The next incretin drugs, one molecule at a time

GLP-1 Drugs Shatter the Surgical Weight Ceiling

Ten years ago, if a pharmaceutical researcher stood up at a conference and claimed a weekly injection could make the average obese adult lose twenty-one percent of their body weight, they would have been escorted from the building — politely, with a parting gift bag if they were lucky, but escorted nonetheless. That number lived exclusively in the territory of bariatric surgery, and surgeons at least had the good faith to use a scalpel to get there.

Then the numbers started arriving, and they did not stop.

Semaglutide went first. In STEP-1, a 68-week phase 3 trial in adults with obesity or overweight but without diabetes, participants on the 2.4 milligram weekly dose lost an average of 14.9 percent of their body weight compared with 2.4 percent on placebo. Nearly seventy percent of people on semaglutide lost at least ten percent of their body weight. On placebo, that figure was twelve percent. The previous generation of weight-loss pharmacology had been getting five to ten percent body weight reduction across the best available compounds, with side effects assertive enough to make the disease seem manageable by comparison. Semaglutide did not inch past that benchmark. It doubled it.

Then tirzepatide arrived and made semaglutide look modest. In SURMOUNT-1, a phase 3 trial in adults without diabetes, participants on tirzepatide at 15 milligrams achieved an average body weight reduction of 20.9 percent. In per-protocol terms, that translates to 17.8 percent versus placebo. Eli Lilly had added six percentage points of weight loss on top of what had just been called a revolution, by building a drug that activates two receptors instead of one.

Completed phase 2 trials across incretin-based obesity therapies as of 2024 showed mean weight loss ranging from 7.4 to 24.2 percent. The top of that range now sits where bariatric surgery operates — procedures that achieve 25 to 35 percent body weight loss, requiring general anesthesia and a surgeon with steady hands. The pharmacological side of this gap is closing.

The data cascade from 2024 to 2026 did not stop at tirzepatide. CagriSema, Novo Nordisk's fixed-dose combination of cagrilintide and semaglutide, produced 20.4 percent mean weight loss in its REDEFINE 1 trial. Amycretin, also from Novo Nordisk, showed 24.3 percent weight loss at 36 weeks in subcutaneous form — preliminary data that has not yet been confirmed across the wider scientific community, but a number now in the published record. And then there is retatrutide, Eli Lilly's triple-receptor agonist currently in phase 3 development. TRIUMPH-1 reported 28.3 percent mean weight loss at 80 weeks. That figure is disputed and requires further confirmation before it can be treated as settled, but even if the final number lands considerably lower, the direction of travel is not subtle.

The first GLP-1 receptor agonist was approved in 2005. For the better part of fifteen years after that, the ceiling for drug-induced weight loss sat at roughly 5.8 percent placebo-subtracted, produced by liraglutide at 3.0 milligrams daily. The field went from 5.8 percent to TRIUMPH-1's 28.3 percent in two decades. Progress in pharmacology rarely looks like that. Progress in pharmacology usually looks like 5.8 to 6.1 percent, and a very confident press release.

Lilly and Novo Nordisk Redefine Obesity Pharmacology

The two companies driving this are Eli Lilly and Novo Nordisk. Lilly controls tirzepatide and retatrutide. Novo Nordisk has semaglutide, CagriSema, and amycretin. Between them they have produced a string of phase 3 results that keep redefining what the word ceiling means in this field, and they are funding several of the largest obesity trials ever run in the process.

What the raw numbers conceal is this: these drugs are not improved versions of the same thing. Each is a test of a distinct pharmacological hypothesis. Tirzepatide tests whether adding GIP receptor agonism to GLP-1 agonism is additive. Retatrutide asks whether glucagon receptor agonism, layered on top of both, is net beneficial rather than catastrophic. CagriSema tests whether amylin, an entirely separate hormonal pathway, meaningfully amplifies GLP-1 stimulation alone. Understanding which hypotheses are winning requires knowing the underlying science — the receptor biology, the peptide engineering, the sometimes contradictory mechanistic logic underneath the numbers. The clinical data is the score. The biology is the actual game, and it is considerably stranger than the scoreboard suggests.

So the drugs are extraordinary. But what are they actually doing inside the body, and why does adding a second or third receptor target keep producing better results? The answer requires a brief detour into the incretin system — a set of gut hormones and their receptors that, until about fifteen years ago, most people outside endocrinology had never thought about, and that now represent one of the most commercially valuable pieces of human biology ever identified.

Start with GLP-1R. The glucagon-like peptide-1 receptor is a class B G protein-coupled receptor. The class B GPCR family contains fifteen members in humans, all sharing a large extracellular domain at their N-terminal end that acts as a docking site for peptide hormones. GLP-1R sits at the center of this story because it is the primary target of every drug mentioned so far. It is expressed in pancreatic beta cells, in the brain, throughout the gastrointestinal tract, and in cardiovascular tissue, which is why the same drug can simultaneously lower blood sugar, reduce appetite, slow stomach emptying, and appear to protect the heart. GLP-1R signals primarily through the stimulatory G protein Gs, which drives cyclic AMP production and downstream insulin secretion. But it also signals through β-arrestin, a second pathway that shapes how the receptor behaves over time — its rate of internalization, its desensitization, and according to emerging structural biology, much of the difference between a drug patients tolerate well and one that sends them sprinting to the bathroom. The balance between these two pathways is called biased agonism, and it is one of the more contested engineering problems in current drug design.

GIP receptor, or GIPR, is GLP-1R's close relative — the two are homologous class B GPCRs that were probably making awkward eye contact at conferences for decades before anyone thought to activate both simultaneously. GIP itself is secreted from K cells in the duodenum after eating, and its receptor is expressed in the pancreas, adipose tissue, brain, and bone. For most of the history of incretin pharmacology, GIPR was treated as the boring one. In obese humans, GIP seemed to lose its insulinotropic effect, which made it look redundant. The assumption was that GLP-1 did the real work and GIP was decorative. Tirzepatide demolished that assumption comprehensively, and the mechanism explaining why remains contested enough that two credible research groups currently have different answers. That contradiction gets its own section.

Glucagon Receptor Risks in Triple Agonist Design

Then there is the glucagon receptor, GCGR. This is where the risk calculus shifts in ways that should make anyone designing a triple agonist slightly nervous. The glucagon receptor is expressed primarily in the liver and kidney. When glucagon binds it, the response is to raise blood glucose — the glucagon receptor promotes liver glycogen breakdown and drives gluconeogenesis. It also stimulates lipolysis and mitochondrial fat oxidation in the liver, and increases energy expenditure. In animal models, glucagon drives thermogenesis in a way that looks genuinely metabolically useful — and there it is, the one time this article had to use that word. The problem is that amplifying glucagon activity in a patient whose glucose regulation is already compromised is the pharmacological equivalent of pressing the accelerator and the brakes at the same time. GCGR antagonists, in clinical trials with type 2 diabetes patients, increased body weight, hepatic fat, and serum lipids. That tells you blocking the receptor is bad, but does not automatically mean activating it is safe. Triple agonists like retatrutide are betting that concurrent GIP and GLP-1 activity provides enough glycemic protection to make glucagon agonism net positive. That bet may be correct. The phase 3 data will say.

Finally, amylin receptors, expressed centrally in the area postrema and the hypothalamus, mediate satiety signals through a pathway entirely separate from incretins. Cagrilintide, the amylin analog in CagriSema, activates both amylin and calcitonin receptors. The significance is that combining amylin receptor activation with GLP-1R activation does not layer two copies of the same signal on top of each other. It runs two orthogonal satiety systems simultaneously, which may account for why CagriSema's clinical numbers exceed what either component produces alone.

The four receptors map onto three physiological functions. Glycemic control comes primarily from GLP-1R stimulating insulin synthesis and secretion from beta cells via a protein kinase A mechanism, with GLP-1 also suppressing glucagon secretion from alpha cells through a paracrine somatostatin pathway. Appetite suppression comes from GLP-1R's central actions — reducing appetite, delaying gastric emptying — plus the amylin pathway running in parallel. Energy expenditure comes from the glucagon receptor's hepatic effects. A drug hitting one receptor system gets one set of effects. A drug hitting two or three is running a parallel experiment in human metabolism in real time, and the results are arriving faster than the underlying theory can process them.

Tirzepatide is a 39-amino-acid synthetic peptide that activates both the GIP receptor and the GLP-1 receptor simultaneously, earning it the nickname twincretin — a name that manages to be both accurate and the best piece of pharmaceutical branding since someone decided "weekly injection" needed a friendlier face. Eli Lilly developed it, and the SURMOUNT program is where it announced itself.

SURMOUNT-1 Validates Tirzepatide's Dual GIP-GLP-1 Mechanism

In SURMOUNT-1, the drug produced mean weight reductions of 19.5 percent at 10 milligrams and 20.9 percent at 15 milligrams in adults with obesity but without diabetes. These numbers came from a drug that was adding GIP receptor activation on top of GLP-1 receptor activation at a time when the field's working assumption was that the GIP receptor was largely useless in obese humans. Tirzepatide did not read that assumption.

The SURPASS program established similar dominance in type 2 diabetes. Across the SURPASS trials, tirzepatide demonstrated superior glycemic control compared to GLP-1 monotherapy, beating semaglutide on HbA1c reduction while simultaneously outperforming it on weight loss. The SURPASS-CVOT cardiovascular outcomes trial then ran tirzepatide against dulaglutide — another GLP-1 agonist with established cardiovascular benefit — over an average of four years in patients with type 2 diabetes and established atherosclerotic cardiovascular disease. Tirzepatide was noninferior on major adverse cardiovascular events. It was not superior. That result sounds disappointing until you consider that dulaglutide had already demonstrated cardiovascular benefit in its own outcomes trial, making it a legitimately tough comparator, and tirzepatide still improved blood glucose, body weight, blood pressure, and blood lipids to a greater degree than dulaglutide across the board. Noninferiority on MACE against a drug that already beats placebo on MACE is not a failure. It is just a high bar.

The SUMMIT trial added a different dimension entirely. In patients with obesity-related heart failure with preserved ejection fraction — a condition where the heart muscle stiffens rather than weakens, and where pharmacological options have historically ranged from limited to "aggressive lifestyle counseling and a sympathetic shrug" — tirzepatide produced a 38 percent reduction in cardiovascular death or worsening heart failure. That number would have seemed implausible for a weight-loss drug a decade ago. It is now in the record.

The head-to-head data against semaglutide comes from SURMOUNT-5, a phase 3 multicenter randomized controlled trial enrolling 750 participants without diabetes, with tirzepatide dosed at 15 milligrams or maximum tolerated dose and semaglutide at 2.4 milligrams or maximum tolerated dose, primary assessment at week 72. The reported figures — tirzepatide producing around 20.2 percent mean weight reduction versus semaglutide's 13.7 percent — are disputed and should not be treated as settled. What is clearer from the same dataset is the distribution of outcomes: approximately 32 percent of tirzepatide-treated participants achieved at least 25 percent weight loss, versus 16 percent in the semaglutide arm. A post hoc analysis classified 44 percent of tirzepatide participants as rapid responders — defined as achieving at least 15 percent weight reduction by week 24 — compared with 21 percent in the semaglutide arm. Rapid responders experienced higher rates of gastrointestinal adverse events, but study completion rates were similar between rapid and non-rapid responders for both drugs.

Real-World Dosing Falls Short of Trial Benchmarks

The real-world picture is predictably less tidy. In actual clinical practice, only 25.9 percent of tirzepatide users reached the 15 milligram target dose. The gap between what a drug achieves at maximum tolerated dose in a controlled trial and what patients actually take is not a rounding error — it is the difference between a phase 3 result and a healthcare system functioning normally under supply constraints, side effect profiles, and the general chaos of human beings trying to take medications on schedule.

What none of this explains is why GIP receptor activation helped at all. Before tirzepatide, the pharmacological logic for targeting GIPR in obese patients was weak. GIP's insulinotropic effect diminishes in obesity, which led most researchers to write off the receptor as a dead end for this population. Tirzepatide's clinical performance forced the field to revisit that conclusion, and the result is an unresolved scientific argument about what, exactly, GIPR activation is doing that GLP-1R activation alone cannot — and whether the answer is what anyone initially guessed.

So tirzepatide beats semaglutide by a wide margin, and the question everyone immediately wanted answered was: why does the GIP receptor matter here? The field had spent years treating GIPR as the boring cousin at the incretin family reunion — the one who shows up, eats all the food, and contributes nothing to the conversation. In obese patients, GIP's ability to stimulate insulin secretion diminishes substantially, which led most researchers to conclude that targeting GIPR in obesity was a waste of molecular resources. Tirzepatide then went ahead and produced twenty-one percent mean weight loss anyway, which put everyone in the awkward position of having to explain results they had largely predicted would not happen.

Several competing hypotheses are now on the table, and none of them has yet won decisively. The first holds that GIPR activation in the hypothalamus and brainstem reduces nausea relative to equivalent doses of GLP-1 agonism alone. The practical implication is that tirzepatide's dual mechanism allows patients to tolerate higher effective doses before the gastrointestinal side effects become intolerable — meaning the drug's superiority might be partly pharmacological and partly a tolerance artifact. You did not build a better weight-loss drug. You built a drug people can actually stay on long enough to lose more weight. That is still better, but it is a different kind of better.

The second hypothesis points to adipose tissue. GIPR is expressed in fat cells, and GIP appears to promote lipolysis — the mobilization of fat stores — while also having direct effects on fat deposition. In animal models, GIP receptor activation drives changes in adipose tissue handling that GLP-1 agonism alone does not produce. Whether this translates cleanly into the weight-loss signal seen in human trials is contested. GIP's effects on fat tissue in humans are real but their net direction, and their relative contribution to the clinical outcome, are not settled.

GIP's Central Anorectic Effects Complement GLP-1 Signaling

A third hypothesis argues that GIP has direct anorectic effects in the central nervous system that complement GLP-1's central actions rather than duplicating them. On this view, the two receptors are running genuinely additive appetite suppression through partially distinct neural circuits. The MC4R knockout mouse data is relevant here: in mice lacking the melanocortin-4 receptor, a critical node in the central appetite regulation pathway, tirzepatide still produced 31.6 percent body weight reduction over 21 days. That suggests tirzepatide's weight-loss mechanism does not depend entirely on MC4R signaling, which is one of the main downstream pathways for GLP-1's central effects. Something else is running.

The fourth hypothesis is the most structurally subversive one. Tirzepatide is not semaglutide with a GIP receptor thrown in. It is a distinct 39-amino-acid peptide with its own binding geometry and receptor engagement profile. The drug's specific architecture may give it a different GLP-1R activation profile than semaglutide regardless of what the GIP receptor is doing. Biased agonism — the balance between G protein and beta-arrestin signaling discussed earlier — is exquisitely sensitive to the precise structure of the peptide docking at the receptor. Tirzepatide might be doing something different at GLP-1R that semaglutide simply does not do, and the contribution of GIPR activation to its superiority could be smaller than everyone assumes.

None of this would matter as much as it does were it not for what Amgen did next. Maridebart cafraglutide, also known as MariTide or AMG 133, is a bispecific antibody under clinical investigation that combines GLP-1 receptor agonism with GIP receptor antagonism — deliberately blocking the GIP receptor rather than activating it. The opposite of tirzepatide's approach to the same receptor, producing additive weight loss above GLP-1 monotherapy in early clinical data. If this holds up as the full clinical picture develops, it means the field is in the following position: activating GIPR helps, and blocking GIPR also helps, and nobody can yet give you a clean mechanistic account of why both are true simultaneously.

This is not a sign that the science is broken. It is a sign that the GIP receptor is doing something in human metabolism that the existing models did not fully capture — and that both agonism and antagonism can disrupt a dysfunctional equilibrium, just from different directions. The GIPR knockout mouse data is a useful reminder that GIP biology extends well beyond appetite: mice lacking the GIP receptor develop a high-turnover osteoporosis phenotype, with reduced bone formation parameters and elevated osteoclast counts, particularly multinuclear osteoclasts. They also show increased plasma calcium after meal ingestion compared to wild-type mice. GIP was doing something in bone before it was implicated in fat. The receptor has a metabolism-wide role that incretin pharmacology is only beginning to map, and the weight-loss story might be a side effect of disrupting a much larger system.

GIP Mechanism Unresolved as Lilly Adds Glucagon

The field's honest position on GIP, as of right now, is this: the receptor matters, the clinical data confirms it, and the mechanism remains genuinely open. That is an uncomfortable place for a field that just rewrote the weight-loss ceiling, but it is where the evidence sits.

Eli Lilly, not satisfied with having rewritten the obesity pharmacology ceiling exactly once, decided the logical next move was adding a third receptor. Retatrutide targets GLP-1, GIP, and glucagon receptors simultaneously — the first triple agonist to reach phase 3 development, and a drug that required advances in peptide engineering that were not straightforwardly achievable when the project began.

The glucagon piece is where the pharmacological logic gets uncomfortable for anyone with a basic understanding of endocrinology, which is presumably most of the people who approved this trial. Glucagon is the hormone that tells the liver to release stored glucose into the bloodstream when blood sugar drops. Deliberately activating the glucagon receptor in a drug aimed at obesity — and in many cases at patients who already have type 2 diabetes — looks, on paper, like a decision that should require a very long meeting and several written sign-offs. The bet is that glucagon does not only raise blood glucose. It also stimulates fat oxidation, drives thermogenesis, increases energy expenditure, and reduces food intake. The dual GLP-1 and GIP activity in the molecule provides enough glycemic protection to offset glucagon's hyperglycemic effects while the body gets the metabolic contributions that neither incretin produces on its own. GLP-1 receptor activation suppresses glucagon secretion from alpha cells through the somatostatin pathway. GIP provides additional insulin stimulus. Between them, the incretin half of the molecule is supposed to hold blood glucose in check while the glucagon half adds metabolic effects that would not otherwise occur. Whether that arithmetic is actually correct in humans over 80 weeks is precisely what the phase 3 data is reporting back.

Building the molecule was not simple. Retatrutide is an engineered amino acid sequence, not a derivative of any naturally occurring human peptide. It was constructed to activate three structurally similar but functionally distinct class B GPCRs simultaneously, with appropriate affinity at each, while remaining stable enough for once-weekly subcutaneous injection. Getting one of those receptor interactions right is a multi-year project. Doing all three in a single molecule required advances in peptide design and amino acid modification that pushed what had been demonstrated in this drug class.

Lilly's phase 3 TRIUMPH-1 trial enrolled 2,339 adults with obesity but without diabetes and ran for 80 weeks. Topline data reported 28.3 percent mean weight loss at 12 milligrams by the efficacy estimand. That number is disputed and preliminary, but it is in the published record. The 9 milligram dose reportedly produced 25.9 percent weight loss. The 4 milligram dose, 19 percent. In a prespecified extension cohort for participants with BMI above 35, the 12 milligram arm reached 30.3 percent weight loss by week 104. Lilly also reported that 65.3 percent of participants on the 12 milligram dose reduced their BMI below 30 kilograms per square meter — from an average starting BMI of 40 — by week 80. These are extraordinary numbers, and the question is whether the full dataset, peer review, and regulatory scrutiny deliver them intact.

Retatrutide Targets Three Receptors, Extends to Diabetes

TRIUMPH-2 moved to adults with both obesity and type 2 diabetes: 1,152 participants, median obesity duration of 21 years, median diabetes duration of 5.6 years, mean BMI of 38.2. Retatrutide at 12 milligrams reportedly achieved up to 20.8 percent weight loss at 80 weeks. The TRANSCEND-T2D-1 trial ran 40 weeks specifically in type 2 diabetes patients and reported HbA1c reductions of approximately 1.7 to 1.9 percent across the 4 to 12 milligram doses, against a placebo reduction of 0.8 percent. In both trials, no weight loss plateau was observed at study end. Participants were still losing weight when the data collection stopped, which is either encouraging or a sign that the trials needed to run longer, and is being presented as the former.

The gastrointestinal side effects are dose-dependent and on theme for this drug class. In TRIUMPH-2, nausea occurred in 14 percent at 4 milligrams and 28 percent at 12 milligrams, compared with 8 percent on placebo. Diarrhea ran 27 to 34 percent across doses. Vomiting in TRANSCEND-T2D-1 ran 15 to 18 percent across doses versus 2 percent on placebo. Discontinuation due to adverse events reached 12 percent at the 9 milligram dose in TRIUMPH-2. One less familiar signal: dysesthesia — abnormal skin sensation — appeared in 6 to 7 percent of participants at higher doses versus 1 percent on placebo. Nobody has a clean mechanistic explanation for that one yet. Pharmacovigilance teams are presumably grateful for the steady work.

The head-to-head that will actually resolve whether adding glucagon receptor activation to the dual incretin framework is worth the design complexity and tolerability trade-off is TRIUMPH-5, a direct comparison of retatrutide against tirzepatide with primary completion estimated in late 2026. Those results have not been published. When they arrive, they will tell the field whether the gap between a dual and triple agonist in phase 3 numbers reflects real pharmacological advantage or something the controlled trial environment produced that real-world practice will flatten. The field currently has its most dramatic efficacy data and one critical missing comparison, and TRIUMPH-5 is where the argument either gets settled or gets considerably more interesting.

Novo Nordisk took a different approach to the multi-receptor problem. While Lilly was engineering single molecules to carry multiple receptor activities simultaneously, Novo Nordisk combined two separately optimized drugs into one formulation. CagriSema pairs semaglutide with cagrilintide, a long-acting analog of amylin, the hormone pancreatic beta cells secrete alongside insulin after every meal.

Amylin acts centrally, through receptors in the area postrema and the hypothalamus, to reduce food intake and slow gastric emptying. The amylin pathway is entirely distinct from the incretin system. Combining amylin receptor activation with GLP-1 receptor activation is not running the same signal twice. It is running two separate satiety programs through two separate neural circuits. Novo Nordisk owned both assets, noticed they both reduced appetite, and combined them.

REDEFINE 1 Tests CagriSema Against Its Components

REDEFINE 1, the 68-week phase 3 obesity trial, enrolled 3,417 participants without diabetes. The trial randomized 2,108 to CagriSema, 302 to semaglutide alone, 302 to cagrilintide alone, and 705 to placebo — a design testing whether the combination outperforms either component alone. The weight-loss figures established earlier hold: 20.4 percent under the treatment-policy estimand, 22.7 percent under the stricter adherence estimand. Semaglutide monotherapy in the same trial reached 14.9 percent. Cagrilintide alone reached 11.5 percent. The combination outperformed either component.

Thirty percent of CagriSema participants achieved the combined target of BMI below 27 and waist-to-height ratio below 0.53 by week 68, versus 3.3 percent on placebo. Under the treatment-policy analysis, 34.7 percent reached at least 25 percent weight reduction and 19.3 percent reached 30 percent or above.

Then comes the body composition data from the prespecified analysis of 252 participants. CagriSema produced a relative fat mass reduction of 35.7 percent from baseline. It also produced a relative lean soft-tissue mass reduction of 14.4 percent. The fat mass number is the one that goes in the headline. Losing lean mass alongside fat is not unique to CagriSema — it occurs with bariatric surgery, with aggressive caloric restriction, with everything that produces large rapid weight loss. But the scale here put the question in front of the whole field, and it does not yet have a satisfying answer.

The GI tolerability data from REDEFINE 1 is the section of the clinical report nobody leads with. Gastrointestinal adverse events affected 79.6 percent of CagriSema participants versus 39.9 percent on placebo. Injection-site reactions ran higher than with semaglutide alone. The events were mostly transient and mild-to-moderate, which means most people continued treatment. A 40-percentage-point gap over placebo is still 40 percentage points, and in clinical practice, that matters.

In type 2 diabetes, REDEFINE 2 enrolled 1,206 patients for 68 weeks. CagriSema achieved 13.7 percent weight loss versus 3.4 percent on placebo, and 73.5 percent of treated patients reached HbA1c of 6.5 percent or below. In REIMAGINE 2, Novo Nordisk reported CagriSema achieving 14.2 percent weight loss against 10.2 percent for semaglutide under the efficacy estimand. The HbA1c advantage was narrower: 1.91 versus 1.76 percentage points. A separate meta-analysis found no statistically meaningful incremental glycemic benefit from adding cagrilintide to semaglutide across the full population. Whether the glycemic advantage is real or population-dependent will take several more trials to settle.

Novo Nordisk filed CagriSema for FDA approval in December 2025. A regulatory decision was expected by the end of 2026. The competitive landscape it enters already includes tirzepatide in widespread clinical use and retatrutide approaching its own regulatory submission in early 2027, carrying phase 3 weight-loss figures that sit above CagriSema's in the published record.

A Rising Pharmacological Ceiling With Unanswered Questions

That record, taken as a whole, has a clear direction and several unresolved details. The pharmacological ceiling for obesity treatment has been raised repeatedly since 2021 — fifteen percent, then twenty-one, then twenty-eight — and shows no obvious sign of stopping. Whether the drug that dominates the decade does so because it hits more receptors, spares muscle better, protects the heart over years, or remains effective after patients stop taking it, those are the questions that will decide the winners. The trials are running. The field that a decade ago could not imagine drug-induced weight loss above ten percent is now arguing in peer-reviewed journals about whether thirty is the floor or the ceiling.