JAMA Netw Open. 2025 Dec 1;8(12):e2548809. doi: 10.1001/jamanetworkopen.2025.48809.
ABSTRACT
IMPORTANCE: Substance use disorders (SUDs) affect millions of of people worldwide and are marked by high relapse rates. Cue-elicited brain activity may correlate with treatment success, but no meta-analysis has identified which brain regions are consistently associated with outcomes.
OBJECTIVE: To evaluate, across the existing literature, whether functional magnetic resonance imaging (fMRI)-based cue reactivity is associated with SUD treatment response (eg, predictive biomarker) or posttreatment neural changes (eg, response biomarker).
DATA SOURCES: A systematic search of PubMed, PsycINFO, Embase, and Cochrane databases covered studies published from database inception through May 2023.
STUDY SELECTION: Studies were included if they reported fMRI-based cue reactivity and treatment outcomes for adults undergoing SUD treatment.
DATA EXTRACTION AND SYNTHESIS: Data extraction followed PRISMA guidelines. Analyses were conducted from October 2024 to January 2025 using Seed-Based d Mapping With Permutation of Subject Images to quantify the magnitude and extent of activation to substance-related cues across studies. Random-effects models were used to generate statistical activation maps.
MAIN OUTCOMES AND MEASURES: Statistical maps of cue-induced activation during drug cue presentation at baseline correlating with a treatment outcome (for predictive biomarkers) as well as in brain regions exhibiting a significant change after treatment (for response biomarkers). Statistical inference used permutation statistics, with thresholds set at P < .005 and cluster extent of 10 voxels.
RESULTS: Fifty-one studies (1787 participants) were included. Cue-elicited activation was observed across studies in the cingulate gyrus, middle frontal gyrus, caudate, and insula. Predictive biomarker regions (n = 14 studies, 456 participants) included the bilateral insula and cingulate, in which lower baseline cue reactivity was associated with better treatment outcomes. Response biomarkers (pretreatment to posttreatment analyses; n = 25 studies, 609 participants) indicated significant reductions in cue reactivity in the anterior cingulate, caudate, accumbens, and insula. Cue-elicited activation in the insula emerged as both predictive and response biomarkers.
CONCLUSIONS AND RELEVANCE: This meta-analysis of cue-induced brain activation in adults with SUDs identified specific regions consistently engaged during cue exposure that may serve as candidate predictive and response biomarkers for treatment outcomes, highlighting the potential role of brain-based biomarkers in informing interventions for SUD.
PMID:41379443 | DOI:10.1001/jamanetworkopen.2025.48809
