Orforglipron blood pressure falls most in stage 2 hypertension in ATTAIN analyses

Eli Lilly's daily pill orforglipron lowered systolic blood pressure more than placebo over 72 weeks in two Phase 3 obesity trials, according to post hoc analyses shown at AHA 2026 in Chicago. The orforglipron blood pressure data, drawn from ATTAIN-1 and ATTAIN-2, showed the largest drops in people who started with stage 2 hypertension. Weight loss explained most of the effect, but not all of it. The drug is already approved in the US as Foundayo, so its stage does not change, and the meaning for the peptide side of the GLP-1 race is narrower than the headline.

What the AHA 2026 analyses tested

ATTAIN-1 enrolled 3,127 adults with obesity but no diabetes and randomly assigned them to one of three once daily orforglipron doses or to placebo, on top of advice on diet and physical activity. ATTAIN-2 used the same doses in 1,613 adults with type 2 diabetes, a body mass index of at least 27 and HbA1c between 7% and 10%. Both trials ran for 72 weeks and supported the FDA approval in April 2026 for chronic weight management. The main ATTAIN-1 paper had already listed systolic pressure among the risk factors that improved, along with waist size, triglycerides and non HDL cholesterol. The new abstract, with Luke Laffin of the investigator team listed first, returns to both trials with a narrower question about blood pressure alone.

The investigators tracked office systolic and diastolic readings over the 72 weeks and compared the arms with a mixed model for repeated measures. They also sorted participants by baseline category under ACC and AHA definitions: normal, elevated, stage 1 hypertension and stage 2 hypertension, the last meaning at least 140 systolic or 90 diastolic. A mediation model then estimated how much of the systolic change ran through weight loss. The orforglipron blood pressure work was presented on 6 November in the Novel Hypertension Therapeutics session of the meeting. These analyses were post hoc, meaning the questions were asked after the trials were designed and run, so they describe patterns in the data rather than test a hypothesis set out in advance.

Systolic blood pressure results in two trials

Across all three doses pooled, orforglipron cut systolic blood pressure by 6.1 mm Hg at 72 weeks in ATTAIN-1, against 0.8 mm Hg on placebo. In ATTAIN-2, the fall was 4.9 mm Hg against 1.5 mm Hg, and both gaps reached P below .001, as reported by Pharmacy Times. That leaves a placebo adjusted gain of about 5.3 mm Hg in people without diabetes and about 3.4 mm Hg in those with type 2 diabetes. Those net figures are our own subtraction of the reported means, not numbers the investigators gave. The reported results pool the three doses, so no dose response can be read from them, and the abstract did not say whether the highest dose added anything over the lowest.

Across all three doses pooled, orforglipron cut systolic blood pressure by 6.1 mm Hg at 72 weeks in ATTAIN-1, against 0.8 mm Hg on placebo.

Diastolic pressure told a split story. In ATTAIN-1, it fell 2.5 mm Hg with orforglipron and 0.9 mm Hg with placebo, a significant gap. In ATTAIN-2, the two arms were level, at 1.3 and 1.4 mm Hg, with a P value of .87. All readings were taken in the clinic. The abstract did not use ambulatory or home monitoring, which captures pressure across a full day and is less prone to a single high reading. It also did not report whether any participant reduced or stopped a blood pressure medicine during the trials. That gap matters, because a drop in office readings means something different when other drugs are changing at the same time.

Stage 2 hypertension and the placebo effect

The biggest falls came in people whose systolic reading started high. In ATTAIN-1, participants with stage 2 hypertension at baseline dropped 15.9 mm Hg on orforglipron and 11.1 mm Hg on placebo. In ATTAIN-2, the same group fell 15.8 mm Hg against 8.2 mm Hg. Stage 1 falls were smaller in absolute terms, but the drug still beat placebo, and every stage 1 and stage 2 comparison reached P below .001. The size of the placebo drops in stage 2 is the point to hold on to. High starting readings tend to drift down on their own at later visits, a pattern called regression to the mean, so the gap between arms carries the signal rather than the raw fall.

Read as gaps, the stage 2 edge looks less dramatic. Our subtraction gives about 4.8 mm Hg over placebo in the ATTAIN-1 stage 2 group, slightly less than the 5.9 mm Hg gap in stage 1, while in ATTAIN-2 the stage 2 gap of 7.6 mm Hg was the widest of any group. Diastolic changes by stage were smaller and closer to placebo, ranging from 3.6 against 2.4 mm Hg in ATTAIN-1 stage 1 to 6.5 against 5.4 mm Hg in ATTAIN-2 stage 2, and the abstract gave no P values for them. Results for people with normal or elevated readings at baseline were not reported, so the claim of benefit across every category cannot yet be checked.

What it means for the peptide GLP 1 race

The mediation model credited weight reduction with about 58% of the systolic change in ATTAIN-1 and 80.8% in ATTAIN-2. On that basis the investigators concluded that the effect was at least partly independent of weight loss. That is a modelled split, not a measured mechanism. In the diabetes trial, weight loss mediation leaves only about a fifth of the change unexplained, and the abstract did not say what might drive the remainder. A post hoc model of this kind cannot separate a direct drug effect from changes in diet, salt intake, activity or other medicines over 72 weeks. The finding is a reason for a planned blood pressure study, not proof that the pill treats hypertension.

Orforglipron is a small molecule, not a peptide, but it competes head on with the peptide GLP-1 drugs: weekly injectable semaglutide and tirzepatide, and Novo Nordisk's oral semaglutide pill, which beat sales forecasts last quarter while Foundayo fell short. The orforglipron blood pressure analysis made no comparison with any of them, so it says nothing about whether the pill lowers pressure more or less than a peptide does. What it adds in the oral GLP-1 contest is a consistent risk factor story next to the ACHIEVE 4 heart safety result, where the pill matched insulin glargine on major heart events. Matching a comparator on safety and trimming office readings is not proof of fewer heart attacks or strokes.

Conclusion

For now, the orforglipron blood pressure data are supportive rather than decisive. They show a consistent systolic drop against placebo in two large Phase 3 trials, larger in stage 2 hypertension and mostly carried by weight loss. They do not show heart outcomes, home readings, medication changes or any edge over peptide GLP-1 injections. The stage stays approved. The next checkpoints are a peer reviewed paper with the missing baseline groups and the European Commission decision that follows the EU committee recommendation.