Summary: In 154 adults receiving second-generation antipsychotics for schizophrenia, with prediabetes and BMI at least 27, semaglutide improved HbA1c and weight versus placebo over 30 weeks. Reported treatment effects were -0.46 percentage points for HbA1c and -9.21 kg for weight; gastrointestinal symptoms were more frequent.
PICO Summary
| Element | Detail |
|---|---|
| Population | 154 adults aged 18-60 years with schizophrenia, second-generation antipsychotic treatment, HbA1c 5.7%-6.4% and BMI at least 27; recruited through community mental health services in two Danish regions. Mean age 38.3 years; 56.5% female. Seventy-seven participants were assigned to each group; 141 completed the trial. |
| Intervention | Subcutaneous semaglutide once weekly, titrated to 1.0 mg/week over eight weeks, for 30 weeks; n=77. This describes the trial regimen, not an individual prescribing recommendation. |
| Comparator | Matching placebo for 30 weeks; n=77. Antipsychotic-treated participants remained the population under study. |
| Outcomes | Primary outcome: change in HbA1c. Reported semaglutide treatment effect -0.46 percentage points (95% CI -0.56 to -0.36). Secondary body-weight effect -9.21 kg (95% CI -11.68 to -6.75). HbA1c below 5.7% was achieved in 81% versus 19% (P<0.001). Physical quality-of-life score improved by 3.75 points (95% CI 1.52-5.98; P=0.001), without a significant effect on mental quality of life or PANSS-6 schizophrenia symptoms. Gastrointestinal symptoms were more frequent with semaglutide. A few semaglutide recipients had more hospitalisations, although serious adverse-event counts did not differ between groups. This does not establish long-term psychiatric or cardiovascular safety. |
| Study Design | HISTORI: multicentre, double-blind, placebo-controlled randomised trial with 1:1 allocation and 30-week follow-up. NCT05193578. This is the primary HbA1c report, distinct from the later insulin-sensitivity analysis of the same trial; the reports are not independent populations. |
Expert Commentary
I regard HISTORI as useful evidence for addressing the metabolic burden accompanying antipsychotic treatment, rather than as a treatment for schizophrenia itself. The primary glycaemic result and the secondary weight result point in the same direction, and the selected population is clinically important. However, thirty weeks is too short to establish durable diabetes prevention or protection from cardiovascular events. Reaching an HbA1c below the prediabetes threshold is not equivalent to proving either outcome. Can I use this with my patients? The findings can inform a coordinated discussion between metabolic and mental health teams for adults resembling those enrolled, subject to the usual assessment of treatment eligibility and tolerability. They do not justify changing effective psychiatric medication or assuming that psychiatric monitoring can be relaxed. Gastrointestinal symptoms and the hospitalisation observations should be included in that discussion, rather than hidden behind a broad statement that treatment was safe. The absence of a significant PANSS-6 difference is reassuring within this trial, but does not exclude uncommon or delayed psychiatric effects. I would also avoid counting the later mechanistic HISTORI publication as an independent replication. Longer follow-up should establish whether these metabolic improvements persist and translate into fewer clinically important complications without compromising psychiatric stability.
References
Ganeshalingam AA, Uhrenholt N, Arnfred S, et al. Semaglutide Treatment of Antipsychotic-Treated Patients With Schizophrenia, Prediabetes, and Obesity: The HISTORI Randomized Clinical Trial. JAMA Psychiatry. 2025;82(11):1065-1074. doi:10.1001/jamapsychiatry.2025.2332. PMID: 40900607.
