Genome-wide study of major depression in 685,808 diverse individuals identifies 697 independent associations, infers causal neuronal subtypes and biological targets for novel pharmacotherapies.

In a genome-wide association study (GWAS) meta-analysis of 685,808 individuals with major depression (MD) and 4,364,225 controls from 29 countries and across diverse and admixed ancestries, we identify 697 independent associations at 636 loci, 293 of which are novel. Using fine-mapping and functional genomic tools, we find 308 high-confidence gene associations and enrichment of postsynaptic density and receptor clustering. Leveraging new single-cell gene expression data, we conducted a causal neural cell type enrichment analysis that implicates dysregulation of excitatory and inhibitory midbrain and forebrain neurons, peptidergic neurons, and medium spiny neurons in MD. Our findings are enriched for the targets of antidepressants and provide potential antidepressant repurposing opportunities (e.g., pregabalin and modafinil). Polygenic scores (PGS) trained using either European or multi-ancestry data significantly predicted MD status across all five diverse ancestries and explained a maximum of 5.8% of the variance in liability to MD in Europeans. These findings represent a major advance in our understanding of MD across global populations. MD GWAS reveals known and novel biological targets that may be used to target and develop pharmacotherapies addressing the considerable unmet need for effective treatment.

Blood-based epigenome-wide analyses of chronic low-grade inflammation across diverse population cohorts.

Chronic inflammation is a hallmark of age-related disease states. The effectiveness of inflammatory proteins including C-reactive protein (CRP) in assessing long-term inflammation is hindered by their phasic nature. DNA methylation (DNAm) signatures of CRP may act as more reliable markers of chronic inflammation. We show that inter-individual differences in DNAm capture 50% of the variance in circulating CRP (N = 17,936, Generation Scotland). We develop a series of DNAm predictors of CRP using state-of-the-art algorithms. An elastic-net-regression-based predictor outperformed competing methods and explained 18% of phenotypic variance in the Lothian Birth Cohort of 1936 (LBC1936) cohort, doubling that of existing DNAm predictors. DNAm predictors performed comparably in four additional test cohorts (Avon Longitudinal Study of Parents and Children, Health for Life in Singapore, Southall and Brent Revisited, and LBC1921), including for individuals of diverse genetic ancestry and different age groups. The best-performing predictor surpassed assay-measured CRP and a genetic score in its associations with 26 health outcomes. Our findings forge new avenues for assessing chronic low-grade inflammation in diverse populations.

A comprehensive hierarchical comparison of structural connectomes in Major Depressive Disorder cases v. controls in two large population samples.

BACKGROUND: The brain can be represented as a network, with nodes as brain regions and edges as region-to-region connections. Nodes with the most connections (hubs) are central to efficient brain function. Current findings on structural differences in Major Depressive Disorder (MDD) identified using network approaches remain inconsistent, potentially due to small sample sizes. It is still uncertain at what level of the connectome hierarchy differences may exist, and whether they are concentrated in hubs, disrupting fundamental brain connectivity. METHODS: We utilized two large cohorts, UK Biobank (UKB, N = 5104) and Generation Scotland (GS, N = 725), to investigate MDD case-control differences in brain network properties. Network analysis was done across four hierarchical levels: (1) global, (2) tier (nodes grouped into four tiers based on degree) and rich club (between-hub connections), (3) nodal, and (4) connection. RESULTS: In UKB, reductions in network efficiency were observed in MDD cases globally (d = -0.076, pFDR = 0.033), across all tiers (d = -0.069 to -0.079, pFDR = 0.020), and in hubs (d = -0.080 to -0.113, pFDR = 0.013-0.035). No differences in rich club organization and region-to-region connections were identified. The effect sizes and direction for these associations were generally consistent in GS, albeit not significant in our lower-N replication sample. CONCLUSION: Our results suggest that the brain's fundamental rich club structure is similar in MDD cases and controls, but subtle topological differences exist across the brain. Consistent with recent large-scale neuroimaging findings, our findings offer a connectomic perspective on a similar scale and support the idea that minimal differences exist between MDD cases and controls.

Distinguishing different psychiatric disorders using DDx-PRS.

Despite great progress on methods for case-control polygenic prediction (e.g. schizophrenia vs. control), there remains an unmet need for a method that genetically distinguishes clinically related disorders (e.g. schizophrenia (SCZ) vs. bipolar disorder (BIP) vs. depression (MDD) vs. control); such a method could have important clinical value, especially at disorder onset when differential diagnosis can be challenging. Here, we introduce a method, Differential Diagnosis-Polygenic Risk Score (DDx-PRS), that jointly estimates posterior probabilities of each possible diagnostic category (e.g. SCZ=50%, BIP=25%, MDD=15%, control=10%) by modeling variance/covariance structure across disorders, leveraging case-control polygenic risk scores (PRS) for each disorder (computed using existing methods) and prior clinical probabilities for each diagnostic category. DDx-PRS uses only summary-level training data and does not use tuning data, facilitating implementation in clinical settings. In simulations, DDx-PRS was well-calibrated (whereas a simpler approach that analyzes each disorder marginally was poorly calibrated), and effective in distinguishing each diagnostic category vs. the rest. We then applied DDx-PRS to Psychiatric Genomics Consortium SCZ/BIP/MDD/control data, including summary-level training data from 3 case-control GWAS ( N =41,917-173,140 cases; total N =1,048,683) and held-out test data from different cohorts with equal numbers of each diagnostic category (total N =11,460). DDx-PRS was well-calibrated and well-powered relative to these training sample sizes, attaining AUCs of 0.66 for SCZ vs. rest, 0.64 for BIP vs. rest, 0.59 for MDD vs. rest, and 0.68 for control vs. rest. DDx-PRS produced comparable results to methods that leverage tuning data, confirming that DDx-PRS is an effective method. True diagnosis probabilities in top deciles of predicted diagnosis probabilities were considerably larger than prior baseline probabilities, particularly in projections to larger training sample sizes, implying considerable potential for clinical utility under certain circumstances. In conclusion, DDx-PRS is an effective method for distinguishing clinically related disorders.

Cognitive ability in early life and risk of depression in adulthood: A systematic review and meta-analysis.

BACKGROUND: There is an established association between cognitive ability and risk of depression, though the direction of this association is unclear. Measuring cognitive ability in childhood, prior to the diagnosis of depression, could help to understand whether childhood cognitive ability is associated with a later diagnosis of depression. This systematic review and meta-analysis explored the association between childhood cognitive ability and risk of depression in adulthood. METHODS: We searched five databases to January 2024. We included studies that assessed cognitive ability in childhood (<18 years) and depression in adulthood. We excluded studies with very specific populations. We pooled each study's most-adjusted correlation coefficient in a random-effects meta-analysis. When studies reported a dichotomous outcome (depression/no depression), we converted the effect size to a correlation coefficient. Subgroup analyses were performed to explore sources of heterogeneity. RESULTS: 18 articles (19 cohorts) were included. There was no association between childhood cognitive ability and depression in adulthood (20 sample populations, N = 45,786, r = -0.04, 95 % CI = -0.09 to 0.01, p = 0.09). Neither age at cognitive assessment, length of follow-up, using a continuous/categorical measure of depression, or sex, significantly influenced the association. We rated most studies as having moderate risk of bias. LIMITATIONS: We limited the literature search to studies written in English. Existing studies were also heterogeneous, often adjusting for a variety of covariates. CONCLUSIONS: Our meta-analysis found no association between childhood cognitive ability and depression in adulthood. Future, longitudinal population-level studies should endeavour to control for potential mediators across the life-course (e.g., demographic and environmental factors).

Childhood cognitive ability and self-harm and suicide in later life.

Background: Self-harm and suicide remain prevalent in later life. For younger adults, higher early-life cognitive ability appears to predict lower self-harm and suicide risk. Comparatively little is known about these associations among middle-aged and older adults. Methods: This study examined the association between childhood (age 11) cognitive ability and self-harm and suicide risk among a Scotland-wide cohort (N = 53037), using hospital admission and mortality records to follow individuals from age 34 to 85. Multistate models examined the association between childhood cognitive ability and transitions between unaffected, self-harm, and then suicide or non-suicide death. Results: After adjusting for childhood and adulthood socioeconomic conditions, higher childhood cognitive ability was significantly associated with reduced risk of self-harm among both males (451 events; HR = 0.90, 95% CI [0.82, 0.99]) and females (516 events; HR = 0.89, 95% CI [0.81, 0.98]). Childhood cognitive ability was not significantly associated with suicide risk among those with (Male: 16 events, HR = 1.05, 95% CI [0.61, 1.80]; Female: 13 events, HR = 1.08, 95% CI [0.55, 2.15]) or without self-harm events (Male: 118 events, HR = 1.17, 95% CI [0.84, 1.63]; Female: 31 events, HR = 1.30, 95% CI [0.70, 2.41]). Limitations: The study only includes self-harm events that result in a hospital admission and does not account for self-harm prior to follow-up. Conclusions: This extends work on cognitive ability and mental health, demonstrating that these associations can span the life course and into middle and older age.

Epigenetic Contributions to Clinical Risk Prediction of Cardiovascular Disease.

BACKGROUND: Cardiovascular disease (CVD) is among the leading causes of death worldwide. The discovery of new omics biomarkers could help to improve risk stratification algorithms and expand our understanding of molecular pathways contributing to the disease. Here, ASSIGN-a cardiovascular risk prediction tool recommended for use in Scotland-was examined in tandem with epigenetic and proteomic features in risk prediction models in ≥12 657 participants from the Generation Scotland cohort. METHODS: Previously generated DNA methylation-derived epigenetic scores (EpiScores) for 109 protein levels were considered, in addition to both measured levels and an EpiScore for cTnI (cardiac troponin I). The associations between individual protein EpiScores and the CVD risk were examined using Cox regression (ncases≥1274; ncontrols≥11 383) and visualized in a tailored R application. Splitting the cohort into independent training (n=6880) and test (n=3659) subsets, a composite CVD EpiScore was then developed. RESULTS: Sixty-five protein EpiScores were associated with incident CVD independently of ASSIGN and the measured concentration of cTnI (P<0.05), over a follow-up of up to 16 years of electronic health record linkage. The most significant EpiScores were for proteins involved in metabolic, immune response, and tissue development/regeneration pathways. A composite CVD EpiScore (based on 45 protein EpiScores) was a significant predictor of CVD risk independent of ASSIGN and the concentration of cTnI (hazard ratio, 1.32; P=3.7×10-3; 0.3% increase in C-statistic). CONCLUSIONS: EpiScores for circulating protein levels are associated with CVD risk independent of traditional risk factors and may increase our understanding of the etiology of the disease.

Disrupted limbic-prefrontal effective connectivity in response to fearful faces in lifetime depression.

BACKGROUND: Multiple brain imaging studies of negative emotional bias in major depressive disorder (MDD) have used images of fearful facial expressions and focused on the amygdala and the prefrontal cortex. The results have, however, been inconsistent, potentially due to small sample sizes (typically N<50). It remains unclear if any alterations are a characteristic of current depression or of past experience of depression, and whether there are MDD-related changes in effective connectivity between the two brain regions. METHODS: Activations and effective connectivity between the amygdala and dorsolateral prefrontal cortex (DLPFC) in response to fearful face stimuli were studied in a large population-based sample from Generation Scotland. Participants either had no history of MDD (N=664 in activation analyses, N=474 in connectivity analyses) or had a diagnosis of MDD during their lifetime (LMDD, N=290 in activation analyses, N=214 in connectivity analyses). The within-scanner task involved implicit facial emotion processing of neutral and fearful faces. RESULTS: Compared to controls, LMDD was associated with increased activations in left amygdala (PFWE=0.031,kE=4) and left DLPFC (PFWE=0.002,kE=33), increased mean bilateral amygdala activation (ß=0.0715,P=0.0314), and increased inhibition from left amygdala to left DLPFC, all in response to fearful faces contrasted to baseline. Results did not appear to be attributable to depressive illness severity or antidepressant medication status at scan time. LIMITATIONS: Most studied participants had past rather than current depression, average severity of ongoing depression symptoms was low, and a substantial proportion of participants were receiving medication. The study was not longitudinal and the participants were only assessed a single time. CONCLUSIONS: LMDD is associated with hyperactivity of the amygdala and DLPFC, and with stronger amygdala to DLPFC inhibitory connectivity, all in response to fearful faces, unrelated to depression severity at scan time. These results help reduce inconsistency in past literature and suggest disruption of 'bottom-up' limbic-prefrontal effective connectivity in depression.

Distinct genomic signatures and modifiable risk factors underly the comorbidity between major depressive disorder and cardiovascular disease.

Major depressive disorder (MDD) and cardiovascular disease (CVD) are often comorbid, resulting in excess morbidity and mortality. Using genomic data, this study elucidates biological mechanisms, key risk factors, and causal pathways underlying their comorbidity. We show that CVDs share a large proportion of their genetic risk factors with MDD. Multivariate genome-wide association analysis of the shared genetic liability between MDD and atherosclerotic CVD (ASCVD) revealed seven novel loci and distinct patterns of tissue and brain cell-type enrichments, suggesting a role for the thalamus. Part of the genetic overlap was explained by shared inflammatory, metabolic, and psychosocial/lifestyle risk factors. Finally, we found support for causal effects of genetic liability to MDD on CVD risk, but not from most CVDs to MDD, and demonstrated that the causal effects were partly explained by metabolic and psychosocial/lifestyle factors. The distinct signature of MDD-ASCVD comorbidity aligns with the idea of an immunometabolic sub-type of MDD more strongly associated with CVD than overall MDD. In summary, we identify plausible biological mechanisms underlying MDD-CVD comorbidity, as well as key modifiable risk factors for prevention of CVD in individuals with MDD.

Comprehensive assessment of sleep duration, insomnia, and brain structure within the UK Biobank cohort.

STUDY OBJECTIVES: To assess for associations between sleeping more than or less than recommended by the National Sleep Foundation (NSF), and self-reported insomnia, with brain structure. METHODS: Data from the UK Biobank cohort were analyzed (N between 9K and 32K, dependent on availability, aged 44 to 82 years). Sleep measures included self-reported adherence to NSF guidelines on sleep duration (sleeping between 7 and 9 hours per night), and self-reported difficulty falling or staying asleep (insomnia). Brain structural measures included global and regional cortical or subcortical morphometry (thickness, surface area, volume), global and tract-related white matter microstructure, brain age gap (difference between chronological age and age estimated from brain scan), and total volume of white matter lesions. RESULTS: Longer-than-recommended sleep duration was associated with lower overall grey and white matter volumes, lower global and regional cortical thickness and volume measures, higher brain age gap, higher volume of white matter lesions, higher mean diffusivity globally and in thalamic and association fibers, and lower volume of the hippocampus. Shorter-than-recommended sleep duration was related to higher global and cerebellar white matter volumes, lower global and regional cortical surface areas, and lower fractional anisotropy in projection fibers. Self-reported insomnia was associated with higher global gray and white matter volumes, and with higher volumes of the amygdala, hippocampus, and putamen. CONCLUSIONS: Sleeping longer than recommended by the NSF is associated with a wide range of differences in brain structure, potentially indicative of poorer brain health. Sleeping less than recommended is distinctly associated with lower cortical surface areas. Future studies should assess the potential mechanisms of these differences and investigate long sleep duration as a putative marker of brain health.

Childhood intelligence and risk of depression in later-life: A longitudinal data-linkage study.

Background: Lower childhood intelligence test scores are reported in some studies to be associated with higher risk of depression in adulthood. The reasons for the association are unclear. This longitudinal data-linkage study explored the relationship between childhood intelligence (at age ∼11) and risk of depression in later-life (up to age ∼85), and whether childhood family structure and adulthood socio-economic and geographical factors accounted for some of this association. Methods: Intelligence test scores collected in the Scottish Mental Survey 1947 were linked to electronic health records (hospital admissions and prescribing data) between 1980 and 2020 (n = 53,037), to identify diagnoses of depression. Mixed-effect Cox regression models were used to explore the relationship between childhood intelligence test scores and risk of depression in later-life. Analyses were also adjusted for childhood family structure (size of family) and adulthood socio-economic and geographical factors (Carstairs index, urban/rural). Results: Twenty-seven percent of participants were diagnosed with depression during follow-up (n = 14,063/53,037). Greater childhood intelligence test scores were associated with a reduced risk of depression in an unadjusted analysis (HR = 0.95, 95% CI = 0.93 to 0.97, P < 0.001), and after adjustment for factors experienced in childhood and adulthood (HR = 0.95, 95% CI = 0.91 to 1.00, P = 0.032). When identifying depression using only hospital admissions data, greater childhood intelligence test scores were associated with a reduced risk of depression following unadjusted analysis (HR = 0.86, 95% CI = 0.82 to 0.90, P < 0.001), and after adjusting for risk factors in childhood and adulthood (HR = 0.94, 95% CI = 0.89 to 0.99, P = 0.026). There was no association between childhood cognitive test scores and depression when identifying cases of depression using only prescribed drugs data. Conclusions: This study provides additional evidence suggesting that higher childhood intelligence predicts reduced risk of later-life depression only when depression is assessed based on hospital admission records. Childhood family structure and adulthood socio-economic and geographical factors did not seem to be substantial confounders.

Integration of datasets for individual prediction of DNA methylation-based biomarkers.

BACKGROUND: Epigenetic scores (EpiScores) can provide biomarkers of lifestyle and disease risk. Projecting new datasets onto a reference panel is challenging due to separation of technical and biological variation with array data. Normalisation can standardise data distributions but may also remove population-level biological variation. RESULTS: We compare two birth cohorts (Lothian Birth Cohorts of 1921 and 1936 - nLBC1921 = 387 and nLBC1936 = 498) with blood-based DNA methylation assessed at the same chronological age (79 years) and processed in the same lab but in different years and experimental batches. We examine the effect of 16 normalisation methods on a novel BMI EpiScore (trained in an external cohort, n = 18,413), and Horvath's pan-tissue DNA methylation age, when the cohorts are normalised separately and together. The BMI EpiScore explains a maximum variance of R2=24.5% in BMI in LBC1936 (SWAN normalisation). Although there are cross-cohort R2 differences, the normalisation method makes a minimal difference to within-cohort estimates. Conversely, a range of absolute differences are seen for individual-level EpiScore estimates for BMI and age when cohorts are normalised separately versus together. While within-array methods result in identical EpiScores whether a cohort is normalised on its own or together with the second dataset, a range of differences is observed for between-array methods. CONCLUSIONS: Normalisation methods returning similar EpiScores, whether cohorts are analysed separately or together, will minimise technical variation when projecting new data onto a reference panel. These methods are important for cases where raw data is unavailable and joint normalisation of cohorts is computationally expensive.

Maternal mental health research in Malawi: Community and healthcare provider perspectives on acceptability and ethicality.

Maternal mental health (MMH) is recognised as globally significant. The prevalence of depression and factors associated with its onset among perinatal women in Malawi has been previously reported, and the need for further research in this domain is underscored. Yet, there is little published scholarship regarding the acceptability and ethicality of MMH research to women and community representatives. The study reported here sought to address this in Malawi by engaging with communities and healthcare providers in the districts where MMH research was being planned. Qualitative data was collected in Lilongwe and Karonga districts through 20 focus group discussions and 40 in-depth interviews with community representatives and healthcare providers from January through April 2021. All focus groups and interviews were audio recorded, transcribed verbatim (in local languages Chichewa and Tumbuka), translated into English, and examined through thematic content analysis. Participants' accounts suggest that biopsychosocial MMH research could be broadly acceptable within the communities sampled, with acceptability framed in part through prior encounters with biomedical and public health research and care in these regions, alongside broader understandings of the import of MMH. Willingness and consent to participate do not depend on specifically biomedical understandings of MMH, but rather on familiarity with individuals regarded as living with mental ill-health. However, the data further suggest some 'therapeutic misconceptions' about MMH research, with implications for how investigations in this area are presented by researchers when recruiting and working with participants. Further studies are needed to explore whether accounts of the acceptability and ethicality of MMH research shift and change during and following research encounters. Such studies will enhance the production of granular recommendations for further augmenting the ethicality of biomedical and public health research and researchers' responsibilities to participants and communities.

Pathway-Based Polygenic Risk Scores for Schizophrenia and Associations With Reported Psychotic-like Experiences and Neuroimaging Phenotypes in the UK Biobank.

Background: Schizophrenia is a heritable psychiatric disorder with a polygenic architecture. Genome-wide association studies have reported that an increasing number of risk-associated variants and polygenic risk scores (PRSs) explain 17% of the variance in the disorder. Substantial heterogeneity exists in the effect of these variants, and aggregating them based on biologically relevant functions may provide mechanistic insight into the disorder. Methods: Using the largest schizophrenia genome-wide association study conducted to date, we associated PRSs based on 5 gene sets previously found to contribute to schizophrenia pathophysiology-postsynaptic density of excitatory synapses, postsynaptic membrane, dendritic spine, axon, and histone H3-K4 methylation-along with respective whole-genome PRSs, with neuroimaging (n > 29,000) and reported psychotic-like experiences (n > 119,000) variables in healthy UK Biobank subjects. Results: Several variables were significantly associated with the axon gene-set (psychotic-like communications, parahippocampal gyrus volume, fractional anisotropy thalamic radiations, and fractional anisotropy posterior thalamic radiations (ß range -0.016 to 0.0916, false discovery rate-corrected p [pFDR] ≤ .05), postsynaptic density gene-set (psychotic-like experiences distress, global surface area, and cingulate lobe surface area [ß range -0.014 to 0.0588, pFDR ≤ .05]), and histone gene set (entorhinal surface area: ß = -0.016, pFDR = .035). From these, whole-genome PRSs were significantly associated with psychotic-like communications (ß = 0.2218, pFDR = 1.34 × 10-7), distress (ß = 0.1943, pFDR = 7.28 × 10-16), and fractional anisotropy thalamic radiations (ß = -0.0143, pFDR = .036). Permutation analysis revealed that these associations were not due to chance. Conclusions: Our results indicate that genetic variation in 3 gene sets relevant to schizophrenia may confer risk for the disorder through effects on previously implicated neuroimaging variables. Because associations were stronger overall for whole-genome PRSs, findings here highlight that selection of biologically relevant variants is not yet sufficient to address the heterogeneity of the disorder.

Psychosis Endophenotypes: A Gene-Set-Specific Polygenic Risk Score Analysis.

BACKGROUND AND HYPOTHESIS: Endophenotypes can help to bridge the gap between psychosis and its genetic predispositions, but their underlying mechanisms remain largely unknown. This study aims to identify biological mechanisms that are relevant to the endophenotypes for psychosis, by partitioning polygenic risk scores into specific gene sets and testing their associations with endophenotypes. STUDY DESIGN: We computed polygenic risk scores for schizophrenia and bipolar disorder restricted to brain-related gene sets retrieved from public databases and previous publications. Three hundred and seventy-eight gene-set-specific polygenic risk scores were generated for 4506 participants. Seven endophenotypes were also measured in the sample. Linear mixed-effects models were fitted to test associations between each endophenotype and each gene-set-specific polygenic risk score. STUDY RESULTS: After correction for multiple testing, we found that a reduced P300 amplitude was associated with a higher schizophrenia polygenic risk score of the forebrain regionalization gene set (mean difference per SD increase in the polygenic risk score: -1.15 µV; 95% CI: -1.70 to -0.59 µV; P = 6 × 10-5). The schizophrenia polygenic risk score of forebrain regionalization also explained more variance of the P300 amplitude (R2 = 0.032) than other polygenic risk scores, including the genome-wide polygenic risk scores. CONCLUSIONS: Our finding on reduced P300 amplitudes suggests that certain genetic variants alter early brain development thereby increasing schizophrenia risk years later. Gene-set-specific polygenic risk scores are a useful tool to elucidate biological mechanisms of psychosis and endophenotypes, offering leads for experimental validation in cellular and animal models.

Blood-based epigenome-wide analyses of 19 common disease states: A longitudinal, population-based linked cohort study of 18,413 Scottish individuals.

BACKGROUND: DNA methylation is a dynamic epigenetic mechanism that occurs at cytosine-phosphate-guanine dinucleotide (CpG) sites. Epigenome-wide association studies (EWAS) investigate the strength of association between methylation at individual CpG sites and health outcomes. Although blood methylation may act as a peripheral marker of common disease states, previous EWAS have typically focused only on individual conditions and have had limited power to discover disease-associated loci. This study examined the association of blood DNA methylation with the prevalence of 14 disease states and the incidence of 19 disease states in a single population of over 18,000 Scottish individuals. METHODS AND FINDINGS: DNA methylation was assayed at 752,722 CpG sites in whole-blood samples from 18,413 volunteers in the family-structured, population-based cohort study Generation Scotland (age range 18 to 99 years). EWAS tested for cross-sectional associations between baseline CpG methylation and 14 prevalent disease states, and for longitudinal associations between baseline CpG methylation and 19 incident disease states. Prevalent cases were self-reported on health questionnaires at the baseline. Incident cases were identified using linkage to Scottish primary (Read 2) and secondary (ICD-10) care records, and the censoring date was set to October 2020. The mean time-to-diagnosis ranged from 5.0 years (for chronic pain) to 11.7 years (for Coronavirus Disease 2019 (COVID-19) hospitalisation). The 19 disease states considered in this study were selected if they were present on the World Health Organisation's 10 leading causes of death and disease burden or included in baseline self-report questionnaires. EWAS models were adjusted for age at methylation typing, sex, estimated white blood cell composition, population structure, and 5 common lifestyle risk factors. A structured literature review was also conducted to identify existing EWAS for all 19 disease states tested. The MEDLINE, Embase, Web of Science, and preprint servers were searched to retrieve relevant articles indexed as of March 27, 2023. Fifty-four of approximately 2,000 indexed articles met our inclusion criteria: assayed blood-based DNA methylation, had >20 individuals in each comparison group, and examined one of the 19 conditions considered. First, we assessed whether the associations identified in our study were reported in previous studies. We identified 69 associations between CpGs and the prevalence of 4 conditions, of which 58 were newly described. The conditions were breast cancer, chronic kidney disease, ischemic heart disease, and type 2 diabetes mellitus. We also uncovered 64 CpGs that associated with the incidence of 2 disease states (COPD and type 2 diabetes), of which 56 were not reported in the surveyed literature. Second, we assessed replication across existing studies, which was defined as the reporting of at least 1 common site in >2 studies that examined the same condition. Only 6/19 disease states had evidence of such replication. The limitations of this study include the nonconsideration of medication data and a potential lack of generalizability to individuals that are not of Scottish and European ancestry. CONCLUSIONS: We discovered over 100 associations between blood methylation sites and common disease states, independently of major confounding risk factors, and a need for greater standardisation among EWAS on human disease.

Meta-analysis of CYP2C19 and CYP2D6 metabolic activity on antidepressant response from 13 clinical studies using genotype imputation.

Cytochrome P450 enzymes including CYP2C19 and CYP2D6 are important for antidepressant metabolism and polymorphisms of these genes have been determined to predict metabolite levels. Nonetheless, more evidence is needed to understand the impact of genetic variations on antidepressant response. In this study, individual clinical and genetic data from 13 studies of European and East Asian ancestry populations were collected. The antidepressant response was clinically assessed as remission and percentage improvement. Imputed genotype was used to translate genetic polymorphisms to metabolic phenotypes (poor, intermediate, normal, and rapid+ultrarapid) of CYP2C19 and CYP2D6. The association of CYP2C19 and CYP2D6 metabolic phenotypes with treatment response was examined using normal metabolizers as the reference. Among 5843 depression patients, a higher remission rate was found in CYP2C19 poor metabolizers compared to normal metabolizers at nominal significance but did not survive after multiple testing correction (OR=1.46, 95% CI [1.03, 2.06], p=0.033, heterogeneity I2=0%, subgroup difference p=0.72). No metabolic phenotype was associated with percentage improvement from baseline. After stratifying by antidepressants primarily metabolized by CYP2C19 and CYP2D6, no association was found between metabolic phenotypes and antidepressant response. Metabolic phenotypes showed differences in frequency, but not effect, between European- and East Asian-ancestry studies. In conclusion, metabolic phenotypes imputed from genetic variants using genotype were not associated with antidepressant response. CYP2C19 poor metabolizers could potentially contribute to antidepressant efficacy with more evidence needed. CYP2D6 structural variants cannot be imputed from genotype data, limiting inference of pharmacogenetic effects. Sequencing and targeted pharmacogenetic testing, alongside information on side effects, antidepressant dosage, depression measures, and diverse ancestry studies, would more fully capture the influence of metabolic phenotypes.