3, multicenter, randomized, double-blind, placebo-controlled trial run across 72 sites in the US, Brazil, China, Japan, and Romania. It enrolled 546 adults with type 2 diabetes whose blood sugar was inadequately controlled on titrated insulin glargine, some also taking metformin, an SGLT2 inhibitor, or both. Participants were randomized 1:1:1:1 to orforglipron at 3 mg, 12 mg, or 36 mg, or to placebo, all added on top of their existing insulin. The trial used a test capsule; on the approved tablet these doses correspond to 2.5 mg, 9 mg and 17.2 mg. Treatment ran 40 weeks, with the dose escalated gradually — starting at 1 mg and stepping up every 4 weeks to the assigned target.
At baseline the groups looked typical for this population: mean A1c of 8.50%, mean body weight of 85.2 kg, mean BMI of 30.8, and a median of 14.6 years living with diabetes. Of the 546 randomized, 507 (92.9%) completed the trial. Stopping the study pill was more common at the two higher doses — 14.4% at 12 mg and 18.4% at 36 mg, compared with 9.2% on placebo (7.3% at 3 mg). Most people who stopped the pill stayed in the trial.
What the data showed
The primary measure was A1c change at week 40. Orforglipron lowered A1c by −1.58%, −1.88%, and −1.82% at the 3 mg, 12 mg, and 36 mg doses, compared with −0.79% on placebo. The treatment differences versus placebo — −0.78%, −1.08%, and −1.03% — were all statistically significant (P < .001). The two higher doses were the primary endpoint and the 3 mg dose a key secondary; all three beat placebo.
The proportion of participants reaching an A1c under 7.0% tracked the same way: 57% at 3 mg, 70% at 12 mg, and 65% at 36 mg, versus 25% on placebo. Body weight fell by −2.6%, −4.8%, and −5.4% across the three doses, against a slight gain of +0.2% on placebo.
Two patterns are worth naming. First, the A1c benefit roughly plateaus between 12 mg and 36 mg — going to the top dose bought very little additional A1c lowering — while weight loss kept increasing with dose. Second, and the reason this matters clinically: orforglipron did not raise the rate of clinically significant low blood sugar compared with placebo. That is the question that matters most when a glucose-lowering drug is added to insulin, and here the answer was reassuring. Severe lows were rare: three people on orforglipron and none on placebo, too few to draw conclusions from.
What this study can't tell us
- 40 weeks is short — durability and long-term safety beyond about nine months are unknown.
- No head-to-head versus injectable GLP-1s such as semaglutide or tirzepatide, so we can’t say it matches those.
- GI side effects were the most common adverse events, and stopping the pill because of side effects was more common at the higher doses; real-world tolerability may differ.
- Not FDA-approved for type 2 diabetes — in the US it is approved only for weight management, and adding it to insulin is not an approved use.
- Industry-funded (Eli Lilly), as most phase 3 drug trials are.
The analyst take: the headline isn’t the size of the A1c drop, which is in the range you’d expect from a GLP-1. It’s the delivery. Orforglipron is a nonpeptide, small-molecule pill taken once daily with no food or water restrictions. Almost every GLP-1 in wide use today is an injection, and the friction of weekly shots is a real barrier to both starting and staying on therapy. A pill with no dosing rules that holds up in a phase 3 trial is meaningful for access and adherence in a way the raw percentages don’t capture.
Who is this for right now? In the US it is not yet approved for type 2 diabetes, though it is available for weight management; in the UK it is already approved for type 2 diabetes. It is most relevant to people already on insulin who are still above target and would rather not add an injection, and to the clinicians who treat them. The right move today is to watch where the regulatory process goes, not to act on a single 40-week study. It is one of two notable GLP-1 readouts from the same week — see how it fits alongside the mazdutide obesity trial.
Sources
- Orforglipron Added to Titrated Insulin Glargine in Type 2 Diabetes: The ACHIEVE-5 Randomized Clinical Trial — JAMA (2026) · doi:10.1001/jama.2026.9512
- ACHIEVE-5 trial registration (NCT06109311) — ClinicalTrials.gov
- Mazdutide and Orforglipron—New Evidence in Obesity and Diabetes (Editorial) — JAMA (2026) · doi:10.1001/jama.2026.10443
Not medical advice. Educational summary of published research. Last reviewed Oct 27, 2026.