Two obesity-drug giants are now fighting over whose ad tells the truth

Biotech & Longevity 4 min 80 sources

Novo Nordisk sued Eli Lilly this week over weight-loss ads it calls misleading, turning the year's biggest drug rivalry into a fight about what a head-to-head trial really proves. Elsewhere: a psoriasis pill and a prostate treatment let their data speak, a knee therapy failed, and polio's endgame stalls.

Key takeaways

Novo Nordisk sued Eli Lilly over weight-loss ads it calls misleading — the fight is over how a real head-to-head trial gets framed, not over whether either drug is safe.
Clean wins landed elsewhere: a psoriasis pill cleared skin in about three-quarters of patients, and a 10-year study showed a targeted prostate treatment matches surgery with far fewer side effects.
A knee-cartilage gene therapy failed its trial, Lilly bet $2.8bn on psychedelics for depression, and polio's global eradication has quietly stalled.

The two companies that dominate the weight-loss business spent this week fighting not over molecules, but over advertising. Around it, quieter results did the opposite — they let the numbers speak. Here is what moved.

The obesity war moves to the courtroom

On Tuesday, Novo Nordisk — the Danish maker of Wegovy and Ozempic — sued its US rival Eli Lilly over what it called “false and materially misleading” ads for Lilly’s competing weight-loss drugs, Zepbound and Mounjaro [9][16]. Novo’s complaint is specific: it says Lilly “intentionally selected outdated studies” that pitted Lilly’s highest doses against lower doses of Novo’s medicines, then presented the result as proof that Lilly’s drugs are simply better [9][58].

The two drugs work in related but different ways. Novo’s semaglutide mimics a gut hormone called GLP-1, which curbs appetite. Lilly’s tirzepatide mimics two gut hormones — GLP-1 and a second one, GIP. Lilly’s defence rests on one trial, SURMOUNT-5, which it calls “the only direct, head-to-head randomised clinical trial” comparing the two drugs for weight loss [9]. A randomised trial splits patients by chance into groups so the comparison is fair — and by that measure, Lilly’s drug won.

So both sides can be technically right. There was a head-to-head trial, and Lilly’s drug did beat Novo’s in it. And Novo can argue the ad campaign — run during big sporting broadcasts and on TikTok and Facebook — stripped out the context a patient would need [9][58]. Novo wants a court to pull the ads and order a “corrective” campaign, and may seek an injunction within days [58]. The stakes are enormous: analysts think the weight-loss drug market could top $100bn a year by 2030 [58]. This is a marketing dispute, not a safety one — no one is claiming either drug is unsafe.

When the data speaks for itself

Two results this week needed no spin. Takeda reported fuller phase 3 data on zasocitinib, a once-daily psoriasis pill — a phase 3 trial is the large, final test before a company seeks approval [1]. Across two studies of 1,801 patients with moderate-to-severe plaque psoriasis, the pill cleared or nearly cleared scalp psoriasis in 77% and 74% of patients, against 7% and 13% on a dummy pill, and beat an existing rival drug too [1]. Scalp, palms and nails are the hard places to treat, so the numbers matter. The caveat is ordinary: this is company-reported data presented at a conference, not yet independent long-term follow-up.

The stronger result came for prostate cancer. A 10-year NHS study led by Imperial College London followed nearly 3,500 men treated with “focal therapy” — destroying just the tumour with high-intensity ultrasound or freezing, instead of removing or irradiating the whole prostate [52]. Most had intermediate- or high-risk cancer, yet 10 years on only two men had died of the disease — outcomes as good as surgery or radiotherapy, but with less than half the risk of side effects like incontinence or loss of sexual function [52]. Regulators had said long-term data was the missing piece; now it exists. Separately, GSK won an early US approval for a lung cancer drug, Jideytro, its entry into that market [34].

A miss, and a bet

Not everything worked. Kolon TissueGene’s cell-based gene therapy for knee osteoarthritis failed its phase 3 trial: a year after injection, the 531 patients who got it did no better on pain and function than those given a placebo — a dummy injection [19]. The therapy has a strange history: South Korea approved it in 2017, then retracted the approval two years later over a mix-up about which cells were actually in it [19]. A second late-stage study reports in October, so Kolon is holding off on next steps.

The week’s big bet was Eli Lilly again — this time buying AtaiBeckley for $2.8bn, a company developing psychedelic drugs for depression [26]. It is the strongest sign yet that a major drugmaker is willing to take seriously a field long dismissed as fringe. Whether these drugs clear regulators is a separate question; a buyout validates the science’s promise, not its proof.

The eradication that keeps not happening

End on the story the headlines skip. Ten years ago, polio looked all but beaten — down to a handful of cases in two countries [64]. It still isn’t gone. Amid funding cuts and political barriers, researchers are now openly asking whether the decades-long eradication campaign can ever finish the job, and what a “plan B” would even look like [64]. It is a reminder that in medicine the last few percent — of a disease, of a population, of certainty — is often the hardest and least glamorous part of all.

02 · Lesson · why it matters

A fair comparison still has an author

A head-to-head result feels like a fact handed down by nature. It isn't. Someone chose the doses, the patients, and the yardstick — and those choices were made before the trial ever ran.

Two truths that don’t fit together

Novo Nordisk and Eli Lilly make the two biggest weight-loss drugs in the world. This week Novo sued Lilly, saying Lilly’s ads that its drug is superior are misleading. Lilly’s reply: our claim rests on the one head-to-head trial that exists, and our drug won it fair and square.

Here is the strange part. Both of those statements can be true at the same time.

There really was a randomised trial. Lilly’s drug really did come out ahead in it. And Novo can still argue the ad leaves out something a patient needs to know — that, by Novo’s account, the trial lined up Lilly’s strongest dose against a weaker dose of Novo’s drug. You don’t have to decide who is right to notice the puzzle. How can a genuine result and a misleading claim be the same result?

The setup does the deciding

Because a comparison is never just a fact. It is a fact wrapped around a stack of choices, and most of the choices are made before anyone measures anything.

Which two things do you compare? At which doses? In which patients? Judged by what — weight lost, or side effects avoided, or cost? Over how many weeks? Every one of those is a decision. Change a single one and the winner can flip. “Highest dose against lower dose” is not a trick hidden in the fine print; it is the whole game, sitting in plain sight, deciding the outcome before the first patient is weighed.

So when a result lands in front of you as “A beats B,” you are seeing the last line of a story whose earlier lines were written by someone with a stake in the ending. The number is real. The setup that produced it is a choice.

You have seen this a thousand times

This is not a quirk of drug companies. It is the shape of almost every comparison you meet.

“Nine out of ten dentists.” Which ten, asked what, given which options? “Clinically proven more effective.” Against what, in whom, measured how? “Our plan is cheaper.” On which basket, over which term, for which customer? The claim is always loud and simple. The setup that made it possible is always quiet and buried. That gap — between the confident conclusion and the invisible conditions — is where most persuasion lives.

None of this requires anyone to lie. You can build a perfectly honest comparison and still build it to flatter your side, just by choosing, in good faith, the ground where you are strongest.

The comparison is aimed at you

Sit with where you stand in this. You are not the referee. You are the person the ad is for — the patient in the waiting room, the shopper reading the box, the reader deciding.

And a comparison built on averages is not a promise to you. A trial reports what happened across a crowd; your body, your case, your circumstances are one point in that crowd, not the headline. You see the winner announced. You almost never see the rules of the match — the doses, the exclusions, the yardstick. The distance between “better on average, under these conditions” and “better for me” is exactly the distance the confident claim is built to hide. You are inside the comparison, being asked to read it as if you were standing above it.

Who gets to set the terms

Now the part that is easy to miss because it looks so normal. Who designs the trials that compare two drugs? Often the companies that make them. Who writes the ads that describe those trials? The same companies. The player picks the pitch, marks the lines, and then reports the score.

This is not a villain to unmask. It is the ordinary arrangement, and it can still produce true results — the trial was real, the win was real. But it means the word “objective” carries someone’s fingerprints. The one who chooses the conditions has a thumb resting on the scale before the weighing starts, whether or not they ever press down. It is precisely because this structure is normal that regulators, courts, and rival companies exist to argue about the setup after the fact — this week’s lawsuit is that argument, out loud.

What a single comparison can honestly tell you

So a comparison shows you one slice of the world, and someone chose which slice.

The trap on one side is cynicism — deciding every claim is a lie, so nothing can be trusted. That is just as blind, because plenty of comparisons are honest and useful. The trap on the other side is taking each claim at face value, as if numbers arrived without authors.

The humbler move sits between them. When a comparison lands in front of you, ask the quiet question the loud claim skips: what was set up here, and by whom? Against what, at what dose, in whom, measured how? You often can’t get the full answer. But holding the question is enough to keep the conclusion loose in your hand instead of certain in your head. A single “A beats B” is one photograph of a vast landscape, taken from an angle someone picked. It can be a true photograph and still show you almost nothing about the part of the ground you are standing on.

03 · Lab · your turn

Build the Winning Ad

Rehearse how choosing the doses and the yardstick lets you make either drug the honest "winner" — the setup decides before anyone is compared.

04 · Hope · carry this

It is a quiet kind of progress that a misleading health claim now draws a lawsuit rather than a shrug — the machinery for holding these claims to account is working, out loud, and the strongest results this week never needed the spin at all.