ReachRxResearch
Home
Research
Sign Up
  1. Home
  2. Research Hub
  3. The multivariate genetic architecture of…

The multivariate genetic architecture of psychiatric and insulin resistance multimorbidity

European Neuropsychopharmacology·July 11Open Access
PsychiatryPractice changingAnorexia NervosaAttention-Deficit/Hyperactivity DisorderInsulin ResistanceMajor Depressive DisorderMetabolic SyndromeObesityObsessive-Compulsive DisorderSchizophreniaType 2 Diabetes MellitusGenome-Wide Association Study / Genomic Structural Equation ModelingNMDA Receptor AntagonistThiazolidinedioneMixedNamendaMemantineRosiglitazone

Summary

View source

What was studied

This study used genomic structural equation modeling (SEM) on GWAS data from 5 psychiatric disorders (ADHD, anorexia nervosa, MDD, OCD, schizophrenia) and 3 insulin resistance (IR)-related conditions (metabolic syndrome, obesity, T2DM; N = 9,725–933,970) to map the shared genetic architecture of psychiatric-IR multimorbidity and identify biological pathways and repurposable drugs.

Key findings

A latent 'Psych-IR' genetic factor was identified, with positive loadings for ADHD, MDD, and IR-related conditions and negative loadings for AN and OCD; multivariate GWAS of this factor yielded 150 associated loci and 366 genes (128 novel), with drug repurposing analysis flagging 6 candidates including memantine and rosiglitazone.

Study limitations

The sample sizes varied widely across contributing GWAS (N = 9,725–933,970), which may unevenly influence factor loadings; the latent factor approach captures shared variance but cannot establish causal direction; findings are largely restricted to populations of European ancestry, limiting generalizability.

Clinical implications

Clinicians should be aware that ADHD and MDD share meaningful genetic overlap with metabolic conditions like obesity and T2DM, supporting routine metabolic screening in these patients. Repurposing candidates such as memantine and rosiglitazone warrant prospective study for psychiatric-metabolic comorbidity.

Caveats

  • Ancestry of the contributing GWAS is not explicitly stated in the abstract but is likely predominantly European, which limits generalizability; this should be confirmed in the full text.
  • Level of evidence is set to 3 (large-scale genetic epidemiology/multivariate GWAS); no experimental intervention or RCT is present, so causal inference is limited.
  • Sample sizes across contributing GWAS vary enormously (9,725 to 933,970), and the impact_flag 'practice_changing' is assigned cautiously — the drug repurposing findings are hypothesis-generating and not yet validated in clinical trials.
Show all 4
  • Study design label reflects a novel multivariate genomic SEM approach — not a standard GWAS; clinicians should interpret loci and gene findings as associative, not mechanistic.

Related Questions

Explore related topics

What is the evidence for metabolic screening in patients with ADHD or major depressive disorder?Are there clinical trials using memantine or rosiglitazone for psychiatric-metabolic comorbidity?How does the chr16p11.2 region relate to both neurodevelopmental and metabolic disorders?

Publication Details

Year
2026
Journal
European Neuropsychopharmacology
Sample Size
n=933,970
Source
View article
Keep scrolling
More content below.
Up Next