The Research Behind Semaglutide: What the Studies Show

Semaglutide is one of the most rigorously tested weight loss medications ever developed. By the time it received FDA approval for chronic weight management in June 2021, it had been studied in randomized controlled trials spanning 16 countries and more than 4,500 participants. Since then, its evidence base has only grown -- a 17,604-person cardiovascular outcomes trial, a 2-year efficacy study, and new research into oral formulations have made semaglutide one of the best-characterized drugs in modern obesity medicine.

This article reviews every major clinical trial: what was studied, who was enrolled, what the numbers showed, and what they mean for you as a patient. No jargon. No hedging. Just the data.

The STEP Trial Program: Overview

Every STEP trial used semaglutide 2.4 mg administered subcutaneously once weekly, the dose that became FDA-approved as Wegovy. Each trial studied a specific clinical question about who benefits most and under what conditions.

The trials were designed with a shared structure: a gradual dose escalation period (starting at 0.25 mg, increasing over 16-20 weeks), a maintenance phase, and end-of-treatment assessment. This dose escalation approach was specifically designed to reduce gastrointestinal side effects -- a lesson learned from earlier GLP-1 receptor agonists like liraglutide.

All four primary STEP trials were published in top-tier journals in 2021 -- the New England Journal of Medicine, JAMA, and The Lancet -- and were the basis of the FDA's approval decision. STEP 5, the 2-year extension, was published in Nature Medicine in 2022.

Source: Wilding JPH et al. N Engl J Med. 2021;384(11):989-1002. | Davies M et al. Lancet. 2021;397(10278):971-984. | Wadden TA et al. JAMA. 2021;325(14):1403-1413.

STEP 1 Through STEP 5: What the Numbers Show

STEP 1: The Pivotal Trial

STEP 1 enrolled 1,961 adults with obesity (BMI 30 or higher) or overweight (BMI 27 or higher) with at least one weight-related comorbidity -- but without type 2 diabetes. Participants were randomized 2:1 to semaglutide 2.4 mg or placebo, both combined with lifestyle counseling. Duration: 68 weeks.

The results redefined what pharmacological weight loss could achieve:

• Mean body weight reduction: 14.9% with semaglutide vs. 2.4% with placebo
• 69.1% of semaglutide patients lost at least 10% of body weight (vs. 12% placebo)
• 50.5% lost at least 15% of body weight (vs. 4.9% placebo)
• 32% lost at least 20% of body weight (vs. 1.7% placebo)

For context, no previously approved weight loss medication had produced average losses anywhere near 14.9% in a randomized controlled trial. Orlistat averages 3-5%. Phentermine/topiramate reaches about 9%. Semaglutide was a step change in what the field thought was achievable.

STEP 2: People with Type 2 Diabetes

STEP 2 enrolled 1,210 adults who had both obesity or overweight and type 2 diabetes -- a population in which GLP-1 medications were known to underperform. The trial compared semaglutide 2.4 mg, semaglutide 1.0 mg (the diabetes dose), and placebo. Over 68 weeks:

• Semaglutide 2.4 mg: 9.6% mean weight loss vs. 3.4% placebo
• Semaglutide 1.0 mg: 7% mean weight loss
• A1C reductions were significant in both semaglutide arms

The lower weight loss versus STEP 1 reflects the biology of type 2 diabetes: insulin resistance and some diabetes medications can blunt the weight loss response to GLP-1 therapy.

STEP 3: Adding Intensive Behavioral Therapy

STEP 3 combined semaglutide 2.4 mg with intensive behavioral therapy (IBT) -- a structured program including low-calorie meal replacements for the first 8 weeks, frequent counseling sessions, and physical activity guidance. Against semaglutide alone, the combination produced marginally greater results.

• Semaglutide + IBT: 16.0% mean weight loss over 68 weeks
• Placebo + IBT: 5.7% mean weight loss

The practical takeaway: semaglutide does most of the work. Adding intensive behavioral therapy adds roughly 1 percentage point of additional weight loss. That is meaningful for some patients -- but it also means that the medication's core efficacy holds even without a structured coaching program.

STEP 4: What Happens When You Stop

STEP 4 was a withdrawal study. All participants completed 20 weeks of open-label semaglutide (reaching a mean 10.6% weight loss). They were then randomized to continue on semaglutide or switch to placebo for an additional 48 weeks.

• Continuing semaglutide group: lost an additional 7.9% body weight over weeks 20-68
• Placebo group: regained 6.9% -- recovering most of what was lost in the first 20 weeks

This trial established a key clinical reality: semaglutide's effects are maintained only with continued use. This is not unique to GLP-1 therapy. It reflects the biology of obesity as a chronic, relapsing disease.

STEP 5: Two Full Years of Data

STEP 5 extended follow-up to 104 weeks (two years) in adults with obesity or overweight without diabetes. It enrolled 304 participants -- smaller than the primary trials but designed specifically to evaluate durability. At week 104:

• Semaglutide group: 15.2% mean body weight reduction
• Placebo group: 2.6% mean reduction
• Waist circumference, blood pressure, fasting glucose, and lipid profiles continued to improve throughout the 2-year period
• No new safety signals emerged compared to the 68-week data

STEP 5 was important because it answered the durability question. Weight loss on semaglutide does not plateau sharply and then reverse at 68 weeks. For people who continue treatment, results hold -- and even modestly improve -- over a 2-year window.

Source: Garvey WT et al. Nat Med. 2022;28(10):2083-2091. (STEP 5) | Rubino D et al. JAMA. 2021;325(14):1414-1425. (STEP 4)

SURMOUNT Trials: Tirzepatide's Clinical Debut

Tirzepatide is a dual agonist: it activates both GLP-1 receptors (like semaglutide) and GIP (glucose-dependent insulinotropic polypeptide) receptors. GIP receptors influence fat cell metabolism and energy expenditure through mechanisms distinct from the GLP-1 pathway. The combination appears to produce additive -- not just additive -- metabolic effects.

SURMOUNT-1

SURMOUNT-1 enrolled 2,539 adults with obesity or overweight and at least one weight-related comorbidity (excluding diabetes). Three tirzepatide doses (5 mg, 10 mg, 15 mg) were compared to placebo over 72 weeks. Results at the highest dose:

• Tirzepatide 15 mg: 22.5% mean body weight reduction
• 57% of participants on 15 mg lost at least 20% of body weight (vs. 3% placebo)
• Tirzepatide 10 mg: 21.4% mean loss
• Tirzepatide 5 mg: 16.0% mean loss

These numbers were unprecedented. For the first time, a clinical trial had produced pharmaceutical weight loss outcomes approaching what is typically seen with bariatric surgery. FDA approved tirzepatide for chronic weight management (as Zepbound) in November 2023.

SURMOUNT-2

SURMOUNT-2 enrolled 938 adults with obesity or overweight and type 2 diabetes -- the diabetic-population parallel to SURMOUNT-1. Over 72 weeks:

• Tirzepatide 15 mg: 15.7% mean body weight reduction
• Tirzepatide 10 mg: 13.4% mean weight loss
• Placebo: 3.3% mean weight loss
• A1C reductions ranged from 2.1% to 2.4% across tirzepatide arms

Even in people with type 2 diabetes -- the harder-to-treat population -- tirzepatide produced weight loss outcomes that matched or exceeded semaglutide's results in non-diabetic patients.

Source: Jastreboff AM et al. N Engl J Med. 2022;387(3):205-216. (SURMOUNT-1) | Garvey WT et al. Lancet. 2023;402(10402):613-626. (SURMOUNT-2)

The SELECT Trial: Cardiovascular Outcomes

SELECT (Semaglutide Effects on Heart Disease and Stroke in Patients with Overweight or Obesity) was the largest trial ever conducted for a weight loss medication. Its enrollment criteria were strict: adults with a BMI of 27 or higher who had pre-existing cardiovascular disease (prior heart attack, stroke, or peripheral artery disease) but no type 2 diabetes. The trial ran across 41 countries.

The primary endpoint was a composite of three major cardiovascular events (MACE): cardiovascular death, non-fatal myocardial infarction, or non-fatal stroke. Results:

• Semaglutide: 6.5% experienced a primary MACE event
• Placebo: 8.0% experienced a primary MACE event
• Relative risk reduction: 20% (hazard ratio 0.80, 95% CI 0.72-0.90, p < 0.001)
• Mean body weight reduction in semaglutide group: 9.4% over follow-up period
• Benefits were consistent across all pre-specified subgroups

Why does this matter? Previous cardiovascular outcomes trials for GLP-1 medications (LEADER for liraglutide, SUSTAIN-6 for semaglutide at diabetes doses) enrolled diabetic patients. SELECT was the first trial to show cardiovascular benefit in people without diabetes. That distinction matters clinically: it broadens the population for whom semaglutide has demonstrated risk reduction beyond glycemic management.

The cardiovascular mechanism is not fully resolved. The 20% risk reduction likely results from a combination of: weight loss itself (which reduces cardiac strain and inflammatory load), direct GLP-1 receptor activity on cardiac and vascular tissue, and improvements in metabolic biomarkers including blood pressure, C-reactive protein, and LDL cholesterol.

Source: Lincoff AM et al. N Engl J Med. 2023;389(24):2221-2232. (SELECT)

Real-World Outcomes vs. Clinical Trials

Randomized controlled trials are the gold standard for establishing efficacy, but they operate in idealized conditions. Trial participants receive regular check-ins, often free medication, and structured lifestyle support. Dropout is actively managed. Real-world patients are more varied: they have more comorbidities, miss doses, switch medications, and may not have the same support structures.

Several real-world datasets have been published since Wegovy's 2021 approval:

• A 2023 analysis of 3,000 Wegovy users from insurance claims data found mean weight loss of 10.2% at 12 months -- lower than STEP 1's 14.9% but still substantially greater than any pre-GLP-1 era medication
• Adherence at 12 months was approximately 65-70% in most real-world cohorts, meaning roughly 30% of patients had discontinued before the year mark
• Patients who remained on medication for 12 months or more showed outcomes closer to trial results
• Gastroenterological adverse events were the primary reason for early discontinuation in real-world settings, consistent with trial data

The clinical takeaway: the trials are a realistic ceiling for what you can achieve under optimal conditions. Most patients who stay on medication and escalate doses as tolerated will land somewhere in the 10-15% weight loss range over 12 months. That is still transformative for most people.

Long-Term Data and Weight Regain After Stopping

The weight regain question is one of the most common concerns people have before starting GLP-1 therapy. The data is clear, and it is worth understanding fully.

The STEP 1 withdrawal extension enrolled 327 participants who had completed the main STEP 1 trial. After the 68-week treatment ended, they were followed for an additional 52 weeks with no medication. Results:

• Mean weight regain: 11.6% body weight over 52 weeks (recovering about two-thirds of the 14.9% initially lost)
• Cardiometabolic improvements (blood pressure, lipids, A1C, waist circumference) also partially reversed
• But: even after the regain, participants remained below their original baseline weight on average

This pattern is consistent with what happens when you stop metformin for blood sugar or statins for cholesterol. The underlying biology reasserts itself. Obesity is a chronic condition driven by dysregulated hunger signaling -- when the medication is removed, that signaling returns. The weight coming back is not a sign that semaglutide "failed." It is a sign that the underlying disease is still present.

And it informs how to think about long-term treatment. Patients who expect semaglutide to "cure" their obesity are likely to be disappointed. Patients who approach it as management of a chronic disease -- like taking a blood pressure medication indefinitely -- tend to have better long-term outcomes.

Source: Wilding JPH et al. Diabetes Obes Metab. 2022;24(8):1553-1564. (STEP 1 extension) | Garvey WT et al. Nat Med. 2022;28(10):2083-2091. (STEP 5)

Ongoing Research: Oral Semaglutide and What's Next

The PIONEER trial program studied oral semaglutide for type 2 diabetes management at doses up to 14 mg. But researchers suspected higher oral doses could produce weight loss comparable to injectable semaglutide. The OASIS trials tested that hypothesis.

OASIS 1: Oral Semaglutide 50 mg

OASIS 1 enrolled 667 adults with obesity or overweight without diabetes. Participants received oral semaglutide 50 mg or placebo once daily for 68 weeks. Results:

• Mean weight loss: 15.1% with oral semaglutide vs. 2.4% with placebo
• 44% of patients achieved at least 15% body weight loss
• The oral formulation is approved in some markets as Rybelsus for diabetes and is under review for weight management use

This is significant because it means injectable administration may not be a permanent requirement for GLP-1 weight loss therapy. An oral option removes one of the primary barriers patients cite before starting treatment.

Amycretin and Next-Generation Combinations

Amycretin is a single-molecule combination of GLP-1 and amylin receptor agonism. Phase 1 data showed mean weight loss exceeding 22% over 36 weeks -- potentially surpassing even tirzepatide's results. Phase 2 trials are underway. Researchers are also investigating triple agonists (GLP-1 + GIP + glucagon) and oral tirzepatide formulations.

The SURPASS-CVOT trial, expected to report in 2025, will provide cardiovascular outcomes data for tirzepatide comparable to what SELECT provided for semaglutide -- answering whether tirzepatide's greater weight loss translates to greater cardiovascular protection.

Source: Knop FK et al. Lancet. 2023;402(10403):705-719. (OASIS 1) | Aroda VR et al. Lancet. 2019. (PIONEER trials)

Safety Profile Across Trials

The safety data for semaglutide is unusually robust. Between the STEP, PIONEER, SUSTAIN, and SELECT programs, semaglutide has been studied in well over 30,000 patients across different doses, populations, and durations. That breadth gives clinicians a detailed picture of both common and rare adverse events.

Gastrointestinal Events (Most Common)

In STEP 1, rates in the semaglutide group vs. placebo:

• Nausea: 44% vs. 16%
• Diarrhea: 30% vs. 16%
• Vomiting: 24% vs. 6%
• Constipation: 24% vs. 11%

The vast majority of GI events were mild to moderate and occurred during the dose escalation phase. They typically resolved within a few weeks as the body adjusted. The clinical strategy for minimizing GI side effects is gradual dose titration -- starting at 0.25 mg weekly and escalating slowly over 16-20 weeks before reaching the full 2.4 mg maintenance dose.

Serious GI adverse events (severe nausea requiring hospitalization, etc.) occurred in less than 1% of trial participants. Discontinuation due to GI events: 4.5% in STEP 1 vs. 0.8% in placebo.

Gallbladder Events

Cholelithiasis (gallstones) was reported in 2.6% of semaglutide participants vs. 1.2% placebo in STEP 1. Rapid weight loss of any cause is a known risk factor for gallstone formation, so some increase was expected. Patients with prior gallbladder disease should discuss this risk with their provider.

Pancreatitis

Acute pancreatitis was observed in 0.3% of semaglutide participants across the STEP trials, compared to 0.1% in placebo groups. This small numerical difference was not statistically significant, and the incidence rate in the semaglutide groups was within the range expected for the general obese population. But semaglutide should be used with caution in patients with a history of pancreatitis.

Thyroid and Cancer

Rodent studies showed dose-dependent thyroid C-cell tumors with GLP-1 receptor agonists. This has not been observed in human clinical data across the STEP trials or SELECT's 3.3-year follow-up. Semaglutide carries a contraindication for patients with a personal or family history of medullary thyroid carcinoma or multiple endocrine neoplasia syndrome type 2 (MEN2) as a precaution. The SELECT trial found no increase in cancer incidence overall.

Cardiovascular Safety in Non-Cardiovascular Patients

For patients without pre-existing cardiovascular disease, the STEP trials showed no cardiovascular harm. Heart rate increased modestly (approximately 1-3 beats per minute on average) with semaglutide, a class effect of GLP-1 agents. Blood pressure decreased, on average, in semaglutide groups -- consistent with weight loss-mediated effects.

Source: Wilding JPH et al. N Engl J Med. 2021. | Lincoff AM et al. N Engl J Med. 2023.

All Major Trials: Side-by-Side Comparison

Every pivotal GLP-1 weight loss trial in one table. Use this to understand how the medications compare across different populations and conditions.