We read with interest the paper by Adachi et al ^{Reference Adachi, Akanuma, Ito, Kato, Hara and Oana1} in which they conclude that early development of interictal psychosis in people with epilepsy may reflect individual vulnerabilities to psychosis, including genetic, rather than being driven by epilepsy-related damage.

As they point out, their conclusion challenges traditional assumptions about the relationship between epilepsy and psychosis, many of which have been based on relatively sparse data. It is of interest that recent molecular genetic findings in psychosis suggest that the comorbidity of psychotic symptoms and epilepsy is a product of shared underlying biological mechanisms. For example, specific genomic structural variants (copy number variants) have been described that predispose to schizophrenia, epilepsy, as well as some other ‘neurodevelopmental’ phenotypes such as autism and intellectual disability. ^{Reference Craddock and Owen2} Individuals with such structural variants do not typically have both schizophrenia and epilepsy, but rather some with a variant have schizophrenia, others have epilepsy, and others have a different phenotype or are unaffected. This means that the relationship cannot be caused simply by ‘toxic’ effects of epileptic seizures on the brain. Rather the finding strongly suggests that one or more genes, the function of which is disturbed by the structural variant, play(s) a role in the pathogenesis of both epilepsy and psychosis.

A second recent observation of potential interest concerns genes encoding ion channels. Ion channelopathies are known to underlie some epilepsies, so it is of great interest that variation within the gene CACNA1C (encoding a subunit of the L-type voltage-dependent calcium channel) is associated with schizophrenia as well as recurrent depression and bipolar disorder. ^{Reference Ferreira, O'Donovan, Meng, Jones, Ruderfer and Jones3,Reference Green, Grozeva, Jones, Jones, Kirov and Caesar4} This suggests that ion channel dysfunction may be also be involved in mood and psychotic illness. Again, this provides support for the possibility that some individuals might experience both psychosis and epilepsy at least in part because of an underlying vulnerability to both.

It is likely that as the understanding of brain function increases we will move closer to understanding the complexities, multiple associations and comorbidities that commonly occur in psychiatric illness. A sufficient number of adequately trained psychiatrists working within appropriate services will be vital for translating this knowledge into benefits for patients. ^{Reference Bullmore, Fletcher and Jones5}