Zepbound results, by the trial

SURMOUNT-1 was a multi-center, randomized, double-blind, placebo-controlled trial run by Eli Lilly and published in the New England Journal of Medicine in 2022 by Ania Jastreboff and colleagues. It's the trial the FDA leaned on to approve tirzepatide — sold as Zepbound for weight management — and it remains the single most-cited data source for what the drug actually does to body weight over time. The design was straightforward: 2,539 adults without type 2 diabetes, all with obesity, or overweight plus at least one weight-related condition such as hypertension, high cholesterol, obstructive sleep apnea, or cardiovascular disease, were randomly assigned in a 1:1:1:1 ratio to one of three tirzepatide maintenance doses or to placebo, alongside standard advice on a reduced-calorie diet and increased physical activity.

Every participant in a tirzepatide arm started at the same low dose — 2.5 mg once weekly — and climbed the dose ladder in steps every four weeks until reaching their randomly assigned maintenance dose of 5 mg, 10 mg, or 15 mg, where they stayed for the remainder of the 72-week trial. That shared starting point matters for reading the results: nobody jumped straight to 15 mg. Reaching it took roughly five months of gradual escalation, which is a big part of why the timeline in Section 04 below looks the way it does.

The headline numbers are the average percent change in body weight from the start of the trial to week 72. On the 15 mg dose, that average was 22.5% — roughly 52 pounds for a participant near the trial's average starting weight. On 10 mg it was 21.4%, and on 5 mg it was 16.0%. Placebo, importantly, wasn't zero: participants assigned to placebo still lost an average of 2.4%, a reminder that diet and lifestyle counseling alone move the needle somewhat, just far less than the drug does.

What jumps out comparing 10 mg to 15 mg is how close the two are — 21.4% versus 22.5%, about a one-point gap. The larger jump happens between 5 mg and 10 mg, a difference of more than five percentage points. That's one reason many prescribers don't treat 15 mg as automatically necessary for a good outcome; for a meaningful share of patients, 10 mg captures most of the drug's effect with a lower dose, and therefore often less gastrointestinal side effect burden and lower monthly cost.

Averages hide how a trial's results are actually distributed, so SURMOUNT-1 also reported the share of participants who crossed specific weight-loss thresholds — a more useful way to understand your own odds than a single mean figure. At the ≥5% threshold, the numbers are strikingly high across every dose: 89% of people on 5 mg and 96% on both 10 mg and 15 mg lost at least one-twentieth of their starting body weight, against 28% on placebo. That's the co-primary endpoint the FDA specifically looked at, alongside the average percent change itself.

The ≥15% threshold is where dose really starts to separate outcomes. Half of participants on 5 mg reached at least 15% loss, compared to about three-quarters on 10 mg and just under four-fifths on 15 mg. Put differently: on the highest dose, a clear majority of people lost enough weight to meaningfully change health markers like blood pressure and joint stress, but roughly one in five did not reach that mark even at the top dose — a useful corrective to any messaging that implies 15 mg guarantees a 20%-plus result.

The ≥20% threshold — a loss most people would describe as dramatic — was reached by about a third of participants on 5 mg, just over half on 10 mg, and nearly two-thirds on 15 mg. On placebo, essentially no one crossed that line, which is the clearest evidence in the whole dataset that the drug, not the diet counseling alone, is driving the outsized results at higher doses.

Read together, these numbers describe a spread, not a single outcome. Someone reading only the 22.5% average for 15 mg might expect that as a personal target; the threshold data shows plenty of people land well above it and plenty land well below it, even at the same dose.

The trial's headline numbers are all measured at week 72, but almost nobody experiences their weight loss as a single jump to that number — it unfolds in distinct phases, and knowing what to expect at each stage helps set realistic week-to-week expectations rather than judging early weeks against a 72-week endpoint.

Weeks 1–4: everyone starts at 2.5 mg regardless of their eventual target dose, which is a low, sub-therapeutic dose meant mainly to let the body adjust. This early period is dominated by gastrointestinal adjustment — mild nausea, appetite changes, occasional constipation or loose stool — more than dramatic weight change. Some early loss does happen, often from reduced appetite and portion size alone, but it's typically modest compared to what follows once the dose climbs.

Months 2–4 (roughly weeks 8–16): this is where the dose ladder starts to matter. Participants heading toward 10 mg or 15 mg step up every four weeks — 5 mg, then 7.5 mg, then 10 mg, and for the 15 mg group, 12.5 mg after that. Appetite suppression tends to become more pronounced at each step up, and this is the stretch where most people report the weight loss becoming clearly noticeable — looser clothing, a visible change on the scale week over week, comments from other people. It's also often when GI side effects are most active, since each dose increase can temporarily reintroduce some nausea or digestive adjustment before the body settles again.

Months 5–15 (roughly weeks 20–65): by around week 20, participants assigned to 15 mg have typically reached their full maintenance dose and remain there for the rest of the trial. Weight loss continues through this stretch but at a gradually decelerating pace — the biggest percentage changes tend to happen in the first several months at a given dose, with each subsequent month adding a smaller increment than the one before. This is normal physiology, not a sign the drug has stopped working; it reflects the body approaching a new, lower energy-balance equilibrium.

Weeks 60–72 and the plateau: SURMOUNT-1's own weight-loss curves show the rate of loss flattening out as the trial approaches its 72-week primary endpoint, without fully leveling off for everyone — some participants were still trending downward when the trial's measurement window closed. In practice, this means the 72-week number is a meaningful snapshot of where the trial stopped measuring, not necessarily the absolute floor of what continued treatment could produce, though the pace of further loss beyond that point tends to be slow for most people who've already reached a stable maintenance dose.

The practical takeaway from this timeline: the first month or two is the wrong window to judge whether the medication is "working" for you, since low starting doses are deliberately conservative and side-effect management, not maximum weight loss, is the priority early on. The clearest signal usually shows up by months three to four, once a therapeutic dose has been reached and held for several weeks.

It's worth saying plainly: SURMOUNT-1 reports averages and thresholds measured across a large, controlled population — it is not a personal prediction, and no two participants in the trial itself lost exactly the same amount of weight even at the same dose. Several factors explain why individual results spread out as widely as the threshold data in Section 03 shows.

The dose you actually reach and stay on matters more than any other single variable. The gap between 5 mg and 15 mg averages — 16.0% versus 22.5% — is larger than the gap between many other factors combined. Not everyone tolerates dose increases at the same pace; some people stop titrating at 5 mg or 7.5 mg because of side effects, cost, or simply because they've reached a weight they're satisfied with, and their results will track closer to the lower-dose averages rather than the 15 mg headline figure.

How long titration takes also shapes the outcome within a fixed observation window. Some prescribers move through the dose ladder faster than the standard four-week steps if side effects are minimal, while others slow down or hold a dose longer if a patient needs more time to adjust. A slower titration means less time spent at the final maintenance dose by any given week, which can translate to a smaller cumulative loss at that same point on the calendar, even if the person eventually reaches the same dose as someone who titrated faster.

Adherence is a more mundane but genuinely significant factor — missed or delayed weekly doses reduce the drug's effect, and real-world adherence is rarely as consistent as a clinical trial's structured dosing schedule with study-visit reminders built in. Diet and activity levels alongside the medication also still matter; SURMOUNT-1's placebo arm, which received the same lifestyle counseling without the drug, still averaged 2.4% loss, showing that behavior changes contribute independently of the medication's pharmacological effect.

Finally, individual metabolic and physiological response varies for reasons science doesn't fully explain yet — the same dose, same adherence, and similar starting weight can still produce different outcomes in different people, which is exactly why the trial reports a spread of thresholds rather than a single fixed number. None of this means the trial data is unreliable; it means the averages describe a population, and your own trajectory is one data point within that wide, real distribution — worth tracking on its own terms rather than benchmarked week-by-week against someone else's number or the trial mean.

SURMOUNT-1's primary measurement window ends at 72 weeks, and that's a study design choice, not a signal that treatment is meant to stop there. Most people who reach a maintenance dose and a satisfactory result continue treatment past that point specifically to hold onto it — obesity is generally treated in clinical practice as a chronic condition requiring ongoing management, similar to how blood pressure or cholesterol medications are continued indefinitely rather than stopped once a target number is reached.

What happens if treatment stops is a question SURMOUNT-1 itself wasn't designed to answer, since it was a fixed 72-week comparison against placebo, not a discontinuation study. That said, the broader clinical understanding of GLP-1 and dual-agonist medications is fairly consistent: appetite-suppressing effects fade once the drug clears the body, and without the medication actively working alongside diet and activity changes, weight tends to trend back upward for many people, sometimes substantially. This is one of the most important framing points in the entire weight-loss-drug conversation — the SURMOUNT-1 numbers describe what happens on the medication, not what happens after it, and treating a 72-week result as a permanent, locked-in outcome misreads what the trial actually measured.

For people who do stay on treatment, it's worth noting that participants who had pre-diabetes at the start of SURMOUNT-1 remained enrolled for an additional 104 weeks beyond the initial 72-week completion date, specifically to study weight trajectory and diabetes progression over roughly three years total — a sign that Lilly and the trial investigators themselves treated the 72-week mark as a checkpoint within a longer treatment horizon, not an endpoint to treatment itself.

The practical implication for anyone using or considering Zepbound: think of the SURMOUNT-1 percentages as what's achievable while actively on a given dose, held consistently, over roughly a year and a half — not as a one-time target to hit and then walk away from. Conversations with a prescriber about long-term dose, tapering, or discontinuation plans are a separate and important discussion from the results data itself.