Connectome-based growth models reveal individual heterogeneity and neurophysiological subtypes of subthreshold depression

Subthreshold depression (StD) confers a high risk for major depression and is characterized by substantial individual clinical heterogeneity. However, the neurobiological substrates underlying this heterogeneity remain largely unknown. Using a large multisite resting-state functional MRI dataset including 1203 healthy participants and 197 individuals with StD, we constructed connectome-based normative models to identify individual brain deviations and biotypes in StD. We highlighted remarkable individual variability in the connectome deviations in StD, leading to the identification of two distinct biotypes. Subtype 1 exhibits severe positive deviations primarily in the default mode regions and negative deviations in the sensorimotor and ventral attention areas, while subtype 2 shows a moderate but opposite deviation pattern. The two subtypes differ significantly in depressive symptoms, gene expression profiles, and treatment responses to bright light therapy. These findings highlight the neurobiological underpinnings of the clinical diversity in StD, emphasizing the necessity for developing personalized interventions for this condition.