There are some limitations to our
There are some limitations to our study. First, the small sample size, even if comparable or higher than most studies, did not offer the possibility to consider other variables, such as frequency of use or age of onset. Second, since most subjects did not report current use, we could not analyze separately current and past users. Third, being a cross-sectional study investigating retrospective information, we cannot exclude a potential for recall bias. Moreover, the instrument adopted for evaluating the use of cannabis is based on subjective answers, whose accuracy can be flawed by imperfect recall and social desiderability bias. However, given the latter one, it Nitrotetrazolium Blue chloride is unlikely that the use of cannabis has been overestimated and has facilitated the results obtained. In this respect, it is to point out that up to date no instrument captures all key aspects of exposure to cannabis use, pointing out the urge of developing validated and standardized tools in order to perform more accurate measurements of the history of use (Lorenzetti et al., 2016). For future considerations, multimodal approaches should be encouraged, such as functional magnetic resonance imaging studies to explore cerebral activation linked to performance in patients with and without history of cannabis abuse while taking into account the genetic variation. Future studies should also take into consideration other candidate genes, such as those involved in glutamate regulation, since cannabis use is associated with reduced prefrontal glutamate levels and glutamatergic abnormalities have been linked to both schizophrenia and cognitive deficits (Rigucci et al., 2017). Finally, considering that polysubstance use is common and equally prevalent in patients with both affective and non-affective psychosis, and the great majority of patients exhibit neurocognitive deficits (Helle et al., 2017), it would be of interest to include patients with other psychiatric diagnoses. In conclusion, this is the first study to evaluate the interaction between cannabis use and the COMT polymorphism on cognitive performance in a sample of patients with schizophrenia. Results showed a significant interaction effect for core cognitive domains, i.e. verbal fluency and processing speed, with worse performance among Val/Val homozygous patients with a history of cannabis use. These results are in line with available data on healthy subjects that highlight a more detrimental effect of cannabis among subjects carrying the Val/Val genotype. Our data also support, in a naturalistic sample, the only study in patients, showing that acute THC administration impairs memory and attentions more severely in Val/Val homozygous than in Met carriers (Henquet et al., 2006). Although a more comprehensive understanding of the mechanisms underlying the interaction between COMT and cannabis is needed, it is known that THC-induced decrease in neuronal firing modulates the memory impairment associated with cannabis use, and the same mechanism may well be implicated when it comes to information processing speed (Pavisian et al., 2015). It could further be hypothesized that both cannabis and COMT polymorphism act through the same neurotransmitter systems and/or have common final effects from different pathways. The implications of these findings encompass both the clinical and research field. Indeed, from a scientific point of view, our results may lead to the identification of novel biological pathways in which cannabis use and COMT are converging, thus helping to further disentangle the mechanisms underlying cognitive deficits in schizophrenia and unravel novel treatment targets. The findings also have clinical application potentials as they underlie the risk associated to cannabis especially in “genetically at risk” subjects, stressing the need for prevention programs. Finally, the results strongly suggest that cannabis use should be routinely assessed in patients and taken into consideration as an important factor to design personalized interventions.