Multidimensional Epigenetic Clocks Demonstrate Accelerated Aging Across Physiological Systems in Schizophrenia: A Meta-Analysis

Importance: Schizophrenia is associated with increased age-related morbidity, mortality, and frailty, which are not entirely explained by behavioral factors. Prior studies using epigenetic clocks have suggested that schizophrenia is associated with accelerated aging, however these studies have primarily used unidimensional clocks that summarize aging as a single “biological age” score.

Objective: This meta-analysis uses multidimensional epigenetic clocks that split aging into multiple scores to analyze biological aging in schizophrenia. These novel clocks may provide more granular insights into the mechanistic relationships between schizophrenia, epigenetic aging, and premature morbidity and mortality.

Study selection: Selected studies included patients with schizophrenia-spectrum disorders and non- psychiatric controls with available DNA methylation data. Seven cross-sectional datasets were available for this study, with a total sample size of 1,891 patients with schizophrenia and 1,881 controls.

Data extraction and synthesis: Studies were selected by consensus Meta-analyses were performed using fixed-effect models.

Main outcomes and measures: We analyzed multidimensional epigenetic clocks, including causality- enriched CausAge clocks, physiological system-specific SystemsAge clocks, RetroelementAge, DNAmEMRAge, and multi omics-informed OMICmAge. Meta-analyses examined clock associations with schizophrenia disease status and clozapine use, after accounting for age and sex.

Results: Overall SystemsAge, CausAge, DNAmEMRAge, and OMICmAge scores demonstrated increased epigenetic aging in patients with schizophrenia after strict multiple-comparison testing. Ten of the eleven SystemsAge sub-clocks corresponding to different physiological systems demonstrated increased aging, with strongest effects for Heart and Lung followed by Metabolic and Brain systems. The causality- enriched clocks indicated increases in both damaging and adaptive aging, though these effects were weaker compared to SystemsAge scores. OMICmAge indicated changes in multiple clinical biomarkers, including hematologic and hepatic markers that support system-specific aging, as well as novel proteins and metabolites not previously linked to schizophrenia. Most clocks demonstrated age acceleration at the first psychotic episode. Notably, clozapine use was associated with increased Heart and Inflammation aging, which may partially be driven by smoking. Most results survived strict Bonferroni multiple testing correction.

Conclusions and relevance: These are the first analyses of novel multidimensional clocks in patients with schizophrenia and provide a nuanced view of aging that identifies multiple organ systems at high risk for disease in schizophrenia-related disorders.