There was a quiet but enormous shift in the way obesity was discussed in Istanbul this week. For the past three years, every major obesity meeting has been a parade of bigger and bigger weight-loss numbers. SURMOUNT-1. STEP 1. TRIUMPH. We watched the ceiling rise from 15% to 21% to 25% to 28%. The question was always: how much can we lose?

ECO 2026 was the first major congress where the question fundamentally changed. The headline trials weren't about how much patients could lose. They were about what happens next. How do we keep it? Can we step down the dose? Can we switch to a pill? What does the body look like a year after the scale stops moving?

This is the conversation the field has needed for years. And it is happening right on the heels of a piece of evidence that should have rattled every prescribing clinician: the BMJ's January 2026 systematic review and meta-analysis, which projected that patients regain to baseline weight an average of 1.7 years after stopping any weight-management medication, with cardiometabolic markers fully reverting within 1.4 years.¹ Cessation isn't a pause. It's a reversal.

Against that backdrop, the trials at ECO 2026 read like a unified statement: stay on therapy, in whatever form works for you. Let's walk through what we actually learned, what the data really says, and what I think it means for how we practice.

1 · SURMOUNT-MAINTAIN

SURMOUNT-MAINTAIN is the trial physicians have been asking for since tirzepatide hit the market. Patients who lose substantial weight on the maximum tolerated dose of tirzepatide (10 mg or 15 mg) routinely ask their physician the same question: can I step down? Until this week, we were answering from physiology, not from data.

Now we have the data. Published in The Lancet simultaneously with the ECO presentation, SURMOUNT-MAINTAIN randomized 378 patients who had lost at least 5% of their body weight over a 60-week open-label run-in to one of three maintenance arms: continued maximum tolerated dose, reduction to 5 mg, or placebo, for 52 more weeks.²

Figure 1

SURMOUNT-MAINTAIN: bodyweight change from baseline through week 112. Continuing MTD held the −22.4% weight loss completely. The 5 mg arm settled at −16.5% to −17.0%. Placebo regressed to −10.1% with rescue, or −4.8% without. Cumulative rescue tirzepatide use by week 112: 11 patients on MTD, 35 on 5 mg, 60 on placebo. Reproduced from Horn et al., Lancet 2026.

The way Dr. Horn and colleagues framed it: patients continuing the maximum tolerated dose were seven times more likely to clear an 80% maintenance threshold compared to placebo, and patients on the reduced 5 mg dose were four times more likely.³ Discontinuation, in other words, looked exactly like discontinuation of a statin or an antihypertensive: the disease comes back.

A few details I want clinicians to sit with. First, the rescue rate. Two out of three patients on placebo needed rescue therapy by the 24-week safety threshold. That is not a slow drift — that is a stampede back to baseline, despite ongoing lifestyle intervention. Second, the 5 mg dose held meaningful but partial benefit; this is a real option for patients with cost, access, or tolerability concerns, but it should be presented as the second-best path rather than the new default. And third, all of the cardiometabolic gains — waist circumference, blood pressure, lipids, glycemic markers — tracked tightly with weight maintenance. The drug isn't doing anything magical to vasculature when patients stop losing weight; the metabolic benefits are anchored to the weight reduction itself.

The blood pressure data deserves a closer look, because it is the single most important translational finding in the trial. SURMOUNT-MAINTAIN participants entered with a mean baseline systolic blood pressure of 126.4 mmHg — already in the normal-to-elevated range, the kind of BP most clinicians wouldn't bother treating. By the end of the 60-week weight-loss phase, mean systolic had fallen by about 9 mmHg across all arms. Then the randomization happened. At week 112:

Translation: stopping tirzepatide doesn't just bring back the weight. It brings back the hypertension. The placebo group's blood pressure benefit was entirely lost. The reduced-dose arm preserved about half of the benefit. Only continuation at MTD held the full 9-mmHg reduction. For a population at elevated cardiometabolic risk, that is not a soft endpoint — it's exactly the kind of finding that drives long-term cardiovascular events.

2 · ATTAIN-MAINTAIN

If SURMOUNT-MAINTAIN answered the "can I taper?" question, ATTAIN-MAINTAIN answered the question every needle-averse patient has been asking since orforglipron's April 2026 FDA approval as Foundayo: can I switch to the pill? The trial, also presented at ECO and published in Nature Medicine, took two cohorts from the original SURMOUNT-5 head-to-head comparison — one that had been on tirzepatide, one on semaglutide — and randomized them to either oral orforglipron 36 mg or placebo for 52 weeks.⁴

The results are nuanced, and the nuance matters.

Two observations here. First, the asymmetry. Patients switching from semaglutide to orforglipron held the line almost completely — barely a kilogram regained over a full year. Patients switching from tirzepatide held most of it, but not all. Five kilograms is not trivial, and it likely reflects what the authors hypothesized in print: moving from a dual GIP/GLP-1 agonist back to a single GLP-1 agonist is a pharmacologic step down, not a lateral move.⁴ The biology you signed up for when you went on tirzepatide isn't fully reproduced by a GLP-1-only molecule, regardless of route.

Second, this is the first prospective evidence that a meaningful number of patients can be transitioned from a high-cost, cold-chain-dependent injectable to a room-temperature pill and not give up most of what they earned. That has enormous implications for global access, especially in middle-income countries where injection storage and supply have been real barriers.

The authors frame it the way I would frame it for a patient considering the switch: this is a trade-off, made consciously, in pursuit of long-term persistence. For some patients, holding 95% of their weight loss while no longer dealing with weekly injections, monthly co-pays for cold-shipped pens, and the logistics of refrigeration is going to be the right answer. For others — particularly those who got to a hard-won target only on tirzepatide MTD — staying on the injectable will remain the right call. Shared decision-making, anchored to data, is finally what this is supposed to look like.

3 · STEP UP sub-analyses

STEP UP, the Phase 3b trial of semaglutide 7.2 mg vs. 2.4 mg vs. placebo, was originally reported in Lancet Diabetes & Endocrinology last year, with a 20.7% per-trial product weight-loss estimand at 72 weeks.⁵ That figure has been used in marketing decks ever since. ECO 2026 brought two sub-analyses that are arguably more important than the headline number.

The first is the early-responder analysis. About one-third of patients on 7.2 mg semaglutide hit a 15% weight loss threshold by week 24 — about six months in. Those patients went on to lose an average of 27.7% of their body weight at week 72. Patients who did not meet that early threshold still lost ~15%, a finding that should reassure clinicians and patients that a slow start is not a treatment failure.⁶ The 24-week mark is shaping up as the clinical decision point we have been looking for: by then, you know whether you're sitting on a strong responder, a moderate responder, or someone who needs a different molecule.

A companion STEP UP analysis from Argyrakopoulou and colleagues approached the same question from the other direction: how long does it actually take to reach each weight-loss threshold? The answer reframes how we should be counseling patients about timelines:

A patient on semaglutide 7.2 mg who is asking when they will hit ≥15% weight loss now has a real answer: about 32 weeks, or roughly seven and a half months. For ≥20%, plan on about ten months. For ≥25%, almost a full year of dose escalation and maintenance. This is the kind of granular timeline data clinicians have been pattern-matching to from memory. Now we can quote it from a trial.

The second STEP UP analysis is the body composition substudy — and this one matters even more, because it directly answers one of the most persistent fears about GLP-1s. The substudy used MRI in 55 STEP UP participants to characterize where the weight loss came from. The answer:

Eighty-four percent of the weight loss was fat. Visceral fat — the metabolically dangerous fat that wraps around the liver and viscera and drives most of obesity's cardiovascular harm — dropped by more than 30%. Lean mass did fall by about 10%, which has been the talking point pulled out of every GLP-1 critique on social media for the past two years. But here is what the critique almost never acknowledges: muscle function was preserved. Sit-to-stand performance, a validated proxy for lower-extremity strength and a meaningful predictor of mortality in older adults, was no different between treated patients and placebo at the end of the trial.

This is the body composition finding that should reframe the conversation. Lean mass loss happens with any caloric deficit — bariatric surgery, very-low-calorie diets, even prolonged moderate restriction. What matters clinically is whether the function the muscle was supposed to do is still being done. On semaglutide, it is.

4 · OASIS 4 sub-analyses

Oral semaglutide 25 mg — approved by the FDA in late 2025 and launched in January — has been the quiet success story of the year. The original OASIS 4 results in NEJM showed a mean 17% weight loss at 64 weeks.⁷ ECO 2026 brought two post-hoc analyses that are clinically more interesting than the average.

The first is the OASIS 4 early-responder analysis, which mirrors the STEP UP framework: about 29% of patients hit at least 10% weight loss by week 16. Those early responders went on to lose 21.6% of their body weight by the end of the trial. Non-early-responders still averaged 11.5%.⁸

The second is the one I think deserves more airtime than it has gotten. Among OASIS 4 participants with poor baseline physical function, 77.3% achieved a clinically meaningful improvement in mobility — bending, standing, sustained activity — compared to 42.9% on placebo.⁹ That is a near-doubling of the functional benefit in the population that needs it the most. We have a tendency in obesity medicine to over-index on weight as the only outcome that matters. The mobility data is a useful reminder that quality of life and the ability to physically participate in your life is at least as important.

5 · ORION & the oral GLP-1 wars

With two oral GLP-1 agents now FDA-approved for obesity — oral semaglutide (Wegovy pill) and orforglipron (Foundayo) — and no head-to-head trial in the field, Novo Nordisk filled the vacuum with ORION, a population-adjusted indirect treatment comparison using data from OASIS 4 (semaglutide oral 25 mg) and ATTAIN-1 (orforglipron 36 mg). The headline finding: oral semaglutide showed approximately 3 percentage points greater mean weight loss, and patients on orforglipron had roughly 14-times higher odds of discontinuing due to gastrointestinal adverse events.¹⁰

I want to be careful here. Indirect comparisons across trials with different populations, protocols, dose-escalation schedules, and timepoints are inherently weaker evidence than randomized head-to-head data. The 95% confidence interval on the GI discontinuation odds ratio in ORION was 2.0–96.0 — an enormous range that signals real uncertainty around the point estimate, even if the direction is plausible.¹⁰ When the lower bound of your effect estimate is 2 and the upper bound is 96, the responsible read is "probably worse, but we don't know by how much."

That said: ORION is consistent with what we already knew about orforglipron's somewhat tougher GI tolerability profile from its own Phase 3 program. The signal is real even if the magnitude is fuzzy. What clinicians should take from this: oral semaglutide and orforglipron are now both on the table, and the choice between them is not just about efficacy. It is also about lifestyle fit (orforglipron has no food or water restrictions; oral semaglutide must be taken on an empty stomach with limited water), cost, GI tolerability, and patient preference. The two oral options will compete for years, and I suspect most clinicians will end up using both, depending on the patient in front of them.

Every day, 4.5 million readers turn to 1440 for their factual news fix. We sift through 100+ sources to bring you a complete summary of politics, global events, business, and culture — all in a brief 5-minute email. No spin. No slant. Just clarity.

Join for free today!

6 · REDEFINE-1 (CagriSema)

REDEFINE-1, Novo Nordisk's pivotal trial for CagriSema — a fixed-dose combination of cagrilintide (a long-acting amylin analogue) and semaglutide — is in pre-registration for FDA approval. The original 22.7% weight loss result at 68 weeks puts it in the conversation with tirzepatide.¹¹ ECO brought two substudies worth knowing about.

The DXA body composition substudy, presented by Eric Ravussin, showed that CagriSema produced a mean 23.9% weight loss, of which 66.9% was fat mass and 33.1% was lean soft tissue. But the more interesting number is what happened in the subgroup of patients who achieved at least 30% weight loss: their proportion of lean soft tissue actually went up — from 51.3% of body mass at baseline to 63.2% post-treatment, while fat proportion fell from 46.3% to 33.2%.¹² Translation: even with substantial absolute lean mass loss, the body composition at the end of treatment was metabolically healthier than the body composition at the start.

The functional readouts told the same story. CagriSema produced an estimated treatment difference versus placebo of +12.9 points on the IWQoL-Lite-CT physical function score (95% CI 11.0–14.7; p<0.001) and +4.0 points on the SF-36v2 physical functioning domain (95% CI 3.4–4.7; p<0.001). Sit-to-stand performance — the validated marker of lower-extremity strength — was unchanged versus placebo (ETD 0.03, 95% CI −0.7 to 0.8; p=NS).¹² Substantial weight reduction, no measurable hit to functional strength. That is the body composition profile we have been hoping these drugs would deliver.

A separate REDEFINE-1 analysis showed clinically meaningful reductions in predicted 10-year atherosclerotic cardiovascular disease risk and in blood pressure, including in participants with resistant hypertension.¹³ Whether the CV signal will translate to hard outcomes is the entire point of REDEFINE-3, the dedicated cardiovascular outcomes trial whose baseline characteristics were also presented at ECO. We will not have those data for several years.

7 · VK2735 (Viking Therapeutics)

Viking Therapeutics presented updated VENTURE-Oral Phase 2 data for VK2735, an oral GLP-1/GIP dual agonist. At 13 weeks, the highest dose produced a mean 12.2% body weight reduction with no plateau. Up to 97% of treated patients achieved at least 5% weight loss; up to 80% achieved at least 10%.¹⁴ Discontinuation due to adverse events was 20% on VK2735 vs. 13% on placebo, with GI events the most common reason — manageable, but worth watching as the dataset matures.

Thirteen weeks is far too short to extrapolate to where this drug ultimately lands at 68 or 72 weeks. The trajectory of GLP-1/GIP combinations through this point has tended to flatten — tirzepatide doesn't plateau in early weeks either, but it eventually does. Still, the early kinetics and tolerability profile are encouraging enough that the ongoing Phase 3 VANQUISH-1 trial (~4,650 adults) is worth tracking as a potential third competitor in the dual-agonist space alongside tirzepatide and the CagriSema/oral pipelines.

Step back from the individual trials and ECO 2026 told a coherent story. Obesity is being treated, finally, like the chronic disease it has always been. Patients on effective therapy do well. Patients off effective therapy regress. The question of how to keep patients on therapy — affordably, tolerably, with options that fit the rhythm of their lives — is the question that will define the next phase of this field.

The framing matters. For decades we have asked patients to bear the moral weight of weight loss alone — to summon willpower, to fail and to keep trying, to interpret weight regain as personal failure rather than disease progression. The ECO 2026 data, taken together, make this framing increasingly difficult to defend. When 67% of placebo patients in SURMOUNT-MAINTAIN needed rescue therapy within months, despite ongoing lifestyle intervention, we are not watching a failure of effort. We are watching biology reassert itself. That is what every other chronic disease does when you stop the medication that controls it.

What I am taking back to clinic this week: be explicit with patients about maintenance from the first visit. Frame initial weight loss as Phase 1 of a multi-phase plan, not a destination. Discuss what tapering, switching, and long-term continuation could look like before the conversation becomes urgent. And for patients who ask whether they can stop — answer them with data, gently. The data are very clear now.

Istanbul didn't give us a new ceiling for how much weight patients can lose. It gave us something more useful: the first generation of evidence on how to make that loss durable. That is the frontier now. That is what the next decade of obesity medicine will be built on.

REFERENCES