BackgroundLithium remains a first-line treatment in bipolar disorder, but individual response is variable. Previous studies have suggested that lithium response is a heritable trait. However, no genetic markers have been reproducibly identified.MethodsHere we report the results of a genome-wide association study of lithium response in 2,563 patients collected by 22 participating sites from the International Consortium on Lithium Genetics (ConLiGen); the largest attempted so far. Data from over 6 million common single nucleotide polymorphisms (SNPs) were tested for association with categorical and continuous ratings of lithium response of known reliability.FindingsA single locus of four linked SNPs on chromosome 21 met genome-wide significance criteria for association with lithium response (rs79663003: p=1·37×10−8; rs78015114: p=1·31×10−8; rs74795342: p=3·31×10−9; rs75222709: p=3·50×10−9). In an independent, prospective study of 73 patients treated with lithium monotherapy for a period of up to two years, carriers of the response-associated alleles had a significantly lower rate of relapse than carriers of the alternate alleles (p=0·03, hazard ratio = 3·8).InterpretationThe response-associated region contains two genes coding for long non-coding RNAs (lncRNAs), AL157359.3 and AL157359.4. LncRNAs are increasingly appreciated as important regulators of gene expression, particularly in the CNS. Further studies are needed to establish the biological context of these findings and their potential clinical utility. Confirmed biomarkers of lithium response would constitute an important step forward in the clinical management of bipolar disorder.

ObjectiveThe assessment of response to lithium maintenance treatment in bipolar disorder (BD) is complicated by variable length of treatment, unpredictable clinical course, and often inconsistent compliance. Prospective and retrospective methods of assessment of lithium response have been proposed in the literature. In this study we report the key phenotypic measures of the “Retrospective Criteria of Long-Term Treatment Response in Research Subjects with Bipolar Disorder” scale currently used in the Consortium on Lithium Genetics (ConLiGen) study.Materials and MethodsTwenty-nine ConLiGen sites took part in a two-stage case-vignette rating procedure to examine inter-rater agreement [Kappa (κ)] and reliability [intra-class correlation coefficient (ICC)] of lithium response. Annotated first-round vignettes and rating guidelines were circulated to expert research clinicians for training purposes between the two stages. Further, we analyzed the distributional properties of the treatment response scores available for 1,308 patients using mixture modeling.ResultsSubstantial and moderate agreement was shown across sites in the first and second sets of vignettes (κ = 0.66 and κ = 0.54, respectively), without significant improvement from training. However, definition of response using the A score as a quantitative trait and selecting cases with B criteria of 4 or less showed an improvement between the two stages (ICC1 = 0.71 and ICC2 = 0.75, respectively). Mixture modeling of score distribution indicated three subpopulations (full responders, partial responders, non responders).ConclusionsWe identified two definitions of lithium response, one dichotomous and the other continuous, with moderate to substantial inter-rater agreement and reliability. Accurate phenotypic measurement of lithium response is crucial for the ongoing ConLiGen pharmacogenomic study.

No abstract

Findings could assist in the formulation of novel concepts of depressive disorders in young people and inform preventive strategies and predictive models for clinical practice.

No abstract

This study provides evidence for a negative association between high genetic loading for SCZ and poor response to lithium in patients with BPAD. These results suggest the potential for translational research aimed at personalized prescribing of lithium.

Lithium is a first-line medication for bipolar disorder (BD), but only one in three patients respond optimally to the drug. Since evidence shows a strong clinical and genetic overlap between depression and bipolar disorder, we investigated whether a polygenic susceptibility to major depression is associated with response to lithium treatment in patients with BD. Weighted polygenic scores (PGSs) were computed for major depression (MD) at different GWAS p value thresholds using genetic data obtained from 2586 bipolar patients who received lithium treatment and took part in the Consortium on Lithium Genetics (ConLi+Gen) study. Summary statistics from genome-wide association studies in MD (135,458 cases and 344,901 controls) from the Psychiatric Genomics Consortium (PGC) were used for PGS weighting. Response to lithium treatment was defined by continuous scores and categorical outcome (responders versus non-responders) using measurements on the Alda scale. Associations between PGSs of MD and lithium treatment response were assessed using a linear and binary logistic regression modeling for the continuous and categorical outcomes, respectively. The analysis was performed for the entire cohort, and for European and Asian sub-samples. The PGSs for MD were significantly associated with lithium treatment response in multi-ethnic, European or Asian populations, at various p value thresholds. Bipolar patients with a low polygenic load for MD were more likely to respond well to lithium, compared to those patients with high polygenic load [lowest vs highest PGS quartiles, multi-ethnic sample: OR = 1.54 (95% CI: 1.18-2.01) and European sample: OR = 1.75 (95% CI: 1.30-2.36)]. While our analysis in the Asian sample found equivalent effect size in the same direction: OR = 1.71 (95% CI: 0.61-4.90), this was not statistically significant. Using PGS decile comparison, we found a similar trend of association between a high genetic loading for MD and lower response to lithium. Our findings underscore the genetic contribution to lithium response in BD and support the emerging concept of a lithium-responsive biotype in BD.

We currently observe a disconcerting phenomenon in machine learning studies in psychiatry: While we would expect larger samples to yield better results due to the availability of more data, larger machine learning studies consistently show much weaker performance than the numerous small-scale studies. Here, we systematically investigated this effect focusing on one of the most heavily studied questions in the field, namely the classification of patients suffering from Major Depressive Disorder (MDD) and healthy controls based on neuroimaging data. Drawing upon structural MRI data from a balanced sample of N = 1868 MDD patients and healthy controls from our recent international Predictive Analytics Competition (PAC), we first trained and tested a classification model on the full dataset which yielded an accuracy of 61%. Next, we mimicked the process by which researchers would draw samples of various sizes (N = 4 to N = 150) from the population and showed a strong risk of misestimation. Specifically, for small sample sizes (N = 20), we observe accuracies of up to 95%. For medium sample sizes (N = 100) accuracies up to 75% were found. Importantly, further investigation showed that sufficiently large test sets effectively protect against performance misestimation whereas larger datasets per se do not. While these results question the validity of a substantial part of the current literature, we outline the relatively low-cost remedy of larger test sets, which is readily available in most cases.