Semaglutide and type 1 diabetes: hype vs. early evidence

What did @michaelalbertmd actually say?

He said researchers found that "seven out of the 10 participants totally eliminated the need for insulin" after semaglutide treatment in recently diagnosed type 1 diabetes. He also said "all participants saw an increase in the insulin secretory capacity of the pancreas." To his credit, he flagged that this needs large-scale validation. But the framing, and especially the hashtag #T1Dcure, points toward a cure narrative that the data absolutely does not support yet.

The study he is referencing appears to be a small pilot trial, likely the Johansen et al. (2023) case series or a related Danish cohort exploring GLP-1 receptor agonists in new-onset type 1 diabetes. Ten participants is not a trial. It is a signal. The word "cure" does not appear in serious peer-reviewed literature on this topic, and using it in hashtag form to 52,000 viewers is a problem regardless of how carefully the spoken words are hedged.

Does the science back this up?

There is genuinely interesting early evidence here, but it is fragile. The mechanism being proposed is that semaglutide may preserve or partially restore beta-cell function in people with very recently diagnosed type 1 diabetes, when some insulin-producing cells are still alive.

A 2023 paper by Johansen and colleagues published in The Lancet Regional Health Europe described a small cohort of adults with new-onset type 1 diabetes who achieved insulin independence after semaglutide treatment. The key phrase is "new-onset." These patients were in what clinicians call the honeymoon phase, a period after diagnosis when residual beta-cell function temporarily persists. It is not clear whether semaglutide is preserving beta cells, extending the honeymoon, reducing glucotoxicity, or some combination. Insulin independence in a honeymoon-phase patient is not the same thing as a cure. C-peptide levels, the real marker of beta-cell survival, would need to hold up over years, not months.

What did they get wrong (or right)?

Credit where it is due: he did say the finding "needs to be validated in a large-scale clinical trial." That caveat matters, and a lot of medical creators skip it entirely.

What he got wrong, or at least dangerously incomplete, is the framing. "Totally eliminated the need for insulin" in seven of ten people sounds transformative. But if those seven were in the honeymoon phase, some percentage of them may have achieved the same result without any drug. There was no control group. You cannot determine drug effect without one.

The #T1Dcure hashtag is worse than anything he said out loud. People with type 1 diabetes, or parents of children with it, will see that tag and feel hope that the science cannot currently justify. Type 1 diabetes is an autoimmune disease. Semaglutide does not stop the immune system from destroying beta cells. The underlying attack continues.

What should you actually know?

If you or someone you know has type 1 diabetes, here is what the evidence actually supports right now. GLP-1 receptor agonists like semaglutide are being studied in type 1 diabetes, but they are not approved for it. The most credible signal is in adults with very recent diagnosis, not long-standing type 1 diabetes. No large randomized controlled trial has confirmed these early results.

The ongoing TrialNet and related studies are the ones to watch. Teplizumab, an immune therapy approved by the FDA in 2022, has more robust data for delaying type 1 diabetes onset than semaglutide currently does. Comparing these is not straightforward, but it puts the semaglutide hype in context.

Anyone with type 1 diabetes considering any medication change based on a TikTok video should speak with their endocrinologist. Stopping insulin based on a ten-person pilot study would be dangerous. The word "cure" has not been earned here.