From Snapshots to Stable Outcomes: rs-fMRI-based Prognosis of Functioning in Patients with Psychosis Risk or Recent-Onset Depression.

Abstract

Background: Early recovery of functioning is critical for favorable outcomes in psychotic and affective disorders. Transdiagnostic brain activity patterns may capture pathways for poor outcomes before clinical manifestation, supporting timely prevention and intervention.

Methods: Using machine learning, we evaluated the transdiagnostic prognostic value of resting-state fMRI fractional amplitude of low-frequency fluctuations (fALFF, slow-5 and slow-4 sub-bands) for functional outcomes in patients at clinical high-risk for psychosis (n=217) or with recent-onset depression (n=198) from the multi-site PRONIA study. Leave-site-out cross-validation assessed geographic generalizability of models across disability and symptoms domains, with outcomes defined as 'snapshots' at 9- or 18-month follow-up or across both timepoints. We examined diagnosis-specific performance, generalization to recent-onset psychosis (ROP, n=140), and negative symptoms, and the added value of fALFF over clinical prognostication.

Results: Transdiagnostic models predicting stable good functioning across follow-ups showed up to 10% higher balanced accuracy (BAC) than 'snapshot' models. Decreased slow-5 fALFF in the default-mode, executive control (EC), and dorsal attentional (DA) networks, and increased fALFF in salience, EC, and DA networks predicted impairment with BAC=67% (Sensitivity=65%, Specificity=70%, P<.001). This model generalized to ROP (BAC=62%, Sensitivity=64%, Specificity=59%, P<.001) and predicted (BAC=65%, Sensitivity=66%, Specificity=65%, P<.001) and was mediated by negative symptoms. Slow-5-based models improved prognostic accuracy over expert ratings in disability (BAC_raters=66%, BAC_{raters+slow-5}=75%, W=1680, P<.001) and symptoms domains (BAC_raters=61%, BAC_{raters+slow-5}=71%, W=1444, P<.001).

Conclusions: We highlighted the prognostic value of fALFF for functional impairment in psychosis-risk and early depression. Leveraging trajectorial information, we identified candidate imaging biomarkers to improve prognostication, supporting personalized prevention and recovery strategies.