Obsessive-Compulsive Disorder (OCD) is a chronic and debilitating psychiatric disorder that affects approximately 1% of the population[1], characterized by recurrent and intrusive thoughts, impulses, or images (obsessions), and repetitive behaviors that are usually performed in response (compulsions)[2]. Individuals with OCD can experience a variety of symptoms that differ widely between people and may fluctuate in presentation and severity throughout their lives[2]. People with OCD also face significant clinical challenges, with an average diagnostic delay of 7 years in young people today[3], and well over 10 years in older adults[4]. Moreover, nearly half of patients fail to respond to first-line treatments[5,6,7]. These factors contribute to an elevated risk of suicide and increased overall mortality[8,9,10].

There is strong evidence for a genetic contribution to OCD, with twin-based heritability estimates reaching 47% [11,12,13,14]. Recent large-scale genomic studies have identified common and rare variants implicated in OCD, with rare variants making a substantial contribution to disease risk, accounting for ~ 10.4% of OCD heritability[15,16,17,18,19,20,21,22,23]. Exploring the genetic contribution to OCD risk may help identify biological processes involved, contributing to a better understanding of the disorder and informing the development of more effective treatments. The aim of this review is to outline recent advances in the genetics of OCD and their implications for clinical translation.

Current Insights Into OCD Based on Common Genetic VariantsGenetic Risk of OCD and Heritability

The genetic risk of OCD is increasingly being characterized through Genome-Wide Association Studies (GWAS, Glossary), which have identified multiple loci (Glossary) associated with disorder risk[22,24,25,26] (Table 1). The largest and most recent GWAS reported the first significant findings for OCD, identifying 30 independent loci[22]. These results support a highly polygenic architecture in which many common variants contribute small effects to disorder risk[22]. In this study, SNP-based heritability (Glossary) was estimated at 6.7% on the liability scale, lower than some previous SNP-based estimates[21,24,27]. This reduction likely reflects heterogeneity in phenotype definition, ascertainment strategies, and sample composition across cohorts. Heritability estimates were higher in clinically ascertained and comorbid subgroups, reaching 16.4% and 13.3%, respectively[22]. Polygenic Scores (PGS, Glossary) derived from this GWAS were significantly higher in cases than in controls (b = 0.41, 95%CI[0.35–0.47], p = 1 × 10− 16)[28].

GWAS have identified loci associated with OCD, yet translating these associations into putatively causal genes is complicated by the fact that the majority of associated SNPs fall in non-coding or intergenic regions. To address this, complementary gene-prioritization strategies can be applied. Positional mapping assigns variants to nearby genes based on genomic location, whereas functional mapping integrates biological annotations, such as gene expression data, to prioritize genes whose regulation may be influenced by associated variants. In the recent OCD GWAS, 251 genes were implicated by at least one of five gene-mapping approaches, and 48 genes were implicated by at least two approaches. Among these, 25 genes were further prioritized as putatively causal using a colocalization-based method. Evaluating these prioritized genes in relation to biological pathways and processes provides insight into the molecular and cellular mechanisms underlying OCD (Fig. 1).

Table 1 Summary of major genetic studies in OCD based on common and rare genetic variantsEvidence Implicating Brain-Related Processes in OCD

Several risk genes were mapped to brain-related processes, including neurodevelopmental and synaptic mechanisms (Fig. 1). Notably, CTNND1, prioritized across three gene-based tests and identified as putatively causal, showed strong colocalization with expression signals in the dorsolateral Prefrontal Cortex (dlPFC). This region has been systematically implicated in the neurocircuitry of OCD[34,35]. In addition, the two genes with strongest converging evidence for causality, WDR6 and DALRD3, are involved in neuronal and synaptic processes[36,37]. On the same genomic locus, CELSR3 was associated with OCD, and has previously been linked to Tourette Syndrome, which is closely related to OCD clinically and genetically, and recognized in the DSM-5 through the “tic-related” specifier for OCD[2,38,39]. CELSR3 was also implicated in rare variant studies of OCD (Sect.  Current insights into OCD based on rare genetic variants)[19].

Fig. 1

Current Genetic Evidence Informing the Neurobiology of OCD Based on Common and Rare Genetic Variants Note. This figure presents a hypothesis-driven synthesis inferred from genes prioritized for OCD. * = significant tissue enrichment; ** = significant cell-type enrichment. ACC, anterior cingulate cortex; CSTC, cortico-striato-thalamo-cortical; dlPFC, dorsolateral prefrontal cortex; MHC, Major Histocompatibility Complex

Established neurobiological models of OCD emphasize dysfunction in distributed brain circuits. The Cortico-Striato-Thalamo-Cortical (CSTC) circuit, comprising primarily the prefrontal cortex, striatum, and thalamus, has long served as the dominant framework for understanding OCD pathophysiology[34,35]. Enrichment analyses from the recent OCD GWAS further support CSTC circuitry in OCD pathophysiology[22]. Significant enrichment was observed in the frontal cortex, as well as in dopamine D1 and D2 receptor-expressing Medium Spiny Neurons (MSNs) within the striatum. Notably, rare variant studies of OCD also implicated these neuronal populations within the striatum (Sect.  Current insights into OCD based on rare genetic variants). Dopaminergic signalling within the striatum has long been associated with OCD as part of CSTC circuitry (Fig. 1)[40].

Accumulating evidence pointing to other networks in OCD, such as frontolimbic, frontoparietal, and cerebellar systems, suggest broader circuit-level dysfunction beyond the classical CSTC model[35,41]. Based on the recent OCD GWAS, significant tissue enrichment of the genetic signal associated with OCD was observed in cortical (Anterior Cingulate Cortex, Cortex) and subcortical regions (Amygdala and Hypothalamus)[22]. Notably, significant enrichment of excitatory neurons was found in the cerebral cortex and the hippocampus (Fig. 1)[22]. Consistent with this, two genes prioritized as putatively causal in the recent OCD GWAS, MEF2C and ZDHHC5, contribute to excitatory neuronal function[42,43,44]. Evidence from mouse models suggests that loss of ZDHHC5 reduces excitatory synapse density in the hippocampus[43]. These findings are consistent with glutamatergic pathways in OCD, consistent with prior OCD GWAS prioritizing glutaminergic genes[24]. Disruption of excitatory-inhibitory balance has been implicated in the pathophysiology of OCD[45].

These genetic findings are complemented by additional evidence from mouse models, Notably, SLITRK5-knockout, HOXB8-null mice and mice with a genetic deletion of SAPAP3 (DLGAP3) exhibited increased anxiety and compulsive grooming behaviour, mirroring OCD features[46,47,48,49,50]. These behaviors were alleviated by Selective Serotonin Reuptake Inhibitors (SSRIs), which are commonly used in the treatment of OCD. Moreover, SLITRK5- knockout mice showed selective overactivation of the orbito-frontal cortex, striatal anatomical abnormalities, and alterations in glutamate receptor composition, consistent with prior evidence regarding OCD neurobiology[48]. These findings may be relevant to humans. SLITRK5 and HOXB8 have been implicated in rare variants studies of Tourette Syndrome and Trichotillomania respectively (Sect.  Current insights into OCD based on rare genetic variants)[46, 51]. Both disorders are related to OCD genetically and classified within the Obsessive-Compulsive and Related Disorders (OCRD) diagnostic group[2]. Overall, although not all genes implicated in mouse models have been replicated in humans, they map on pathways consistent with established OCD neurobiology, including synaptic function, glutaminergic signaling, and significant enrichment in key brain regions.

Evidence Implicating Immune-Related Processes in OCD

Beyond genes implicated in brain-related processes, several risk genes for OCD map to the Major Histocompatibility Complex (MHC) locus, a region that is central to adaptive immune function[22]. A link between OCD and the immune function has been documented previously[52]. For example, many immune-related conditions sometimes present with obsessive-compulsive symptoms[53]. These include Pediatric Autoimmune Neuropsychiatric Disorders Associated with treptococcal infections, and Pediatric Acute-onset Neuropsychiatric Syndrome[52]. Consistent with this, significant genetic correlations were observed between OCD and several autoimmune disorders[22]. Notably, some of these associations were negative, specifically with Crohn’s Disease, Ulcerative Colitis, and Inflammatory Bowel Disease[22]. Family-based studies also supported this pattern, reporting a negative genetic correlation between OCD and Type 1 Diabetes[54]. However, an overall risk of autoimmune disorders was increased among relatives of individuals with OCD. This contrasts with GWAS-based findings[54], suggesting unresolved heterogeneity in this relationship. Current evidence suggests that immune involvement in OCD may characterize a specific subtype, rather than reflecting a unifying mechanism across the disorder as a whole[52].

The prioritization of MHC genes is not unique to OCD and has been observed in other psychiatric disorders including Autism Spectrum Disorder (ASD), Attention-Deficit/Hyperactivity Disorder (ADHD), and Anxiety Disorders[55,56,57]. However, the pattern of negative correlations is somewhat unusual compared with other psychiatric-immune relationships. Within the MHC locus, TRIM27 was prioritized as a putatively causal gene for OCD[22]. This gene has been associated with autoimmune diseases, particularly psoriasis and Crohn’s disease, the latter showing a negative genetic correlation with OCD[58]. In addition, QRICH1, an MHC gene associated with OCD in the recent GWAS, was also implicated in rare variant studies of OCD (Sect.  Current insights into OCD based on rare genetic variants) (Fig. 1). Mouse models have raised evidence for immune processes taking place in the brain that may contribute to OCD-like behavior[42,49]. These processes are thought to involve microglia, the resident immune cells of the brain. In these models, disruptions in microglial function of the HOXB8 and MEF2C genes induced OCD-like and repetitive behaviors, respectively[42,49]. Although the latter was originally studied in the context of ASD, MEF2C was implicated in the recent OCD GWAS[22].

Extending gene-level findings further, pathway analyses conducted on the 251 genes prioritized for OCD revealed significant enrichment in chromatin structure-related pathways[59]. Chromatin remodeling regulates several processes including neuronal differentiation, synaptic gene expression, and immune signaling. Therefore, disruptions may alter neurotransmission and immune function, potentially contributing to OCD symptoms and their variability in response to stress or infection. Evidence from rare variants also supports the involvement of chromatin-related processes in OCD (Sect.  Current insights into OCD based on rare genetic variants).

Insights from Obsessive-Compulsive Symptoms

While ~ 1% of individuals meet criteria for a clinical diagnosis of OCD, a substantially larger proportion of the population experiences subclinical Obsessive-Compulsive Symptoms (OCS). These are thought to lie on a continuum, with clinically diagnosed OCD representing the extreme end. Therefore, investigating the genetic architecture of OCS provides an opportunity to study larger and more representative population samples, and to assess the extent to which OCS and OCD share underlying biological mechanisms. Several GWAS have been conducted for OCS[30,32,60,61,62] (Table 1). SNP-based heritability estimates for OCS (~ 4.1%) are lower than for OCD (~ 6.7%)[23, 30], which is typically observed for non-clinical phenotypes[63]. Symptom-level GWAS have identified loci that have been mapped to specific risk genes[32,