Thyroid dysfunction and Research Domain Criteria framework-based depression phenotypes: Insights from a historical...

Depression is a clinically heterogeneous disorder, making it difficult to identify its biological underpinnings, including potential links to dysregulation of the hypothalamic-pituitary-thyroid axis, which plays a central role in the pathophysiology of depression. The Research Domain Criteria (RDoC) framework provides a multi-dimensional approach to parsing depressive phenotypes. This study investigates the association between thyroid-stimulating hormone (TSH) abnormalities and depressive symptoms mapped to RDoC domains.

In a retrospective cohort from 2000 to 2021, adults with euthymic (PHQ-9 < 5) and euthyroid (TSH >0.3-4.2 mIU/L) baseline status and at least two TSH-PHQ-9 paired assessments within six months were included. PHQ-9 items were mapped to four RDoC-aligned domains: Negative Valence-Externalizing, Negative Valence-Internalizing, Arousal/Regulatory (AR) Systems, and Cognitive/Sensorimotor (CS) Systems. Linear mixed-effects models adjusted for covariates were used to assess the association between subsequent TSH dysfunction (low TSH ≤0.3 mIU/L and high TSH >4.2 mIU/L) and domain scores.

Among 6191 adults (mean age, 50.2 years; 67% female; 92.4% White), abnormal TSH levels are associated with higher AR Systems scores, with higher mean AR scores for low TSH (2.54; 95% CI, 2.38-2.70) and high TSH (2.57; 95% CI, 2.44-2.69) compared with normal TSH (2.40; 95% CI, 2.29-2.50; p ≤ .04). No significant differences were found in other RDoC domains.

Our findings support the utility of a biologically informed, dimensional framework in depression research. Elevated AR Systems symptoms in the context of abnormal TSH levels may suggest a possible narrower phenotype, with implications for personalized treatment if validated and further investigation into the thyroid-depression interface.