Viking Therapeutics released topline results on September 22 from a trial that asked a question the obesity field has been circling for years without answering. Participants who had lost weight on weekly injections of VK2735 were moved onto less frequent schedules for three months. Most of them kept the weight off, the gastrointestinal side effects during that stretch were no worse than placebo, and the company's shares jumped while the two incumbent obesity leaders fell. The results are real and the caveats are substantive, and both deserve the same attention.
What the trial was built to measure
The study enrolled about 180 adults with obesity and otherwise healthy participants. Everyone received weekly subcutaneous VK2735 or placebo for 21 weeks. Weekly dosing produced 16 to 19 percent weight loss depending on the dose, against roughly zero on placebo, and the loss was still progressing at week 21 with no sign of a plateau.
At that point participants were re-randomized for 12 more weeks of maintenance: monthly injections, every-other-week injections, continued weekly injections, or placebo. The study's stated objectives were safety, tolerability and pharmacokinetics. Change in body weight was an exploratory endpoint. That distinction matters when the results are described as a demonstration of efficacy, because the trial was not designed or powered to prove one.
Participants reaching the maintenance phase were not people who had failed treatment. They were people who had responded to it, which is the group every clinician eventually has to make a decision about.
The numbers, and the arm that carries the argument
Following the company's topline release, the every-other-week arms retained a combined 90 percent of the weight lost during induction, against 61 percent for placebo. The best-performing arm in that group, participants who moved from 17.5 milligrams weekly to 10 milligrams every other week, retained 97 percent. The monthly arms retained a combined 85 percent, with the strongest of them, a straight transition from 17.5 milligrams weekly to 17.5 milligrams monthly, at 90 percent. Separation from placebo appeared within four weeks, and the p values ranged from 0.004 to 0.0002.
Two details in those tables deserve more attention than the headline figure. The first is the least-drug arm. Participants who dropped from 17.5 milligrams weekly to 5 milligrams every other week retained 83 percent, and the 10 milligram monthly arm retained 82 percent, both below their better-dosed comparators. Extending the interval worked in this study, and so did taking less drug, but the smallest exposures gave back the most ground. The relationship between dose and maintenance is not fully mapped by a 12-week study with arms of 10 to 16 patients.
The second is the placebo group, which retained 61 percent of its prior loss over three months. These were participants who had been receiving weekly VK2735 during induction and were switched to placebo, so that figure is a withdrawal result, and it is a milder one than the withdrawal trials of other drugs have produced. Some maintenance after stopping is normal in a study where everyone is being weighed monthly and talking to a clinician.
A small exploratory cohort that stayed on weekly 17.5 milligrams lost a mean of 21.7 percent from baseline by week 33, roughly 22 percent placebo-adjusted, still without a plateau, which suggests the induction phase in this trial was stopped by the protocol rather than by the pharmacology.
The tolerability data are the part that makes the schedule change plausible in the first place. Treatment-emergent adverse events ran 38 percent on every-other-week dosing and 49 percent on monthly dosing, both below the 52 percent reported on placebo, and one participant in the every-other-week group discontinued because of an adverse event. Rates of nausea, vomiting, diarrhea and constipation were close to placebo across the maintenance arms. That is a different pattern from the induction phase, where gastrointestinal effects are the standard reason people stop. The company also noted that the study used an accelerated two-week titration schedule and still produced gastrointestinal event rates similar to its earlier Phase 2 trial, which titrated over three weeks.
Why maintenance is the part of obesity medicine without an answer
The evidence that weight comes back after treatment stops is not in question. In SURMOUNT-4, a randomized withdrawal trial of tirzepatide published in JAMA, participants who continued treatment lost an additional 5.5 percent of body weight between weeks 36 and 88 while those switched to placebo regained 14 percent. Roughly 90 percent of the continued-treatment group kept at least 80 percent of their initial loss, against about 17 percent of the placebo group.
What has been missing is the step after that finding. A review in Diabetes, Obesity and Metabolism this year synthesized randomized withdrawal trials, meta-analyses and real-world cohorts covering more than 289,000 patients and concluded that discontinuation is associated with 60 to 90 percent weight regain within a year and a reversal of cardiometabolic gains, and recommended treating discontinuation as a high-risk clinical transition. An expert panel writing in the journal Obesity reached a plainer conclusion: the optimal maintenance dose and the duration of pharmacotherapy remain undefined.
Practice has filled the gap with habit. Patients who reach a goal weight are often moved to a lower dose or a longer interval, or told to pause treatment, because that is what the evidence from other chronic diseases suggests and because the alternative is paying for a weekly injection indefinitely. Viking's trial is the first randomized, placebo-controlled test of that habit for an incretin drug rather than the first demonstration that stopping costs weight. The distinction between tapering and quitting is precisely the thing no one had randomized.
Real-world data complicates the picture in a useful way. A Cleveland Clinic cohort of nearly 8,000 patients who stopped semaglutide or tirzepatide, published in March, found an average weight change of 0.5 percent regain a year later among patients treated for obesity, far less than the trials predicted, because 19.6 percent restarted the drug and 35.2 percent moved to another treatment. Individual outcomes varied widely, with 55 percent gaining and 45 percent holding or losing. Average figures hide both the patients who did fine and the ones who did not.
What twelve weeks of maintenance cannot settle
The study's limits are worth stating plainly. Three months is a short window for a disease that is treated for decades, and the question a clinician faces is what happens over years, not what happens between week 21 and week 33. The maintenance arms were small, the efficacy endpoints were exploratory, and the trial compared drug schedules against placebo rather than against each other in a way that would isolate whether the interval or the total dose is doing the work. Nobody discontinued treatment entirely and then restarted it, which is the path a large share of patients follow outside trials.
The population is also narrower than the market. Participants had obesity and were otherwise healthy, so the results say nothing directly about patients with type 2 diabetes, established cardiovascular disease or the kidney complications that now drive much of the prescribing in this class. And the tolerability finding, while encouraging, comes from a group that had already tolerated 21 weeks of weekly dosing. The people who quit early are not in the maintenance phase.
Why the dosing interval became a competitive variable
Obesity drugs are chronic therapies, and chronic therapies are judged by persistence as much as by potency. A monthly injection competes with a daily pill on convenience and with a weekly injection on the number of clinical touchpoints it requires in a year. Brian Lian, Viking's chief executive, framed the results as a route to improved adherence and sustained weight management, and Louis Aronne, a former president of the Obesity Society who consults for the company, described them as "a basis for improved dosing flexibility for both clinicians and their patients."
There is a commercial tension underneath that framing. Fewer doses per patient per year is less revenue from each patient, in a market that analysts project could reach $150 billion by 2035. Companies are betting that longer intervals win more patients than they lose in volume, and that the maintenance market will be larger than the discontinuation market it replaces. Viking shares rose sharply on the data while shares of Eli Lilly and Novo Nordisk fell the same day, which is the market taking a position on that bet.
What comes next
Viking plans a second part of the study to test oral maintenance dosing, which would extend the same question to the tablet formulation the company is developing alongside the injectable. The drug is already in Phase 3 trials for obesity, and it is one of several programs from smaller developers trying to break into a market two companies currently dominate. Oral candidates from other manufacturers are in late-stage development as well, and the competition is likely to be decided on the same axis this study measured: not peak weight loss, but how little treatment a patient can stay on and still keep the result. A maintenance claim would be a new kind of label in this class: current approvals authorize a dose and a schedule, not a transition between schedules, and regulators have not yet been asked to bless the idea that a patient can step down in frequency without stepping down in outcome.
For patients, the practical reading is modest and worth keeping in proportion. The trial suggests that a person who responds well to an incretin drug may be able to hold the result on a schedule that is easier to live with, which is a meaningful option for anyone who has considered stopping because weekly injections are a burden rather than because they wanted to stop. It does not suggest that the drugs can be set aside after a goal is reached. If anything, the placebo arm is the argument against that: three months off treatment cost 39 percent of the weight that had been lost, in a trial where everyone was still being monitored.
The larger point is about what kind of evidence the field now needs. Proving that a drug produces weight loss has become the easy part. Proving how little of it a patient can take and still keep the benefit is the question that will decide what obesity treatment looks like a decade from now, and this study is a first randomized answer for one drug in one schedule.
Primary sources
- Viking Therapeutics, Viking Therapeutics Announces Positive Topline Results from Maintenance Study of GLP-1/GIP Agonist VK2735 Demonstrating the Promise of Multiple Maintenance Dosing Regimens (September 22, 2026).
- Aronne LJ, Sattar N, Horn DB, et al. Continued Treatment With Tirzepatide for Maintenance of Weight Reduction in Adults With Obesity: The SURMOUNT-4 Randomized Clinical Trial. JAMA. 2024;331(1):38-48. doi:10.1001/jama.2023.24945.
- Diabetes, Obesity and Metabolism, Clinical Management of Weight Regain and Cardiometabolic Consequences After Discontinuation of GLP-1 Receptor Agonists (June 2026), for the discontinuation review and the 289,000-patient synthesis.
- Obesity, From Guidelines to Clinical Practice: Expert Perspectives on Long-Term Management of Obesity, for the conclusion that the optimal maintenance dose and duration remain undefined.
- Nasdaq and Motley Fool market coverage, Why Viking Therapeutics Stock Skyrocketed on Tuesday (September 23, 2026), for the induction-phase figures and the same-day move in obesity shares.