Abstract and Introduction
Abstract
Background and Hypothesis: Schizophrenia has been robustly associated with multiple genetic and environmental risk factors. Childhood adversity is one of the most widely replicated environmental risk factors for schizophrenia, but it is unclear if schizophrenia genetic risk alleles contribute to this association.
Study Design: In this systematic review and meta-analysis, we assessed the evidence for gene-environment correlation (genes influence likelihood of environmental exposure) between schizophrenia polygenic risk score (PRS) and reported childhood adversity. We also assessed the evidence for a gene-environment interaction (genes influence sensitivity to environmental exposure) in relation to the outcome of schizophrenia and/or psychosis. This study was registered on PROSPERO (CRD42020182812). Following PRISMA guidelines, a search for relevant literature was conducted using Cochrane, MEDLINE, PsycINFO, Web of Science, and Scopus databases until February 2022. All studies that examined the association between schizophrenia PRS and childhood adversity were included.
Study Results: Seventeen of 650 identified studies met the inclusion criteria and were assessed against the Newcastle-Ottawa Scale for quality. The meta-analysis found evidence for gene-environment correlation between schizophrenia PRS and childhood adversity (r = .02; 95% CI = 0.01, 0.03; P = .001), but the effect was small and therefore likely to explain only a small proportion of the association between childhood adversity and psychosis. The 4 studies that investigated a gene-environment interaction between schizophrenia PRS and childhood adversity in increasing risk of psychosis reported inconsistent results.
Conclusions: These findings suggest that a gene-environment correlation could explain a small proportion of the relationship between reported childhood adversity and psychosis.
Introduction
Schizophrenia is a heritable condition with estimates of around 80% in twin and family studies.[1] There have been recent major advances in our understanding of psychiatric genetic risk, driven by international collaborations, falling costs of genotyping technologies, and methodological advances.[2] Genome-wide association studies (GWAS) have established that genetic risk for schizophrenia is polygenic, with a substantial contribution made by a large number of common genetic variants that individually increase risk to a small extent.[3,4] Polygenic risk scores (PRS) are derived by calculating a weighted sum of genetic risk alleles present in an individual's genome.[5]PRS is a measure of an individual's inherited liability to developing a disorder or trait and, in schizophrenia, has accounted for increasing proportions of heritability as the power in GWAS datasets has increased, currently explaining up to 8% of the variance in liability.[6]
Environmental risk factors have also been shown to contribute to the likelihood of developing schizophrenia; evidence exists for illicit drug use (in particular cannabis),[7–10] obstetric complications,[11–13] season of birth,[14] urbanicity,[15–17] and migration.[18,19] Childhood adversity is one of the most widely replicated environmental risk factors for psychosis,[20] but the association has also been extended to schizophrenia.[21] Childhood adversity is common worldwide and could involve anything that presents a serious threat to a child's physical or psychological well-being such as trauma, abuse, neglect, parental death or separation, and bullying. A large meta-analysis across prospective cohorts, case-control, and cross-sectional design studies found that patients with psychosis were 2.78 (95% CI = 2.34, 3.31) times more likely to report being exposed to childhood adversity or trauma than controls.[22] Furthermore, this association has demonstrated a "dose-effect" whereby as the severity of the childhood trauma increased, so did the likelihood of developing psychosis.[21]
Although evidence indicates that schizophrenia arises from both genetic liability and environmental exposures, how these risk factors combine to ultimately lead to disorder remains unclear. There is ongoing debate about whether genetic and environmental effects act independently, are associated via gene-environment correlation, and/or via gene-environment interaction. Table 1 provides a summary of these terms. A gene-environment correlation refers to how an individual's genes can influence the environment they are exposed to. This can be either passive, evocative, or active. A passive gene-environment correlation refers to the association between the child's genetics and the environment in which they are raised, whereby parents produce a home environment that is influenced by their genotypes.[23] An evocative gene-environment correlation describes the association between a child's genetically influenced behavior and the reaction from others in their environment to that behavior.[24] Lastly, an active gene-environment correlation indicates an association between an individual's genetically influenced traits and the selection of specific environments (such as a shy child seeking a different environment to an outgoing child). Twin and adoption studies have highlighted the importance of gene-environment correlations and interactions in human genetic research.[25] The presence of a gene-environment correlation suggests that genetics may be a confounder that explains, at least in part, why an environmental risk factor is associated with a disorder or outcome. A gene-environment interaction alternatively suggests that an individual's genetics can control (moderate) how sensitive they are to environmental exposures and whether those exposures will increase the risk of an outcome.[26,27]
The first studies investigating the relationship between genetic risk and childhood adversities used twin and family designs and found significant within-pair differences, indicating that at least part of the association between childhood trauma and psychosis is likely to be causal and not driven by gene-environment correlation.[28–30] A recent qualitative review also found that childhood adversity was associated with psychosis largely independent of genetic liability.[31] However, this review only included studies with individuals experiencing psychosis and where the genetic factor was analyzed as a mediator, thus limiting the studies that were included.
The aim of the current study was to conduct a systematic review and meta-analysis of the existing literature from both population and clinical samples investigating the gene-environment correlation between schizophrenia polygenic risk and reported childhood adversity. To our knowledge, this is the first quantitative review of this evidence. Secondly, we aimed to summarize the evidence regarding gene-environment interaction between schizophrenia polygenic risk and childhood adversity for the outcome of psychosis or schizophrenia case/control status. Better understanding the relationship between schizophrenia risk alleles and childhood adversity may enhance our knowledge of the mechanisms underpinning the development of psychosis and schizophrenia.
Schizophr Bull. 2022;48(5):967-980. © 2022 Oxford University Press
