GLP-1s and Prediabetes: Where the Evidence Stands

Two facts sit side by side in this subject and refuse to merge. The first: no GLP-1 medicine is approved for prediabetes, and the word does not appear in any of the current labels. The second: there is a substantial body of trial evidence in people who had prediabetes, and it points consistently in one direction. Articles usually pick the fact that suits their argument and drop the other. This page keeps both, because the gap between them is the actual subject.

What prediabetes is, briefly

Prediabetes is a label applied when blood glucose values are higher than normal but not in the range used to diagnose diabetes. This site does not interpret lab values, and it will not start here: what counts as higher than normal, measured how and judged against what, is a question for the clinician ordering the test, and Blood Work on a GLP-1: Numbers Worth Keeping covers how to keep the record straight once that conversation exists. What matters for this page is only the category: a range of values, not a fixed state, and one that a person's own results, and only those, place them in or out of.

What the labels say, which is nothing

The word prediabetes does not appear anywhere in the current prescribing labels for Wegovy, Ozempic, Zepbound, Mounjaro or Saxenda. There is no prediabetes indication on any of them. The approved uses are consolidated at What GLP-1s Are Actually Approved to Treat, and prediabetes is not on the list.

That has a consequence worth stating plainly. Any use of these medicines specifically to prevent diabetes in someone with prediabetes is off-label: legal, common, and prescriber-driven, but outside what the label describes. The term is defined in GLP-1 Words Decoded: A Plain-English Glossary, and its main practical effect here is that the label, the document written to tell patients and prescribers what a medicine is for, has nothing to say about this purpose. Off-label prescribing is a judgment a clinician makes with an individual's history in front of them, and this article recommends nothing either way.

What the trial analyses reported

Now the other fact. Across the large weight-management and cardiovascular outcomes trials, the researchers examined what happened to participants who entered those trials with prediabetes, and the findings have been consistent. Fewer of those participants progressed to type 2 diabetes on the medicine than on placebo, and more of them returned to normal glucose values. The direction holds across semaglutide analyses from the STEP weight-management programme and the SELECT cardiovascular outcomes trial, and across tirzepatide analyses from SURMOUNT-1, one of which followed its prediabetes cohort well beyond the original treatment period.

What this article will not do is attach a figure to that sentence, and the reason deserves explaining rather than apologizing for. Published results for the size of the effect vary between analyses, between populations and between time points, and several of the most-circulated numbers come from company announcements rather than peer-reviewed primary reports. The direction is consistent and well supported. The magnitude depends on which analysis you read. A page that quotes one number for this question is offering a confidence nobody has, and we prefer the honest version: consistent direction, unsettled size.

Two cautions travel with the finding itself. These analyses describe groups of trial participants, and a statement about groups cannot be converted into a statement about any reader. And the word prevention is doing work the evidence has not paid for: what was observed is that fewer people in the medicine groups crossed the diagnostic line during the observation period, which is not the same claim as that the medicines prevent diabetes.

GLP 1 Tracker AppWeight, doses and how you felt, dated and private on your phone. Whatever question you bring to an appointment, the record is what makes the conversation specific. Get the app

What kind of evidence this is, and the weight underneath it

One more structural point matters as much as the results. The STEP, SELECT and SURMOUNT trials were designed to study weight and cardiovascular outcomes; the prediabetes findings come from analyses of subgroups within those trials, some planned in advance and some conducted afterward. A trial designed from the start to test diabetes prevention would enroll people with prediabetes, define progression as its primary outcome, and run long enough to measure it. That trial is not what these analyses are.

This is not a dismissal; subgroup analyses of large trials are real evidence, and when they point the same way across programmes, molecules and years, that consistency means something. But it is a description of rank. Regulators approve indications on the strength of trials built to answer the question at hand, which is a large part of why the gap in the first paragraph exists: encouraging subgroup findings on one side, an indication requiring different evidence on the other. Both facts are true at once, and neither cancels the other.

There is one more honest complication, and the analyses themselves raise it. Much of the apparent effect appears to travel with the weight loss rather than being independent of it: participants who lost more weight tended to be the ones whose glucose values improved. That matters for a practical reason. If the benefit rides substantially on weight loss, then what happens after the medicine stops is an open question in a way it would not be for a direct, weight-independent effect on glucose metabolism. The analyses do not settle that question, and neither does this page.

It also explains why this subject resists tidy headlines. A reader who hears that a medicine fixes prediabetes has heard a claim about mechanism that the subgroup analyses do not make. A reader who hears that losing weight improves glucose has heard something long established in general medicine that these trials happened to illustrate again, at scale, in a population on these medicines. The second framing is closer to what the evidence supports, and it is noticeably less exciting.

Who decides, and what this means for you

Nothing on this page tells an individual reader what their own results mean or what their trajectory will be. Prediabetes is a range, not a fixed state; whether a person sits in it, moves within it or leaves it is a matter of their own lab record read by their own clinician. Whether a GLP-1 is part of that picture is a prescriber's judgment, weighing history, other conditions, and everything else a label cannot see. This article recommends no prescription and discourages none, and it will not close with a nudge dressed as a summary.

The one honest closing observation is this. A reader with prediabetes who is already taking one of these medicines for weight is in the population those analyses looked at, which is genuinely interesting, and is still not a prediction. The right use of that fact is conversational: it is a reasonable thing to raise with a prescriber, alongside the lab record, at an appointment prepared the way GLP-1s and Blood Sugar: When Lows Become Possible describes. The prescriber has the label, the analyses and your history. This page has only the first two, which is exactly why it stops here.