Functional neuroimaging subtypes of obsessive-compulsive disorder: A systematic review and meta-analysis.

In 30 seconds

This systematic review and meta-analysis synthesized 40 functional neuroimaging subtyping studies involving approximately 8,150 patients with obsessive-compulsive disorder (OCD). Consensus clustering identified three neurobiological subtypes with distinct connectivity patterns, symptom profiles, and reported treatment responses: Limbic-Hyperactive, Fronto-Striatal-Hypoconnected, and Global-Disrupted. Support vector machine classification assigned patients to subtypes with 81.5% accuracy, while distinguishing OCD from healthy controls had limited multisite generalizability (AUC 0.567-0.673). The authors propose a framework for personalized treatment selection but note that prospective validation is needed.

Key findings

• Consensus clustering identified three subtypes: Limbic-Hyperactive (approximately 40%), Fronto-Striatal-Hypoconnected (approximately 35%), and Global-Disrupted (approximately 25%).
• Subtype assignment by support vector machine achieved 81.5% accuracy.
• OCD-versus-control classification showed limited multisite generalizability, with AUC 0.567-0.673.
• The subtypes differed in reported symptom profiles and treatment responses, including favorable CBT response in Limbic-Hyperactive and SSRI response in Fronto-Striatal-Hypoconnected.

Why it matters

The findings suggest that OCD may comprise neurobiologically distinct groups rather than a single uniform condition, with potential relevance to symptom patterns and treatment response. A reproducible imaging-based classification could eventually help refine treatment selection. However, the reported classification performance and multisite limitations mean these results should be viewed as a research framework, not an established clinical tool.

What to keep in mind

The authors state that prospective validation is required; OCD-versus-control classification also showed limited generalizability in multisite settings.