ERROR-RELATED BRAIN ACTIVITY AS AN ENDOPHENOTYPE OF OBSESSIVE-COMPULSIVE DISORDER

Abstract

abstract Endophenotypes are conceived as measurable but not directly observable features of mental disorders that link clinical symptoms with their genetic underpinnings. They promise to guide the search for specific causal gene variants but also to understand mechanisms of symptom generation more precisely. Several criteria have been listed that must been fulfilled by a trait to qualify for an endophenotype, like association with the disorder, presence in biological relatives of patients with the disorder, independence of symptom state, and heritability. Obsessive-compulsive disorder (OCD) is etiologically poorly understood but appears to be partially under genetic control. Identification of endophenotype candidates is therefore a viable strategy to make progress in understanding OCD. We conducted a series of studies evaluating the applicability of the error-related negativity (ERN) as an endophenotype of OCD. The ERN is an event-related brain potential elicited by the commission of errors but also by actions that are potentially false. We found larger ERN amplitudes in OCD patients compared to healthy controls in several independent studies case-control studies, with effects sizes (Cohen’s d) between 0.5 and 1.0 [2] . A twin study showed that shared genetic factors account for 47% of ERN variance [1] . Associations of enhanced ERN amplitudes with specific gene variants are preliminary as yet pointing to effects of genes coding serotonergic and dopaminergic neurotransmission. In a family study, first-degree relatives of OCD patients also showed enhanced ERN [2] , [3] , what was independently confirmed by another research group. To examine time stability and independence from symptom state, we measured ERN before and after completion of CBT in a group of 45 OCD patients. While symptoms improved, ERN remained unchanged supporting the notion that ERN enhancement is not a consequence of acute symptom state. In a cross-sectional study, associations between specific subtypes or symptom dimensions of OCD and ERN amplitudes could not be detected. These findings let us conclude that ERN is a stable trait that may be on the pathway between OCD-relevant genes and clinical symptoms. Further studies were conducted to examine mechanisms that influence ERN amplitudes. Intermittent punishment of errors elicited larger ERN amplitudes, which remained stable throughout an extinction phase. Furthermore, the ERN enhancement in OCD disappeared when patients were given a second task concurrently, possibly due to diminished resources available for performance monitoring. The ERN appears the best-confirmed endophenotype candidate of OCD to date. Now, larger studies must follow to identify the genetic variants contributing to enhanced ERN, to locate the brain areas driving the ERN effect, and to elucidate the connections between characteristics of error monitoring and the generation of specific symptoms. Direct modification of the ERN is a new challenge that might result in improved treatment of OCD. Finally, as ERN enhancements are also seen in other disorders like generalized anxiety disorder, social anxiety disorder and depression, a transdiagnostic concept is needed to obtain better accounts of psychopathology in OCD and other disorders.