The headline from the Phase 3 TRIUMPH-1 obesity trial, reported by Eli Lilly on May 21, 2026, is a number that's hard to put in context: 28.3% mean body weight loss at the 12 mg dose over 80 weeks, with the BMI ≥35 subgroup reaching 30.3% at 104 weeks. The quieter number is the 4 mg low-dose arm, which produced about 19% mean weight loss — a dose tier that hadn't been tested in the Phase 2 trial at all, and that hints at a tolerability-friendly path through the same mechanism.
If you're a man somewhere in your 30s or 40s, your weight has been creeping for a few years despite the usual interventions, and a clinician has either suggested or already prescribed a GLP-1 or triple-agonist class drug, the mean-of-the-trial number is mildly useful and your own number is the only one that actually matters. You will not lose the average. You will lose your number. The question is whether the trajectory you are on at week 6 matches what the trial cohort experienced at week 6 — and if not, whether the gap is signal or noise.
This post is about the data infrastructure that turns "is it working?" from a vibe into an answerable question. It is not medical advice. It is not a recommendation to use or seek any of these drugs. Decisions about pharmacotherapy belong with you and a clinician who has your full medical picture.
Retatrutide is investigational and not FDA-approved as of May 2026. Semaglutide (Wegovy) and tirzepatide (Zepbound) are FDA-approved for chronic weight management. Compounded retatrutide is not legal. The tracking framework below applies to any incretin-based therapy your clinician prescribes — the specifics of dosing, escalation, and side-effect management are between you and your prescriber.
What TRIUMPH-1 Reported
The trial enrolled adults with obesity (BMI ≥30) or overweight with cardiometabolic risk factors. Topline results announced by Lilly:
- 28.3% mean body weight loss at 12 mg over 80 weeks — approximately 70.3 lbs.
- 30.3% mean weight loss at 104 weeks in the BMI ≥35 subgroup.
- Approximately 19% mean weight loss at the previously untested 4 mg dose — approximately 47.2 lbs.
Full data — including detailed adverse-event rates, lean-mass changes, time-to-effect curves, and subgroup analyses — is expected at the American Diabetes Association 2026 Scientific Sessions in June. Until then, these are press-release numbers and worth treating as such. They are the largest mean weight-loss numbers published for any obesity drug, but they are not yet peer-reviewed.
For our purposes, the trial numbers are not a target — they are a reference curve. If your trajectory looks meaningfully off the reference curve at the same time point, that's information worth bringing back to your prescriber.
The Three Things You Want a Baseline On
The week before you start any incretin-based therapy is the highest-information week in the entire experiment. Everything you measure later is being interpreted relative to it. A vague "I was around 240 last winter" baseline means you can't answer the question "did this work" except in extremes. A clean baseline means you can answer it at week 6.
Three things, all in one Saturday morning:
1. Body composition, not just weight. A DEXA scan if you can get one — most major metropolitan areas have a $50–$150 cash-pay option. The reason DEXA matters here and didn't matter as much in the pre-GLP-1 era is that incretin-based weight loss reliably includes a meaningful fraction of lean mass loss. The published estimates for semaglutide (Wilding et al., NEJM 2021) put lean-mass loss at roughly 25–40% of total weight lost. Whether retatrutide's glucagon-receptor component changes that ratio is not yet known. Either way, "you weigh 30 lbs less" and "you have 30 lbs less of fat" are very different sentences. If a DEXA isn't accessible, a multi-frequency BIA scan is a distant second choice, and tape-measure circumferences (chest, waist at navel, hip widest point, thigh, biceps) are the third.
2. Strength benchmarks. Two lifts you do reliably, two rep-max numbers, the date. The specific lifts don't matter — what matters is that you can repeat the test the same way 8 weeks later. A common pairing: a hinge (deadlift or RDL), and a push (bench or overhead press). Record load, reps, and bar speed if you have a tracker that does it. Strength loss tracks lean mass loss closely. If your strength stays flat while your weight drops, that's the success case. If your strength drops faster than your weight, that's a signal to adjust protein, training, or pace.
3. Resting metabolic markers. Fasted morning bloodwork: HbA1c, fasting glucose, fasting insulin, lipid panel, ALT/AST, and resting heart rate. If you have an Apple Watch or Garmin, pull the prior 30-day resting heart rate average so you have a wearable-derived baseline as well. Triple agonists in particular produce a modest dose-dependent resting heart rate increase via the glucagon-receptor component — typically 5–10 bpm in Phase 2 — and you want to know your starting point.
Total cost: roughly $150–$400 for the DEXA + bloodwork. Time: a Saturday morning and 30 minutes at the gym. The leverage on a 12-month experiment is large.
What to Track Each Week During Titration
GLP-1 and triple-agonist class drugs use multi-month dose-escalation protocols. Nausea, fatigue, and altered satiety are concentrated during escalation; weight loss is gradual; body composition changes are gradual on top of weight changes. Daily tracking would drown signal in noise. Weekly is the right cadence.
Once a week, same day, same conditions:
- Body weight, morning, after first bathroom visit, before food or drink, on the same scale.
- Waist circumference at the navel — same time as weight, no clothes interfering.
- Average daily steps for the prior 7 days (your wearable already has this).
- Resting heart rate average for the prior 7 days. Watch for the dose-dependent rise.
- Protein intake in grams for the prior 7 days, averaged. The minimum-effective dose for muscle protein synthesis in caloric deficit is ≈1.6 g/kg/day; the upper end is ≈2.2 g/kg/day. See our GLP-1 muscle-loss body recomposition guide for the underlying evidence on why this number matters more on incretin therapy than off it.
- Resistance training sessions logged that week. The TRIUMPH-1 trial protocol did not require resistance training, but every protocol guide for incretin-based weight loss includes it as the primary lean-mass-preservation countermeasure.
Subjective markers, same once a week:
- Energy 1–5.
- Sleep quality 1–5.
- GI side effects (none / mild / moderate / severe).
- Hunger relative to baseline 1–5 (1 = much less hungry, 5 = much more hungry).
What to Track Monthly
The body-composition signal lags weight by weeks. Re-checking it weekly would just be noise. Monthly is the right cadence for the composition layer:
- Repeat strength benchmarks. Same two lifts, same rep range, same warmup.
- Tape-measure circumferences (chest, waist at navel, hip, thigh, biceps).
- Photo, consistent lighting and pose. Front, side, back. This is the cheapest body-composition tool you have access to, and the cumulative photo strip is often more useful than any single instrument.
What to Track Quarterly
The deeper signals run on a quarterly cadence:
- DEXA scan at month 4 and month 8, if accessible. The month-4 scan answers the lean-mass-loss question early enough to course-correct.
- Full fasted lipid + metabolic panel. HbA1c, fasting glucose, fasting insulin, ALT/AST, complete lipids.
- Resting heart rate delta from baseline. If it's risen by more than 10 bpm at quarter end, that's a conversation with your prescriber.
Reading the Reference Curve
The TRIUMPH-1 mean curve is roughly: a slow start during the first 4–8 weeks of titration, then a steady decline of roughly 1.5–2% body weight per month at the maintenance dose, with the curve beginning to flatten somewhere between months 12 and 18. The 80-week endpoint at 12 mg was 28.3% mean loss; the 104-week extended endpoint in the BMI ≥35 subgroup was 30.3%.
Three patterns to watch for in your own data:
On-curve. Your weekly weight change is within roughly half a percentage point of the trial mean for your time point. Strength is flat or slightly down. Waist circumference is dropping faster than weight in absolute terms (a good sign for fat-vs-lean ratio). Resting heart rate is up 5–8 bpm and stable. Protein intake is sustained at ≥1.6 g/kg/day. The interpretation: working as expected.
Off-curve, weight-only. Weight is dropping faster than the trial mean, but strength is dropping in parallel, the DEXA shows >40% lean-mass loss, and protein intake is below 1.4 g/kg/day. The interpretation: caloric deficit is too aggressive for the protein and training inputs. The fix is usually upstream of the drug — more protein, more lifting, slower pace — and a conversation with your prescriber about whether dose escalation should pause.
Off-curve, plateau-early. Weight stalls for more than 4 consecutive weeks at the maintenance dose, well before the trial's natural plateau window. Waist circumference also stalls. Resting heart rate elevation hasn't fully developed (which can indicate sub-clinical dosing). The interpretation: dose-response may need a conversation. This is also where adherence to dose, injection technique, and storage matter — none of the trial curves apply if the dose isn't actually being delivered.
Confounders You Want to Log
The hard part of personal data isn't measuring — it's separating one variable from another. Things that confound an incretin-based weight-loss signal:
- Strength-training program changes. New volume, new lift, new tempo, new shoes.
- Travel or major schedule changes. Hotel sleep, restaurant food, schedule chaos.
- Other medications started or stopped. Especially anything with appetite, energy, or weight effects.
- Illness or injury. A two-week bronchitis can produce a transient weight drop that looks like a dose-response.
- Alcohol or social-eating events. Carry over 24–72 hours.
- Sleep loss runs. A week below 6 hours can blunt weight loss substantially.
A one-sentence weekly note ("started squatting twice a week," "flew Tuesday, three restaurant meals") is enough to interpret a weird data point three months from now.
How This Fits With the Other GLP-1 Tracking Guides
This is the body-composition / metabolic protocol for triple-agonist response. For joint pain — which the TRIUMPH-4 trial flagged as a major secondary signal — see our knee-pain tracking protocol on GLP-1 and triple-agonist therapy. For the protein-and-training-specific mitigation of lean-mass loss, see the GLP-1 muscle-loss body recomposition guide. For the mechanism and trial-by-trial breakdown of retatrutide vs tirzepatide vs semaglutide, see the PeptideWise retatrutide profile.
The four-stream model — drug + weight + body composition + training — over 12 months is what turns a treatment decision from a guess into a measured trajectory.
Be the first to try Prova
We're building an app to track whether glp1-triple-agonist-tracking actually works. Join the waitlist.
What This Post Doesn't Cover
This is the tracking framework. It is explicitly not:
- A recommendation about whether to start any GLP-1, dual-agonist, or triple-agonist therapy.
- Guidance on which specific drug, dose, or escalation schedule is appropriate.
- A substitute for a prescriber who has your full medical picture, comorbidities, and prior interventions.
- Advice on stopping, switching, or adjusting any prescribed medication.
The tracking framework also applies to non-pharmacological weight-loss interventions — they don't have the same trajectory as TRIUMPH-1, but the four-stream model (intervention + weight + body composition + training) is exactly the same.
The TRIUMPH-1 numbers will look different by the time the full ADA 2026 presentation is published in June. The framework for tracking your own response won't.