Can a GLP-1 Keep Prediabetes From Becoming Diabetes?

Your last physical came back with one new line on it: prediabetes. Maybe it was an A1C of 5.9%, or a fasting glucose that had crept into the 110s. Your doctor said something reassuring and something vague in the same breath — it's not diabetes, but let's keep an eye on it — and you walked out with a printout and a low hum of worry, especially if a parent has type 2.

If you've gone looking since, you've probably hit the same question I keep seeing: can one of these weight-loss shots really stop the slide? Tirzepatide, semaglutide, the Zepbound-and-Wegovy crowd. There's finally a three-year trial that speaks to it directly, and the answer is genuinely encouraging. It also comes with two catches that most headlines skip right past.

Here's the real number, both catches, and what someone sitting exactly where you are should take from it.

Prediabetes is a fork in the road, not a verdict

Prediabetes means your blood sugar is higher than normal but not yet in diabetes range. In the US, that's an A1C of 5.7% to 6.4%, or a fasting glucose of 100 to 125 mg/dL. It isn't a disease so much as a warning light, and a reversible one at that.

That last part matters. A fair number of people with prediabetes go on to develop type 2 diabetes within a few years. Plenty of others never do. The standard playbook for staying in the second group is well worn: lose some weight if you're carrying extra, move more, eat in a way you can sustain, and in some cases take metformin. None of that is glamorous, and all of it works to some degree.

So the honest framing of the GLP-1 question isn't "drug versus lifestyle." It's whether a medication that drives serious weight loss can bend the odds further than diet and movement alone, and what you give up to find out.

What three years of SURMOUNT-1 showed

SURMOUNT-1 was a phase 3, randomized, double-blind, placebo-controlled trial in 2,539 adults with obesity. The interesting subgroup for our question is the 1,032 who also had prediabetes at the start. They stayed on treatment for 176 weeks — a 72-week main phase plus a roughly two-year extension — and were then followed for about 17 weeks after the drug stopped.

Now the headline finding. Over the 176-week treatment period, just 1.3% of the tirzepatide group progressed to type 2 diabetes, against 13.3% on placebo, for a hazard ratio of 0.07. Eli Lilly, in its own summary, put that at roughly a 94% lower risk of progression across the pooled tirzepatide doses.

One nuance worth keeping straight, because it's where careful and careless reporting split. That 94% is the efficacy estimand: it leans on people who stayed on the drug. On the more conservative treatment-regimen basis, which counts everyone as randomized regardless of whether they stuck with it, the reduction works out to about 93%. Either way you slice it, the gap is large. And these aren't two separate wins: the 1.3% versus 13.3% gap (hazard ratio 0.07) is the treatment-regimen estimand — about a 93% lower risk — while Lilly's 94% is the pooled-dose efficacy estimand. Same finding, counted two different ways, not two different results.

In the prediabetes subgroup of SURMOUNT-1, type 2 diabetes showed up in 1.3% of people on tirzepatide versus 13.3% on placebo over 176 weeks. That is a real, sizable difference on a hard outcome — the kind of result that earns a closer look, and the closer look is the rest of this article.

Why the risk drops: it comes down to the weight

This isn't a mysterious blood-sugar trick. The mechanism is mostly the weight.

Excess body fat, especially around the middle, makes your cells less responsive to insulin. Your pancreas compensates by pumping out more, and for a while that keeps glucose in check. Until it can't. Then the numbers drift up into prediabetes, then diabetes. Take a meaningful amount of weight off and you reverse a chunk of that insulin resistance. Tirzepatide happens to be very good at taking weight off, which is the engine behind the diabetes result.

Those figures are the treatment-regimen estimand. Looking only at people who stayed on treatment (the efficacy estimand), the 15-mg dose averaged about −22.9% in the prediabetes subgroup, versus −2.1% on placebo. Different accounting, slightly bigger numbers, same story.

One thing to read carefully: those are percentages, not pounds. The trial reported percent change in body weight. For someone starting at 220 lb, a 19.7% drop pencils out to about 43 lb, but that's just arithmetic on an example I picked, not a number the trial published. Your starting weight sets your own figure.

Semaglutide points the same way (the STEP data)

Tirzepatide isn't the only molecule sending this signal, which is reassuring. One drug, one trial is a thinner branch to stand on than two.

Semaglutide, the molecule in Wegovy and Ozempic, was studied in the STEP 1 trial. Among participants who started with prediabetes, 84.1% on semaglutide 2.4 mg had returned to normal blood sugar by week 68, compared with 47.8% on placebo. That's a slightly different yardstick — "back to normoglycemia" rather than "didn't progress to diabetes" — but it leans in the same direction: drive the weight down with a GLP-1, and a lot of people move back toward the normal-glucose side of the line.

That's two molecules and two trials pointing the same way, which is a sturdier place to reason from than any single result.

Catch one: it's not an approved "diabetes-prevention" drug — yet

Here's the first place the headlines get ahead of the label. As strong as the SURMOUNT-1 result is, in the US the FDA has not approved any of these drugs to prevent diabetes. The approved uses are narrower than people assume.

Read that table again with prediabetes in mind. The weight-loss brands are cleared for obesity and a few weight-related conditions, not for preventing diabetes. The diabetes brands are cleared for managing blood sugar in people who already have type 2 diabetes, also not for preventing it. So using any of them in prediabetes, where you don't yet have diabetes, is off-label by definition. That's legal and not unusual, but it's a different thing from "FDA-approved to stop diabetes," and it shapes coverage and cost in ways we'll get to.

A note for readers outside the US: the FDA only speaks for the American market. Regulators in the EU, the UK, the Gulf, and elsewhere draw their own indication lines, and approvals shift over time. Whatever your local authority has cleared these drugs for, "diabetes prevention" is unlikely to be on that list today.

Catch two: stop the drug, and the protection softens

The second catch is the one that quietly changes the whole calculation. The protection isn't a one-and-done. It runs on the drug.

SURMOUNT-1 followed people for about 17 weeks after treatment ended. In that off-drug window, type 2 diabetes turned up in 2.4% of the former tirzepatide group versus 13.7% on placebo, for a hazard ratio of 0.12. Still a clear gap. But notice it's weaker than the 0.07 seen during treatment, and the weight that had come off started creeping back, with some people drifting from normal blood sugar back into prediabetes.

The way to hold this: a GLP-1 here behaves more like a blood-pressure pill than an antibiotic. It manages a risk for as long as you take it. Stop, and the effect doesn't vanish overnight, but it does begin to fade — because the weight does.

That reframes prediabetes treatment with these drugs as a long-term commitment, not a short course. Which is exactly the kind of thing to weigh up front, not discover a year in.

So you're prediabetic — what now?

Put the two catches together and the picture isn't discouraging, it's just honest. The drugs work. They aren't a quick fix or an official prevention tool, and the benefit tracks how long you take them. So how do you actually use that?

Start with the boring foundation, because it's the one part that's free, carries no off-label asterisk, and makes everything else work better. The standard prevention toolkit — losing some weight, regular movement, fewer refined carbs, decent sleep — isn't the consolation prize you fall back on if you can't get a GLP-1. It's the base the medication sits on top of, and the part that keeps doing something even on a week you skip the gym. Think "and," not "or."

Metformin is the other low-cost option your doctor may raise. It's been used in prediabetes for years, it's inexpensive, and it has a long safety record, even though it's gentler on weight and blood sugar than a GLP-1.

The GLP-1 conversation gets more serious if you also carry obesity, since that's where these drugs are actually approved and where the weight effect is largest. If a weight-loss dose is on the table for your weight anyway, the prediabetes benefit comes along as a meaningful bonus, which is essentially the population SURMOUNT-1 studied. None of this is a decision to make from an article. It's the agenda for a visit, with your A1C, your weight, your family history, and your budget all in the room.

Cost and access, realistically

This is where the off-label status bites. In the US, list prices for these drugs run north of $1,000 a month, and because preventing diabetes isn't an approved use, insurance is unlikely to cover a prescription written for prediabetes alone. People who pursue it often end up paying cash, sometimes through manufacturer direct-pay options that come in lower than list but are still a real monthly line item.

Coverage usually opens up only when there's an approved indication attached — obesity, for instance, or type 2 diabetes once it actually develops. That's a frustrating irony: the surest path to having it covered can be to wait until your prediabetes becomes the disease you were trying to avoid. It's a real bind, and naming it is more useful than pretending the price tag is small.

Outside the US, the math differs. National systems like the UK's NHS, and other public and private insurers, set their own rules on who qualifies and what's reimbursed. The drug may be the same molecule worldwide; what you'll pay for it is anything but uniform.

Questions worth bringing to your doctor

You'll get a better conversation by walking in with specifics than with "should I be on the weight-loss shot?"

• Given my A1C, weight, and family history, how high is my actual risk of progressing to type 2 diabetes over the next few years?
• Does the SURMOUNT-1 data change what you'd consider for someone in my situation, or is lifestyle change the right first move for me?
• I have prediabetes but not obesity — does that change whether a GLP-1 is even appropriate, or whether it'd be covered?
• If we did start one, what's the realistic plan if I ever need to stop, given that the effect fades off the drug?
• Where does metformin fit for me, and how would we know if it's enough on its own?
• What would this cost me out of pocket, knowing insurance probably won't cover it for prediabetes?

None of these are gotcha questions. Any decent clinician will be glad you read the trial instead of the headline — it means the decision gets made with you, not handed to you.

Where this leaves you

The data is good news, and worth stating plainly. In people with both obesity and prediabetes, tirzepatide cut three-year progression to type 2 diabetes to 1.3% versus 13.3% on placebo, and semaglutide moved most of a prediabetes group back to normal blood sugar in STEP 1. If you carry extra weight along with that borderline A1C, this is a real tool, not hype.

The two catches keep it honest. It isn't approved to prevent diabetes, so using it there is off-label and usually uninsured. And the protection leans on the drug: stop, and the risk starts climbing back toward where it began. Both halves are true at the same time; neither cancels the other.

So the move isn't a pharmacy and it isn't panic. It's a conversation where your own numbers finally enter the picture, with lifestyle change running underneath whatever you decide. Prediabetes is a fork in the road precisely because you still get to choose the turn.

This article draws on published clinical trials and peer-reviewed research, including SURMOUNT-1 and STEP 1, and is meant for information rather than medical advice. Any decision to start, stop, or change a medication belongs with your doctor, who can weigh it against your full history.

Start managing your GLP-1 with Blueshot

AI coaching, injection scheduling, and weight tracking in one app