OCS is both be common in clozapine-treated patients and associated with psychosis severity and clozapine load. Here, mediation analyses suggested that psychosis severity generates checking behaviour indirectly by inducing obsessions. After psychosis remission, checking compulsion correlated with clozapine plasma levels. The transition from psychosis to obsession and compulsion was found to be moderated by genetic variants associated with serotoninergic and glutamatergic transmission. To our knowledge, the association between clozapine-associated OCS/OCD and serotonin genetic variants has not been previously described.
This study has some limitations. Firstly, its naturalistic design is inferior to an experimental study. Experimental methodologies, however, are unfeasible as the potentially lethal side effects of clozapine prohibit its administration to healthy volunteers. Long-term longitudinal studies in schizophrenia patients after clozapine initiation could provide a more detailed characterisation, but the latency of the OCS onset (up to a decade) may render such studies impractical. Given this, we consider the large sample and longitudinal follow-up of this study a good balance between practical feasibility and rigour. Secondly, OCS were evaluated using a self-rated scale, blind to the clinician, and further work might benefit from replication using a clinician-rated scale such as Y-BOCS. Again, however, this may be impractical in a busy clinical environment. Thirdly, the patient sample included in this study’s genetic analyses was relatively small (n=100), and these findings should therefore be taken as exploratory and in need of replication.
Our results broadly align with previous research in terms of OCS prevalence (52%) and the predominance of checking and obsessing symptoms. Our sample is representative of a typical clozapine-treated cohort, with male predominance (79%), an average prescribed clozapine dose of ~330 mg/day, typical treatment length (~14 years) and frequent prescription of a second antipsychotic (35%). However, this study’s sample size (204 cases) and volume of standardised OCS assessments (734 face-to-face assessments) is more extensive than any previous research.
We considered clozapine-associated OCS as a dynamic phenomenon that fluctuates in intensity according to psychosis severity, in line with recent work by Schirmbeck and colleagues [12], and is associated with the clozapine load. We dissected OCS into its two main components (obsessions and compulsions) and applied a mediation model that identified an association of psychosis severity (particularly severity of reality distortion symptoms, such as persecutory delusions) with obsessive thoughts. Obsessive thoughts, in turn, precipitated checking behaviour. This may initially be understood as a goal-directed, safety-seeking behaviour in acutely psychotic, paranoid patients. As one patient explained; ‘I need to check everything is in place as my upstairs neighbour comes to steal my stuff’. In patients achieving psychosis remission, we found that checking severity correlated with clozapine plasma levels, suggesting a role of clozapine in perpetuating checking as a non-goal directed action, or habit. Indeed, persistently impaired safety signalling in OCD has been described [33], limiting patients’ ability to assign safety after verification.
Appreciating the distinct roles of psychosis and clozapine in OCS development is essential to understand the apparent discrepancies in previous cross-sectional studies. Point-prevalence based on questionnaire scores fluctuate according to the severity of the psychosis (figure 1 and 2), clozapine load (figure 3), concomitant medications (e.g. antidepressants) or even the tools used to assess the symptoms. In this sense, for instance, a psychosis-driven, goal-directed checking might not be considered OCS but part of a delusion. Here we circumvented this risk, by using a patient-reported OCS questionnaire.
Our hypothesised two-phase process of OCS development is based on cognitive neuroscience frameworks of habit formation [23, 24] and shows the potential for embedding cognitive neuroscience into clinical practice. In our context, clozapine treated patients in psychosis remission would experience checking compulsion as an antipsychotic-induced habit. Notably, a two-phased model integrates not only the present results but also the results of previous studies [32][15, 16, 35] and patients’ narratives, in which psychosis-induced goal-directed behaviour could become a habit. This model is flexible regarding the content of the repetitive behaviour as it would also accommodate other triggers, such as the influence of stressful events recently reported [36].
A plausible mechanism for the persistence of repetitive behaviour following psychosis remission may be clozapine’s antagonism of 5-HT2C and 5-HT2A receptors [1]. A decrease in serotonin neurotransmission causes perseveration in reversal learning tasks, the hallmark of cognitive inflexibility in compulsive behaviours, including OCD [37]. In humans, dietary tryptophan depletion (which acutely reduces serotonin in the brain) promotes habitual over goal-directed control [38]. Remarkably, a recent mouse-model study showed similar results. Mice receiving clozapine significantly increased their grooming time (as a proxy of compulsion). This behaviour then reverted on administration of fluoxetine, a selective serotonin reuptake inhibitor. Interestingly, the same effects were seen in both the wild type and the Sapap3 knockout mice (a well-known animal model of OCD), suggesting a genetic vulnerability to clozapine-induced habit formation [39].
The genetic vulnerability was explored in this work, in which we identified a distinct role of the serotonin and glutamatergic SNP on psychosis-to-obsession and obsession-to-compulsion transition respectively (figure 4 and supplementary material). We replicated previous findings involving the glutamate pathway [9], such as GRIN2B. More importantly, we are here first reporting the serotonin pathway involvement, as described in pure OCD. Importantly, however, drawing firm conclusions about the role of specific variants will require replication in larger samples. Nevertheless, our identification several genetic variants of the serotonin pathway (SCL64A, 5HTR2A and 5HTR2C) moderating the psychosis-obsession-compulsion transition is notable and may offer clues to future preventative or therapeutic approaches. The converging evidence form this study, and others [39], indicates that interventions directed to enhancing serotonin function are crucial for the effective treatment of clozapine-induced OCS [39], as in pure OCD. Nevertheless, we did not explore specifically the effect of medication modifications in the OCS severity, and further research is needed in this area.
In conclusion, the onset of significant OCS in clozapine treated patients is a puzzling phenomenon in which people suffering from one disorder (schizophrenia) seem to transition to a second (OCD). Here, we offer an explanatory model for this phenomenon informed by cognitive neuroscience’s insights surrounding habit formation, in which initial compulsions arises as a by-product of florid paranoid psychosis, and are then perpetuated in predisposed subjects after psychosis remission by clozapine’s anti-serotoninergic action. Better understanding the different phases of the phenomenon may inform clinicians' therapeutic decisions.