Semaglutide
Ozempic / Wegovy / Rybelsus (GLP-1 receptor agonist)
The Ground Truth Score
four plain questions, never one numberProven drug, real risks
Bottom line
Semaglutide is one of the rare peptides with genuine grade-A evidence, multiple large RCTs and three FDA approvals for objective, measurable weight and glycemic endpoints, but it carries a boxed thyroid-tumor warning, a real GI burden, and rebound weight regain when stopped.
Does the science back it?
Do real people feel it?
Is it safe?
Could it be placebo?
Can you legally get it?
FDA-approvedPrescription medicineAn approved drug with a label, a manufacturer, and a prescribing route. This is the only tier where what is in the vial is guaranteed by anyone. Check the indication, though: approval is always for something specific, and it is rarely the use this compound is popular for.
Approved by FDA (Ozempic/Wegovy/Rybelsus). Note separately that FDA proposed in April 2026 to remove semaglutide from the 503B bulks list, narrowing the compounded route while the branded product stays approved.
Legal status as of 2026-04-30. This is a separate axis from the grades above, and it does not move them. A vote about who may legally prepare a compound is not evidence that the compound works.
"Do real people feel it?" is anecdote, not proof, weighted up because the science is thin, never because it beats a trial. And "could it be placebo?" is not an insult: if you feel better, that's real to you. The point is only to know whether you're paying peptide prices for an expectation.
Why is the evidence this thin? It's mostly economics →
Dose at a glance
full dosing ↓FDA-labeled regimens (reported, not prescribed here): Wegovy for obesity titrates over ~16+ weeks.
Reported, not prescribed. Verify your vial and your math.
First documented human use
Phase 2/3 human trials began in the early-to-mid 2010s; first FDA approval as Ozempic (injectable, type 2 diabetes) on December 5, 2017; first obesity approval as Wegovy on June 4, 2021. Unlike most peptides on this site, semaglutide has completed numerous large controlled human trials.
The pitch
What people claim it does
Stated plainly and neutrally, exactly as you'll hear it. We grade each one below.
- Once-weekly injection (or daily oral) GLP-1 agonist that reliably produces ~15% body-weight loss in non-diabetic adults over ~68 weeks (STEP 1) and strong HbA1c reduction in type 2 diabetes.
- Beyond weight: FDA-approved for cardiovascular risk reduction and, as Ozempic, for slowing kidney disease progression, benefits demonstrated in dedicated outcome trials (SELECT, SUSTAIN 6, FLOW), not extrapolated.
- Endpoints are objective and measurable (scale weight, HbA1c, MACE events), so the effect is not a subjective placebo artifact, this is the high-evidence exception in the peptide world.
The data behind each bullet
What actually backs it
Once-weekly semaglutide 2.4 mg produced mean ~14.9% body-weight loss vs ~2.4% on placebo over 68 weeks in non-diabetic adults (STEP 1, NEJM 2021).
Large double-blind placebo-controlled RCT (n=1961), published NEJM; replicated across STEP program.
PubMed: STEP 1 (Wilding, NEJM 2021) ↗Reduces major adverse cardiovascular events in overweight/obese adults with established CV disease and no diabetes (SELECT trial); basis for the March 2024 Wegovy CV indication.
Large randomized cardiovascular outcomes trial; FDA-approved indication.
PubMed: SELECT CV outcomes (Lincoff 2023) ↗Reduces HbA1c and lowers cardiovascular events in type 2 diabetes (SUSTAIN 6, 104-week RCT, n=3297).
Double-blind placebo-controlled cardiovascular outcomes RCT in T2DM.
PubMed: SUSTAIN 6 (Marso, NEJM 2016) ↗Most weight is regained after stopping: in the STEP 1 extension, participants regained ~11.6 points of body weight over the year after withdrawal (net ~5.6% retained).
Pre-specified RCT extension; objective DXA/scale data.
PubMed: STEP 1 withdrawal extension (Wilding 2022) ↗FDA boxed warning: causes thyroid C-cell tumors (incl. medullary carcinoma) in rodents; human relevance not determined. Contraindicated in personal/family history of MTC or MEN2.
FDA-approved label boxed warning (Wegovy/Ozempic); rodent data is grade-D for the tumor signal but the warning itself is regulatory.
FDA label: Wegovy (DailyMed/accessdata) ↗Acute pancreatitis (including fatal hemorrhagic/necrotizing cases) and acute gallbladder disease (cholelithiasis ~1.6% vs 0.7% placebo) reported.
FDA label adverse reactions + clinical trial incidence data.
FDA label: Wegovy adverse reactions ↗Weight loss includes meaningful loss of lean/skeletal muscle mass alongside fat, raising metabolic-health concerns especially in older adults.
Human body-composition sub-studies and reviews; magnitude and clinical significance still debated; some recovery seen in animal models.
PubMed: semaglutide lean mass / body composition ↗Prove it
Can you tell if it's working?
The honest test: if it's working, is there any objective way to see it — a blood marker, a scan, a wearable? For a lot of peptides the answer is no, and that's worth knowing before you spend on labs or gear.
DEXA (visceral fat + fat mass) and a home CGM are the strongest objective proof; waist tape, scale, and BP cuff corroborate it cheaply.
Bloodwork
A routine lab actually tracks whether this is working.
Semaglutide is a GLP-1 receptor agonist that augments glucose-dependent insulin secretion, suppresses glucagon, and slows gastric emptying; its glucose-lowering action is captured directly by routine HbA1c and fasting glucose, so efficacy in dysglycemic users is genuinely readable on standard bloodwork. In metabolically healthy weight-loss users glycemia starts near-normal, so blood is a weak proxy there.
Clear decline from baseline in dysglycemic users (SUSTAIN program reported roughly 1.0-1.5% absolute HbA1c reduction in T2D vs placebo), trending toward the <7% treatment goal / normalization
GLP-1 agonism raises glucose-dependent insulin and suppresses glucagon, lowering average glycemia integrated over the ~3-month red-cell lifespan
Lower fasting glucose vs baseline; in T2D, movement toward the 70-99 mg/dL range
Reduced hepatic glucose output and improved beta-cell response lower fasting glycemia; shifts faster than HbA1c so it reads earlier
Draw baseline
Before first dose, fasting draw (capture native glycemia while still drug-naive)
Retest
~12 weeks (after titration to maintenance dose and long enough for HbA1c to re-equilibrate over the ~120-day RBC lifespan); fasting glucose can move by 4-8 weeks
Read this before you spend on labs
HbA1c and fasting glucose are legitimate, orderable efficacy readouts, but only in users who start dysglycemic (prediabetes/T2D). In a metabolically healthy person taking semaglutide purely for weight loss, HbA1c and glucose start near-normal and may barely move even while the drug is clearly working; the real efficacy signals there are weight, waist circumference, and blood pressure, none of which are bloodwork. So bloodwork verifies the glycemic effect cleanly in dysglycemic users and is a weak proxy in normoglycemic weight-loss users. Fasting insulin and triglycerides were dropped from this panel: fasting insulin moves bidirectionally (GLP-1 directly augments insulin secretion, so it can rise even as insulin resistance falls) and triglycerides are a downstream consequence of weight loss, not a drug-specific efficacy readout.
Other ways to track it
DEXA scan — total fat mass, visceral adipose tissue (VAT), and lean mass (regional + total)
A working result shows total fat mass and visceral adipose tissue falling, with lean mass dropping too but proportionally less, so lean-as-a-percent-of-body rises. Track absolute lean mass deliberately: some lean loss is expected and normal, but if lean is falling as fast as fat, that is the signal to add resistance training and protein, not evidence the drug is failing. VAT reduction is the honest cardiometabolic payoff beyond the scale.
STEP 1 DEXA substudy: semaglutide 2.4 mg reduced total fat mass and regional visceral fat and raised the lean-mass proportion. VAT and fat mass are estimated from the scan, making DEXA the cleanest objective composition readout and the one that distinguishes real fat loss from water/lean loss.
14-day consumer CGM (Dexcom Stelo / Abbott Lingo / Libre) — mean glucose, time-in-range, and glycemic variability (SD/CV)
A working result shows lower mean glucose, higher time-in-range, tighter variability (lower SD/CV), and blunted post-meal spikes. The signal is largest in people with diabetes or prediabetes; in an already-normoglycemic lean user the mean barely moves and the honest readout is reduced post-meal excursions and variability rather than a dramatic drop.
CGM studies show semaglutide improves time-in-range and lowers glucose SD and post-prandial peaks in dysglycemic users. CGM captures the drug's core glycemic mechanism with far more resolution than a single HbA1c, and matches the direct blood finding; its usefulness is largest in prediabetic/diabetic users.
Home scale (weekly weight), tape-measure waist circumference, and validated upper-arm BP cuff
A working result shows steady weight decline (the pivotal trials centered on roughly mid-teens percent bodyweight over ~68 weeks, so expect a real downward trend over months, not a linear weekly drop), shrinking waist circumference, and frequently a modest fall in blood pressure. These are the cheapest honest confirmers and move in the same direction as DEXA.
Weight loss is the primary efficacy endpoint of the STEP program (high-quality RCT evidence), with waist circumference and systolic blood pressure as validated secondaries. Waist tracks VAT reduction; these are the accessible stand-ins when DEXA/CGM aren't available.
Urine albumin-to-creatinine ratio (UACR) — orderable spot urine lab
Only trackable in someone with elevated baseline albuminuria (diabetic kidney disease / CKD): a working result shows UACR declining. If your baseline UACR is normal, this readout stays flat and is NOT a useful tracker for you, so it only applies to the CKD/albuminuric subgroup.
The FLOW trial established that semaglutide slows kidney-disease progression and reduces albuminuria in diabetic CKD. UACR is a validated, orderable urine efficacy readout, but it is subpopulation-specific rather than a general tracker.
Bottom line
Semaglutide is one of the most objectively trackable peptides available: blood (HbA1c/fasting glucose) is direct, and beyond blood, DEXA (fat mass + visceral fat + lean) and a home CGM (time-in-range, variability) give clean efficacy proof, with waist/weight/BP as cheap confirmers. Three honest caveats: (1) watch absolute LEAN mass on DEXA, since some lean loss is expected and excessive loss is the main downside to catch; (2) CGM is a strong readout mainly in prediabetic/diabetic users — in a normoglycemic person it barely moves; and (3) skip VO2max and sleep/HRV wearables as efficacy proof — the cardiac trials used 6-minute-walk distance, not measured VO2max, and wearable VO2max/sleep changes mostly reflect the weight-loss denominator, so they are confounded, not validated readouts. Urine UACR only helps if you start with elevated albuminuria.
Logging baseline vs retest is the whole game. The Truth Tracker is our free log for bloodwork, DEXA, genetics, and wearable data — bring your own numbers.
How to track it →Mechanism
How it's assumed to work
Assumed · theoretical pathway
Confirmed (approved drug, not assumed): a GLP-1 analogue (~94% homology to human GLP-1) that selectively activates the GLP-1 receptor, stimulating glucose-dependent insulin secretion, suppressing glucagon, slowing gastric emptying, and acting on hindbrain/hypothalamic appetite circuits to reduce energy intake. Structural modifications (position-8 substitution resisting DPP-4, C-18 fatty-diacid chain binding albumin) give it its long half-life.
Dosing & handling
What users and clinicians report
FDA-labeled regimens (reported, not prescribed here): Wegovy for obesity titrates over ~16+ weeks. 0.25 mg weekly x4, then 0.5, 1.0, 1.7, to a 2.4 mg once-weekly maintenance dose (investigational 7.2 mg studied in STEP UP). Ozempic for diabetes: 0.25 mg starter x4 weeks, then 0.5-2.0 mg weekly. Oral Rybelsus/oral Wegovy taken once daily. Slow titration is deliberate to limit nausea.
Doses above are the approved labels for context only, this is a prescription drug requiring medical supervision, lab/history screening (thyroid/MTC, pancreatitis), and titration. Self-dosing from compounded or gray-market vials is where most dosing-error harm occurs; concentration and units must be verified against the actual product, not assumed. Do not exceed labeled maintenance dose without prescriber coordination.
Timing & food
Injectable: once weekly on the same day, any time of day, with or without food. Oral Rybelsus: once daily on an empty stomach with ≤4 oz water, ≥30 minutes before other food/drink/medication (food sharply reduces oral absorption). Slow weekly up-titration over ~16 weeks is the standard to minimize nausea; many users report taking the weekly shot before a lighter-eating day to manage GI effects.
Half-life
Approximately 1 week (~165 hours), enabling once-weekly subcutaneous dosing; the long half-life is driven by strong albumin binding (reduced renal clearance) and DPP-4 resistance.
Reconstitution sensitivity
The FDA-approved injectable products (Ozempic/Wegovy) ship as pre-filled pens, no reconstitution required, which is a safety advantage. Compounded/gray-market lyophilized 'research' semaglutide does require reconstitution and is where most dosing errors occur; verify mg per vial and resulting concentration against the actual label before any math, never assume vial size.
Real-world signal
What people actually report
Anecdote, not proof, weighted because the science is thin. Here's the record, graded on volume, consistency, and how credible the sources are.
Volume
Very high, semaglutide is among the most-discussed drugs online; one analysis covered 410,198 Reddit posts with 67,008 self-reported semaglutide users (May 2019-June 2025), alongside enormous coverage on dedicated forums, TikTok, and mainstream press.
Consistency
Highly consistent: users reliably report meaningful weight loss (commonly ~8-18% over 6-12 months) and a predictable, mostly transient GI side-effect profile (nausea, fatigue, vomiting, constipation). The two dominant narratives, 'it works' and 'the nausea / the regain after stopping are real', recur across independent communities and match the trial data.
Source credibility
Mixed-to-high. The volume includes many genuine first-hand patient reports that align with RCTs (raising credibility), but the space is heavily commercialized, telehealth prescribers, weight-loss clinics, and compounding pharmacies have strong financial incentives, so affiliate and clinic-sponsored content is discounted. The independent Reddit analysis and press coverage carry more weight than vendor testimonials.
- Widely reported as highly effective for weight loss and appetite suppression ('food noise' reduction), with many users citing ~10-20% body-weight loss over 6-12 months, consistent with trial data.
- Nausea is the dominant complaint, typically worst in the first 8-12 weeks and at each dose increase, then improving; vomiting, fatigue, constipation, and diarrhea are also commonly described.
- A recurring frustration is weight regain after stopping and the sense of open-ended dependence on the drug, matching the STEP 1 withdrawal data showing most weight returns within a year off-treatment.
- Growing community discussion of muscle/strength loss and the need to pair the drug with resistance training and high protein, plus warnings about cost, shortages, and risks of compounded or gray-market sources.
Placebo risk, Low
Low placebo risk: the headline endpoints are objective and externally measurable, scale weight, DXA body composition, HbA1c, and adjudicated cardiovascular/renal events, and were established in large double-blind placebo-controlled trials where semaglutide vastly outperformed placebo. The effect is pharmacological, not driven by subjective self-report.
Risk panel
What could go wrong
Adverse events
High rate of GI adverse effects: nausea (~20% at 1 mg, up to ~44% at the 2.4 mg obesity dose), vomiting, diarrhea, constipation, and fatigue. In a 67,000-user Reddit analysis, ~43.5% self-reported at least one side effect, GI predominating. Most GI effects are dose-titration related and improve over 8-12 weeks but drive a meaningful share of discontinuations.
Theoretical concerns
Rodent thyroid C-cell/medullary carcinoma signal (boxed warning) with undetermined human relevance; theoretical concerns around pancreatic effects, diabetic retinopathy worsening (seen in SUSTAIN 6), gastroparesis/delayed emptying complicating anesthesia, and possible aspiration risk during sedation. Lean-mass loss may harm long-term metabolic and functional health if not offset by resistance training and adequate protein.
Contraindications
Contraindicated in personal or family history of medullary thyroid carcinoma (MTC) or Multiple Endocrine Neoplasia syndrome type 2 (MEN2), and in known hypersensitivity. Caution/avoid with history of pancreatitis, severe gastroparesis, and active gallbladder disease. Pregnancy: discontinue (weight loss not appropriate in pregnancy). Use of unregulated compounded/gray-market "research" semaglutide carries dosing-error and purity risks the FDA-approved product does not.
Honest unknowns
Long-term (>2-3 year) effects on muscle, bone, and weight maintenance; durability of cardio-renal benefit after stopping; optimal strategy to prevent rebound regain (most weight returns within ~1 year off-drug); whether the rodent thyroid signal translates to humans at population scale; effects of very-high-dose (7.2 mg) regimens over time; and outcomes in people using compounded or off-label sources outside trial conditions.
Confound watch
Pivotal trials paired semaglutide with structured lifestyle intervention (diet + activity counseling), so real-world results without that support may be smaller, and weight regain is faster when lifestyle support is also withdrawn (STEP 1 ext. vs STEP 4). Community 'before/after' reports often omit diet/exercise changes, dose, and duration. Discount weight-loss-clinic and telehealth/affiliate sources that profit from prescriptions or compounded product.
History
Discovery → first use → status
- 2012-2016Semaglutide developed by Novo Nordisk as a long-acting, albumin-bound GLP-1 analogue; phase 2/3 human trials run.
- 2016SUSTAIN 6 cardiovascular outcomes trial in type 2 diabetes published (NEJM).
- 2017-12-05FDA approves injectable semaglutide as Ozempic for type 2 diabetes.
- 2019Oral semaglutide approved as Rybelsus (first oral GLP-1) for type 2 diabetes.
- 2021-03-18STEP 1 obesity RCT published in NEJM (~14.9% weight loss).
- 2021-06-04FDA approves Wegovy (semaglutide 2.4 mg) for chronic weight management.
- 2024-03-08Wegovy approved for cardiovascular risk reduction (based on SELECT).
- 2025Higher-dose (7.2 mg) STEP UP data (~21% weight loss) reported; oral Wegovy for weight management advances/approved; Ozempic gains chronic-kidney-disease indication.
Verification
The COA standard, applied
Cross-checked against FDA-approved labeling (Wegovy/Ozempic, accessdata.fda.gov), the STEP 1 RCT (NEJM 2021) and its withdrawal extension, SUSTAIN 6 and SELECT cardiovascular outcome trials, the NCBI StatPearls semaglutide monograph, and a published 67,000-user real-world Reddit side-effect analysis (Nature Health). PubMed citations use the canonical search-term URL format; the underlying trials are well-indexed and resolvable (PubMed served a bot CAPTCHA to the automated fetcher, not a broken link).
The full verification standard →Sources
Where this comes from
- FDA label - Wegovy (semaglutide) injection ↗· Boxed warning, contraindications (MTC/MEN2), pancreatitis/gallbladder, adverse-reaction incidence.
- FDA label - Ozempic (semaglutide) injection ↗· Original 2017 diabetes approval label and mechanism/PK.
- PubMed - STEP 1 obesity RCT (Wilding, NEJM 2021) ↗· ~14.9% weight loss vs placebo over 68 weeks; pivotal efficacy.
- PubMed - STEP 1 withdrawal extension (weight regain, 2022) ↗· Most weight regained within ~1 year of stopping.
- PubMed - SUSTAIN 6 cardiovascular outcomes (T2DM) ↗· 104-week CV outcomes RCT; retinopathy signal.
- PubMed - SELECT cardiovascular outcomes (overweight/obese, no diabetes) ↗· Basis for the 2024 Wegovy cardiovascular indication.
- NCBI StatPearls - Semaglutide monograph ↗· Approval timeline, indications, mechanism, safety overview.
- Nature Health - Self-reported side effects of semaglutide/tirzepatide online ↗· 67,008 self-reported users; GI side-effect frequencies (community signal).
- ClinicalTrials.gov - STEP 1 (NCT03548935) ↗· Registry record for the pivotal obesity trial.
The four lenses reflect the evidence and the real-world record as of the last review and will change as data arrives. Real-world signal and reported feedback are anecdote, not proof. Nothing here is medical advice or a prescription.