The GLP-1 Supercycle Has One Crack: Two-Thirds of Patients Quit Within a Year.

In 2024, a single molecule did something almost no drug in history has done: across three branded products, semaglutide booked roughly $29.3 billion in sales for one company.¹ Ozempic for diabetes, Wegovy for weight, Rybelsus in a pill — same compound, different doors into the same surging demand. The numbers read like a gold rush because they are one. But gold rushes have a quiet feature the headlines skip: most of the people who stake a claim walk away within the year. This one has the same problem, and it sits at the center of every forecast.

The official story is that the GLP-1 boom is a demand story — a giant, underserved population finally getting an effective obesity drug, and the only question is how high the curve goes. That framing is wrong in one decisive way. The drugs work while you take them and stop working when you stop. So the real contest was never about getting people to start. It's about keeping them on. The supercycle is real. The crack runs straight through the middle of it.

The numbers are enormous, and they don't agree with each other

Start with the figure everyone repeats: the "$100 billion-plus GLP-1 market by 2030." It sounds like consensus. It isn't. Goldman Sachs Research actually cut its 2030 anti-obesity forecast to $95 billion from a prior $130 billion — and trimmed its US peak number from $95 billion to $70 billion — citing price erosion, a patient population that segments into very different willingness to pay, an uncertain Medicare unlock, and tighter payer dynamics.⁴ Morgan Stanley, meanwhile, sees $190 billion — but by 2035, not 2030, and for the broader class that includes diabetes use, more than doubling the $79 billion in sales it recorded in 2025.⁵ Different years, different perimeters, different drugs counted. Stack them in a single sentence and you manufacture a number nobody actually forecast.

Notice what Goldman is really saying. It didn't lower its number because demand fell. Demand is climbing — J.P. Morgan estimates roughly 25 million Americans on GLP-1 treatment by 2030, up from about 6 million in 2024.⁸ Goldman lowered the number because the economics per patient got harder to defend: lower realized prices, payers narrowing who qualifies, and a Medicare door that may or may not open. The forecast that fell wasn't a forecast of interest. It was a forecast of capture.

The drug that everyone starts and most people quit

Here is the mechanism the demand story can't see. A peer-reviewed JAMA Network Open study followed 125,474 patients who started a GLP-1 drug, and found that most discontinued within a single year. Among patients without type 2 diabetes — the weight-management population, the one the whole obesity supercycle is built on — 64.8% stopped within twelve months. Among those with diabetes, it was roughly 47.4%.⁶ Nearly two-thirds of the headline market evaporates inside a year of starting it.

This would be a footnote if stopping were harmless. It isn't. These drugs deliver benefit through continuous use; the clinical record is consistent that pausing the medicine reverses much of the result — randomized withdrawal trials including STEP 4 and SURMOUNT-4 show rapid weight regain within a year of stopping, and a systematic meta-regression found roughly 60% of lost weight returned within twelve months of cessation.¹⁰¹¹ So a 65% one-year drop-off isn't a satisfied customer leaving — it's lost revenue and a patient sliding back toward where they started. The lifetime value of a GLP-1 patient is not the prescription. It's the renewal, year after year, against side effects, cost, and the human tendency to stop a daily or weekly burden once the urgency fades. That is why durable GLP-1 revenue is an adherence business wearing a pharmacology costume.

Why the pill and the dual agonist are really adherence moves

Read the recent product moves through the retention lens and they stop looking like a feature race. On December 22, 2025, Novo Nordisk won FDA approval for the Wegovy pill — once-daily oral semaglutide, the first oral GLP-1 cleared specifically for weight management, with 16.6% mean weight loss in the treatment-adherent arm of its OASIS 4 trial.³ A pill isn't strictly more effective than the proven injectable that came before it — at comparable doses, clinical results between the two formats are similar, and injectable semaglutide's bioavailability advantage is well established. What it is, is easier to keep taking. Remove the weekly needle and you remove one of the most common reasons people quit. The format is the retention strategy.

The competitive threat reads the same way. In the SURMOUNT-5 head-to-head trial, tirzepatide produced about 47% greater average weight loss than semaglutide over 72 weeks — 20.2% versus 13.7%.⁹ But tirzepatide isn't a purer GLP-1; it's a dual GIP/GLP-1 receptor agonist, and that second mechanism is part of why it pulled ahead. More efficacy is itself an adherence tool: bigger, faster results are what keep a wavering patient from walking away. The whole front line of this market — pill versus shot, single versus dual agonist — is a fight over who can make staying on the drug feel worth it.

Isn't this still the biggest pharma story of the decade?

The fair objection is that I'm burying a genuine megatrend under one caveat. And it's a strong objection. Semaglutide at roughly $29 billion in a single year is real money, the label has expanded beyond weight into cardiovascular risk reduction,² and the secondary effects are large enough to dent other industries — J.P. Morgan estimates GLP-1 adoption could cut $30–$55 billion of annual food-and-beverage revenue by the early-to-mid 2030s.⁸ A trend that reshapes the grocery aisle is not a mirage. All true. But none of it answers the discontinuation problem; it sharpens it. A bigger eligible population with a 50–65% annual drop-off is a leakier bucket, not a fuller one. The honest read is that the supercycle is genuine in its ceiling and fragile in its floor — and the floor is set by access policy, realized price, and adherence infrastructure, exactly the variables Goldman named when it cut the number.⁴ The trend is not in doubt. The capture is.

So the decision facing every player in this market isn't whether to mine the GLP-1 seam — it's whether to build for the rush or for the renewal. Chasing peak demand means optimizing for the first prescription, the launch, the headline weight-loss percentage. Building for durable revenue means treating obesity the way the body actually treats it: as a chronic condition that comes back the moment you stop managing it. The companies that internalize that will spend less on getting people to start and more on the unglamorous machinery of keeping them on. The gold rush will reward the prospectors loudly and the toll-keepers quietly. The molecule was the easy part. The hard part — the part the forecasts keep relearning — is that a drug you have to take forever is only worth what people are willing to keep paying for, forever.