STEP 8 puts a number on the gap: semaglutide 15.8% versus liraglutide 6.4%

Researchers published the first phase 3b head-to-head trial of two GLP-1 drugs used for weight management in JAMA on January 11, 2022. Over 68 weeks, adults taking once-weekly subcutaneous semaglutide 2.4 mg lost an average of 15.8% of their body weight, compared with 6.4% for adults taking once-daily subcutaneous liraglutide 3.0 mg ^[2][3].

The STEP 8 trial enrolled 338 adults at 19 US sites, with enrollment running from September 11 to November 26, 2019, and follow-up ending May 11, 2021 ^[2]. Participants had a body mass index of 30 or higher, or 27 or higher with at least one weight-related condition, and did not have diabetes ^[2]. At the start, the average age was 49, 265 participants (78.4%) were women, average body weight was 104.5 kg (about 230 pounds), and average BMI was 37.5 ^[2].

People were randomly assigned in a 3:1:3:1 pattern to semaglutide 2.4 mg with a 16-week dose escalation (n=126), matching placebo, liraglutide 3.0 mg with a 4-week escalation (n=127), or matching placebo, all on top of diet and physical activity counseling ^[2]. The two placebo groups were pooled (n=85) ^[2]. Participants who could not tolerate semaglutide 2.4 mg were allowed to stay on 1.7 mg; those who could not tolerate liraglutide 3.0 mg had to stop treatment and could restart the 4-week titration ^[2]. The semaglutide-versus-liraglutide comparison was open-label, meaning participants and investigators knew which active drug was being used, while each active drug was blinded against its matched placebo ^[2].

The numbers

The difference in average weight change was 9.4 percentage points in favor of semaglutide (95% CI, −12.0 to −6.8; P < .001) ^[2]. The pooled placebo group lost 1.9% ^[2].

The gap was wider at higher weight-loss thresholds. Reaching 10% or more weight loss: 70.9% with semaglutide versus 25.6% with liraglutide (odds ratio 6.3) ^[2]. Reaching 15% or more: 55.6% versus 12.0% (odds ratio 7.9) ^[2]. Reaching 20% or more: 38.5% versus 6.0% (odds ratio 8.2) ^[2]. All three comparisons were statistically significant ^[2].

Side effects tell a more nuanced story. Gastrointestinal adverse events — nausea, vomiting, diarrhea and similar problems — were reported by 84.1% of the semaglutide group and 82.7% of the liraglutide group, essentially the same ^[2]. But 27.6% of the liraglutide group stopped treatment for any reason, roughly double the 13.5% in the semaglutide group ^[2]. Overall, 319 participants (94.4%) completed the trial and 271 (80.2%) completed treatment ^[2].

Why it matters for patients

Both drugs are GLP-1 analogues approved in the US for long-term weight management, and until this trial there was no phase 3 comparison between them ^[2]. JAMA noted that five drugs are FDA-approved for long-term weight management in adults — orlistat, phentermine/topiramate, naltrexone/bupropion extended release, liraglutide, and semaglutide — yet fewer than 1% of US adults with obesity use one ^[1].

For someone weighing whether a weekly or daily injection matters, STEP 8 suggests the choice is not just about convenience. Average weight loss was about 2.5 times greater with semaglutide, and far more people crossed the 15% and 20% marks ^[2]. At the same time, stomach-related side effects were common in both groups, so switching from daily to weekly dosing did not mean avoiding them ^[2].

The higher discontinuation rate in the liraglutide arm is worth noting, though the trial design may have contributed: people who could not tolerate the top liraglutide dose had to stop and restart titration, while semaglutide users could simply stay on a lower dose ^[2].

Limits and context

The trial was funded by Novo Nordisk, which makes both drugs ^[3]. Company representatives helped design and run the study and analyze the data, and a company-funded medical writer helped draft the manuscript, though the sponsor did not have the right to veto publication or choose the journal ^[3]. Several authors are Novo Nordisk employees or shareholders ^[2].

The study was relatively small at 338 people, ran for 68 weeks, and the active-drug comparison was open-label ^[2]. Results were first presented at the Obesity Society's annual meeting on November 5, 2021, before publication ^[3]. The trial is registered as NCT04074161 ^[2]. STEP 8 did not test how either drug compares with tirzepatide or any other newer agent, and it did not measure cardiovascular outcomes.