The Number That Moved The Goalposts

The bar was five percent

Most people have no idea how low the bar used to be.

For an obesity drug to be considered a success by regulators, it historically needed to beat placebo by around five percent of body weight. Five percent. On a 100 kilogram person, five kilograms, and you had a viable product.

That number was not cynicism. It was a realistic read of what anyone could deliver, and five percent genuinely does move blood pressure, triglycerides and diabetes risk. The bar was set where the science was.

Here is what the science was, laid out in order.

The ladder

The early generation. Orlistat, phentermine-topiramate, naltrexone-bupropion. Approved for long-term use, and honest about what they were. Single digit percentages, with side effect profiles patients frequently abandoned.

Liraglutide, 3.0 mg daily. Average 8.40 kg over 56 weeks. A daily injection, roughly a year, for about eight kilograms. In its moment that was a real achievement.

Semaglutide, 2.4 mg weekly. Average 15.3 kg over 68 weeks. This is the one that became a household name, and it nearly doubled liraglutide.

Tirzepatide, the dual agonist. GLP-1 plus GIP. HbA1c reductions of 1.06 to 1.94 percent, weight loss from 0.9 up to 11.3 kg depending on dose.

Look at the shape of that. Decades of grinding, and each generation buys you a few more kilograms than the last. Steady, incremental, respectable.

Then the phase 2 retatrutide data arrived.

24.2 percent

Jastreboff and colleagues ran 338 adults with obesity, or overweight with weight-related conditions, on weekly retatrutide or placebo for 48 weeks.

At the top dose of 12 mg, mean body weight reduction was 24.2 percent.

Not 24.2 kilograms in a heavy cohort. 24.2 percent of body weight, on average, in under a year.

The review makes the comparison that everyone in the field made privately: that figure sits in the range previously reserved for bariatric surgery. The authors note the obvious downstream implication, that fewer patients may end up in an operating theatre, with the costs and complications that involves.

Set it against the ladder and the discontinuity is the story. Semaglutide roughly doubled liraglutide. Retatrutide did not extend the line. It stepped off it.

The number nobody put in the headline

Weight was not the most interesting outcome in that trial.

Among participants who entered with prediabetes, 72 percent reached normal blood glucose by week 48, against 22 percent on placebo.

Prediabetes is the waiting room for a lifelong condition, and for most of medical history the exit from that room was a lifestyle programme with modest, poorly maintained success rates. Seven in ten, in under a year, on a weekly injection, is a different category of claim.

A separate phase 2 trial by Rosenstock ran 281 people who already had type 2 diabetes, against placebo and against dulaglutide. HbA1c reductions beat both comparators at most doses at 24 weeks, and weight reductions beat both at 36 weeks.

The head-to-head against dulaglutide at 36 weeks is worth stating plainly:

Comparable glucose control, eight times the weight loss.

It was not only the scale

The metabolic panel moved with it. Waist circumference down as much as 13.2 cm. Systolic blood pressure down up to 8.8 mmHg. Diastolic down up to 3.9 mmHg. Triglycerides down as much as 35 percent.

That collection of numbers has a name. It is the metabolic syndrome, and it is normally managed with four separate prescriptions, four sets of side effects, and four appointments.

Then phase 2 had to survive phase 3

This is where most exciting drugs go quietly wrong. Phase 2 runs on a few hundred carefully selected people. Phase 3 runs on thousands of ordinary ones, for longer, and the effect usually shrinks.

The review, written in 2024, could only point forward at the TRIUMPH programme and say the phase 3 data would decide it.

That data has now arrived.

TRIUMPH-1 enrolled 2,339 participants on 4 mg, 9 mg, 12 mg or placebo. At 80 weeks:

• 4 mg: 17.6 percent
• 9 mg: 23.7 percent
• 12 mg: 25.0 percent
• Placebo: 3.9 percent

The phase 2 figure held. At seven times the sample size and nearly twice the duration, 24.2 percent became 25.0 percent.

A subgroup of 532 participants with a BMI of 35 or above who tolerated their dose were escalated further and followed to 104 weeks, reaching up to 30 percent body weight reduction. On the 12 mg arm, 45.3 percent of participants lost more than 30 percent of their body weight.

Nearly half of one arm, past thirty percent.

What actually changed

The interesting thing is not that a bigger number exists. Bigger numbers arrive regularly and most of them do not replicate.

The interesting thing is what the conversation is now about.

For sixty years the question was how to get someone to lose fifteen pounds and keep them off. That question consumed careers, industries and an enormous amount of public blame aimed at individuals who were, as Part 1 laid out, fighting a control system rather than a habit.

The question on the table now is how much weight loss is too much, whether these drugs should be taken indefinitely, and what the cost and access picture looks like when the effective option is a permanent weekly injection.

Those are real problems. They are also a completely different set of problems, and you only get to argue about them once the first question stops being the bottleneck.

Before you get carried away

Every number above is a mean. Means conceal the people at both tails, and a 25 percent average contains people who lost far less.

And there is a column in these trials that the headlines consistently leave out. People stopped taking this drug, and the reasons matter.

Part 4 covers the organ that turned out to respond even more dramatically than the scale did. Part 5 covers the column nobody prints.

This series covers a peer-reviewed narrative review of published trial data plus subsequently released phase 3 topline results. Retatrutide has not been approved by the FDA. Nothing here is medical advice.