Abstract:Background: Antipsychotic drugs are the first-choice therapy for psychotic episodes, but antipsychotic treatment response (AP-R) is unpredictable and only becomes clear after weeks of therapy. A biomarker for AP-R is currently unavailable. We reviewed the evidence for the hypothesis that functional magnetic resonance imaging functional connectivity (fMRI-FC) is a predictor of AP-R or could serve as a biomarker for AP-R in psychosis. Method: A systematic review of longitudinal fMRI studies examining the predictive performance and relationship between FC and AP-R was performed following PRISMA guidelines. Technical and clinical aspects were critically assessed for the retrieved studies. We addressed three questions: Q1) is baseline fMRI-FC related to subsequent AP-R; Q2) is AP-R related to a change in fMRI-FC; and Q3) can baseline fMRI-FC predict subsequent AP-R? Results: In total, 28 articles were included. Most studies were of good quality. fMRI-FC analysis pipelines included seed-based-, independent component- / canonical correlation analysis, network-based statistics, and graph-theoretical approaches. We found high heterogeneity in methodological approaches and results. For Q1 (N = 17) and Q2 (N = 18), the most consistent evidence was found for FC between the striatum and ventral attention network as a potential biomarker of AP-R. For Q3 (N = 9) accuracy's varied form 50 till 93%, and prediction models were based on FC between various brain regions. Conclusion: The current fMRI-FC literature on AP-R is hampered by heterogeneity of methodological approaches. Methodological uniformity and further improvement of the reliability and validity of fMRI connectivity analysis is needed before fMRI-FC analysis can have a place in clinical applications of antipsychotic treatment.

fMRI connectivity as a biomarker of antipsychotic treatment response: A systematic review