Prevention of Antipsychotic-Induced Weight Gain: Beyond Metformin

I have read with great interest the guideline published by Carolan et al. on the use of metformin to prevent antipsychotic-induced weight gain.¹ Its methodological robustness and stepwise approach constitute a valuable contribution to the preventive management of cardiometabolic risk in psychosis. We fully concur with the need for proactive intervention from the earliest stages. The aim of this letter is to reinforce and expand that framework by proposing that the algorithm explicitly incorporate the fatty liver index (FLI)² as an accessible tool for the early screening of metabolic dysfunction-associated steatotic liver disease (MASLD) and, in particular, its inflammatory phenotype, metabolic dysfunction-associated steatohepatitis (MASH).

The recent redefinition of non-alcoholic fatty liver disease (NAFLD) as MASLD has underscored its pathogenetic link to the metabolic syndrome beyond body weight alone.³ In a 3-year prospective study of first-episode psychosis, 25.1% of patients developed an FLI ≥ 60, which was associated with >7% increases in body mass index (BMI) and dyslipidemia.⁴ Subsequent data have shown that patients with an intermediate FLI experience unfavorable metabolic progression,⁵ an observation consistent with studies in Asian and non-obese populations.⁶ Unlike classical indicators of metabolic dysfunction—BMI, HbA1c, blood pressure, or lipid profile—the FLI provides an indirect estimate of hepatic steatosis by combining routinely obtained variables (Gamma-Glutamyl transferase (GGT), triglycerides, waist circumference, and BMI).

Integrating it into screening algorithms would allow the early detection of hepatic involvement—before the onset of MASH or shortly thereafter—and a more refined stratification of metabolic risk in this vulnerable population. In the original population-based derivation cohort, adding FLI to a model including BMI and triglycerides increased the area under the receiver-operating curve for ultrasound-defined steatosis from 0.79 to 0.84 (P < .01) and yielded a 9% net re-classification improvement.² Similarly, in the Rotterdam Study, an FLI ≥ 60 improved cardiovascular risk re-classification by 17% compared with the Framingham score and remained independently associated with incident NAFLD after adjustment for HbA1c, blood pressure, and lipid profile.⁷ In a biopsy-proven cohort, an FLI ≥ 60 conveyed an adjusted odds ratio of 3.4 (95% CI 2.1-5.2) for ≥F² fibrosis independently of BMI and glycemic status.⁸ The 2024 joint European Association for the Study of the Liver (EASL)/European Association for the Study of Diabetes (EASD)/European Association for the Study of Obesity (EASO) guidelines recommend its use as a first-line, non-invasive marker, together with the Hepatitis steatosis index (HSI) or Fibrosis-4 index (FIB4), in clinical contexts where MASLD is suspected.⁹ We acknowledge the diagnostic value of advanced techniques such as transient elastography (FibroScan®) and the FIB-4 index for the assessment of significant fibrosis;¹⁰ however, we consider them unsuitable for initial screening in outpatient psychiatric settings: they identify moderately advanced stages of disease and their availability is limited outside hepatology services.¹¹ By contrast, the FLI offers a cost-effective alternative that can be readily incorporated into routine clinical practice. We therefore propose that it be added—alongside other markers—to the original article as a complementary measure within the metabolic-dysfunction box.

Regarding treatment, it is well established that metformin provides only a modest weight reduction (3%-5%), and although a limited hepatic benefit has been suggested in the general population, no conclusive data exist on its impact on MASLD/MASH in patients with antipsychotic-induced weight gain.⁹ By contrast, glucagon-like peptide-1 receptor agonists (GLP-1RAs) confer multi-organ benefits, including ≈20% weight loss, histological resolution of MASH, and cardiovascular as well as renal protection.^12,13 Emerging psychiatric data are likewise encouraging: liraglutide shows good tolerability,¹⁴ and semaglutide achieves clinically meaningful weight loss in antipsychotic-treated cohorts,¹⁵ while a recent systematic review supports the class as a whole in this setting.¹⁵ Accordingly, we propose that an FLI ≥ 60 be formally added to the “advanced metabolic risk” box of Figure 1; once this threshold is reached, patients should transition directly from lifestyle measures or metformin to a GLP-1RA, consistent with current hepatology and endocrinology guidance and the proven histological benefits of this drug class.^12,13 This refinement does not replace the stepwise schema of Carolan et al. but rather sharpens it by introducing a liver-centric trigger that accelerates escalation to a therapy capable of simultaneously addressing obesity, MASH, and overall cardiometabolic risk.

In summary, we suggest:

1. Including the FLI as a complementary biomarker to traditional indicators in screening algorithms for metabolic dysfunction in patients treated with second-generation antipsychotics.

2. Adding an FLI ≥ 60 to the “advanced metabolic risk” box of Figure 1 and, once this threshold is reached, considering early escalation to GLP-1RAs—particularly when additional risk factors are present—as a strategy for expanded cardiometabolic prevention.

3. Prioritizing the development of dedicated clinical trials to validate this strategy in psychiatric populations and to identify early predictors of metabolic disorder with and without associated MASLD.

We appreciate the opportunity to contribute to this clinical debate and hope that our proposal will be regarded as a constructive extension of the excellent work by Carolan and colleagues.¹