Scroll through GLP-1 news for a week and you’ll see a dozen headlines built around a single number: “12% weight loss,” “30% weight loss,” “21% weight loss.” Those numbers get treated as directly comparable, like batting averages. They’re not. Two drugs can both report “20% weight loss” and mean genuinely different things, depending on trial design choices that never make it into the headline. Once you know what to look for, you can read a press release in about ninety seconds and know whether the number inside it is actually impressive — or dressed up to look that way.
Start with the phase, because it tells you how much to trust the number
Drug development moves through phases, and the number attached to a Phase 2 trial and a Phase 3 trial carry very different weight, even when they look identical on the page.
Phase 1 trials are small — often dozens of healthy volunteers — and exist mainly to check safety and dosing, not efficacy. You’ll rarely see weight-loss percentages worth citing from Phase 1 alone.
Phase 2 trials, often broken into 2a and 2b, are where efficacy signals first show up in a real patient population, typically a few hundred people, like Structure Therapeutics’ 230-person aleniglipron trial. These numbers are genuinely informative but come with a real catch: smaller trials produce noisier results, and dose-response patterns can shift once a larger, more diverse Phase 3 population is tested.
Phase 3 trials are the big, expensive, multi-thousand-patient studies that regulatory approval actually rests on — think Novo Nordisk’s OASIS program for oral semaglutide, or Lilly’s Triumph program for retatrutide. A Phase 3 number is the one worth building real expectations around. A Phase 2 number is a promising early signal that may or may not survive contact with a larger trial.
The practical rule: when a headline doesn’t specify the phase, that’s often not an accident. Check before you get excited about a number from a 200-person study as if it were settled science.
Is the number placebo-adjusted, or raw?
This is the single most common way GLP-1 trial numbers get inflated in casual reporting. A raw result tells you how much weight the treatment group lost, period. A placebo-adjusted result subtracts out how much the placebo group also lost — because in nearly every weight-loss trial, the placebo group loses some weight too, simply from the extra attention, monitoring, and lifestyle counseling built into a clinical trial protocol.
Oral semaglutide’s OASIS 4 trial is a clean example of how this gets reported: 13.6% weight loss in the treatment group, against 2.2% in the placebo group. Some coverage reports the 13.6% headline figure; a more careful outlet reports the roughly 11.4-point placebo-adjusted difference. Neither number is “wrong,” but they’re answering slightly different questions — “how much weight did people on the drug lose” versus “how much of that loss is actually attributable to the drug itself.” When you’re comparing two different drugs’ headline numbers, make sure you’re comparing the same version of that number for both, or the comparison is meaningless.
What’s the actual endpoint, and at what dose?
Trials almost always test multiple doses simultaneously, and the headline number nearly always comes from the highest dose tested — which is not necessarily the dose you’d actually be prescribed. Structure Therapeutics’ aleniglipron Phase 2b trial is a good example of why this matters: the drug scaled cleanly by dose, from 9.0% at 45 mg up to 12.1% at 120 mg. If a headline reports “aleniglipron: 12% weight loss” without the dose, a patient starting at a lower dose (which is standard practice for GLP-1 titration) may reasonably expect a smaller real-world result, at least initially.
You should also check the timeframe. A 36-week result and a 104-week result aren’t the same kind of claim, since GLP-1-driven weight loss tends to keep climbing for the first year or more before leveling off. Retatrutide’s Triumph-1 trial reported roughly 30% weight loss, but that number is anchored to 104 weeks — nearly two full years — which is a meaningfully different claim than a 36-week Phase 2b readout, even when the percentages get casually stacked next to each other in comparison charts.
Who was actually in the trial?
This is the variable that gets flattened out of headlines most aggressively, and it’s arguably the most important one for deciding whether a trial’s results say anything about your likely outcome. Retatrutide’s own Triumph program shows this clearly within a single drug: Triumph-1, run in patients with obesity but no major complicating conditions, produced the headline 30% figure. Triumph-2 and Triumph-3, run in patients who also had type 2 diabetes or established cardiovascular disease, produced meaningfully smaller results — around 21% and 23% respectively. RBC Capital Markets analyst Trung Huynh put the pattern plainly: those results “met expectations for patients with underlying conditions who often lose less weight.” That’s not a knock on the drug. It’s a reminder that “GLP-1 drugs produce X% weight loss” is never really a complete sentence — it always has an implicit “…in this specific population” attached, and that population matters enormously if you’re trying to predict your own results.
A five-question checklist for the next headline you read
Next time a GLP-1 headline crosses your feed, run it through five quick questions before deciding how impressed to be: What phase was this trial? Is the number placebo-adjusted or raw? What dose and timeframe does it reflect? Who was actually studied — healthy-weight-adjacent patients, or people with diabetes and cardiovascular disease? And is this a peer-reviewed publication, a conference presentation, or a company press release — since press releases are written by marketing teams, not scientists, and tend to lead with the most flattering framing available.
None of this is about being cynical toward the science, which in this drug class has genuinely been remarkable over the past five years. It’s about giving yourself the tools to tell the difference between a real breakthrough and a press release doing its job a little too well.
Frequently asked questions
Why do some GLP-1 trials report different weight-loss numbers for the same drug?
Usually because they’re reporting different doses, different timeframes, different patient populations, or a different phase of development — all of which are legitimate reasons for numbers to vary, even for the same underlying drug.
Is a Phase 2 trial result trustworthy at all?
Yes, as a genuine early signal — Phase 2 data is a real, peer-reviewed-quality data point, not a rumor. It’s just statistically less certain than a Phase 3 result and more likely to shift as the drug moves through larger trials.
What’s the fastest way to check if a headline number is placebo-adjusted?
Look for the placebo group’s own result reported alongside the treatment number. If you only see one number with no comparison group mentioned, treat it as unconfirmed until you can find the full trial data.
Sources: Structure Therapeutics aleniglipron Phase 2b data — ScienceDaily; Lilly retatrutide Phase 3 Triumph trial results — BioPharma Dive; Oral semaglutide OASIS 4 approval coverage — AJMC.