Biernacka, Joanna M.

ABSTRACT Objectives To investigate pharmacological treatment patterns in individuals with bipolar disorder (BD) with and without comorbid substance use disorder (SUD) and anxiety disorder (AX), we leveraged the Global Bipolar Cohort to analyze cross‐regional practices across North America, Europe, and the Pacific. Methods Fourteen cohorts contributed aggregate data on pharmacotherapy, demographics, diagnostic subtypes, and comorbidities. Proportional meta‐analyses using generalized linear mixed models were conducted to examine prescription trends and identify clinical differences. Results The sample ( N = 11,521) was 60% female and 84% Caucasian. Participants were categorized into four mutually exclusive groups based on comorbidity status: those with comorbid AX only, comorbid SUD only, both AX and SUD, or neither. The AX+SUD subgroup showed higher rates of attention‐deficit/hyperactivity disorder (ADHD), post‐traumatic stress disorder (PTSD), rapid cycling, obesity, and unemployment, reflecting a more severe clinical profile. Regional variations were notable: North American cohorts reported higher prevalence of AX and SUD than European and Pacific cohorts. Antidepressants use for AX were more common in Europe and the Pacific, while North American prescribing patterns were more variable. Benzodiazepine use was high among individuals with SUD across all regions. Lithium and first‐generation antipsychotic prescriptions varied, with higher rates observed in Europe. Conclusions Findings underscore the heterogeneity of BD and the influence of comorbid AX and SUD on illness burden and treatment. Regional prescribing variations underscore the need for context‐specific guidelines. Gaps in data on medication‐assisted treatment for SUD point to areas for future research. These insights can support more individualized and effective care for complex BD presentations.

ABSTRACT Background This study aimed to explore the effect of family history (FH) and polygenic risk scores (PRS) for multiple psychiatric disorders on lithium response in bipolar disorder (BD). Method The sample included patients from the Mayo Clinic BD Biobank. FH in first‐degree relatives was obtained via the patient questionnaire. Linear regression models tested the association of FH and PRS for BD, major depressive disorder (MDD), schizophrenia (SCZ), alcohol use disorder (AUD), anxiety disorder (ANX), and attention‐deficit/hyperactivity disorder (ADHD) with lithium response measured by the Alda A score adjusted for the Alda B sum score. Post hoc models additionally adjusted for clinical covariates (age at onset, rapid cycling, and BD subtype). Results 716 patients had Alda scores for lithium response (61.9% females, mean age = 43.4 years). FH‐BD ( β = −0.57, p = 0.015), AUD ( β = −0.44, p = 0.045), and ANX ( β = −0.61, p = 0.050) were associated with worse lithium response. After adjusting for clinical covariates, associations for FH‐BD and FH‐ANX were no longer significant. PRS‐BD was associated with better response ( β = 0.22, p = 0.041), while PRS‐MDD ( β = −0.28, p = 0.01) and PRS‐ADHD ( β = −0.24, p = 0.028) were associated with worse response, with effects largely unchanged after adjusting for clinical covariates. In the combined model considering all FHs simultaneously, only FH‐BD remained significantly associated with worse lithium response ( β = −0.71, p = 0.039). In the combined model for PRS, only PRS‐BD ( β = 0.31, p = 0.008; better response) and PRS‐MDD ( β = −0.29, p = 0.020; worse response) showed significant associations. Conclusions In this exploratory study, FH‐BD, AUD, and ANX were associated with poor response, though only FH‐AUD remained significant after adjusting for clinical factors. PRS‐BD predicted better response, while PRS‐MDD was associated with worse response. These findings suggest independent and potentially complementary roles of FH and PRS. Integrating both familial and genetic risk measures may improve personalized treatment strategies and outcomes for BD patients.

Abstract Antidepressants are widely prescribed for major depressive disorder, yet only one-third of patients achieve remission after initial treatment. Previous genome-wide association studies (GWAS) of clinically assessed antidepressant response combined multiple antidepressant classes, potentially obscuring class-specific effects. This study focused on selective serotonin reuptake inhibitors (SSRIs), often first-line due to better tolerability. Data from 15 cohorts across four ancestries were integrated: European (N = 3887; 11 studies), East Asian (N = 1068; 4), African (N = 277; 1), and Admixed American (N = 250; 1). GWAS of non-remission and percentage improvement were conducted within cohorts, followed by ancestry-specific meta-analyses and trans-ancestry meta-regression. Single nucleotide polymorphism (SNP)-based heritability was estimated in European samples. Polygenic scores were used for leave-one-out prediction and to assess shared genetic architecture with psychiatric traits. Gene-level and gene-set enrichment analyses were also performed. No genome-wide significant variants were identified for either outcome in any ancestry-specific or trans-ancestry analyses. However, trans-ancestry meta-regression yielded eight independent loci with suggestive associations (p < 1 × 10 −5 ) for non-remission and 17 for percentage improvement. Gene-set analyses revealed nominal enrichment of the serotonergic synapse pathway for non-remission. SNP-based heritability estimates were not significantly different from zero for either outcome. Better SSRI response was nominally associated with lower genetic predisposition to major depressive disorder, post-traumatic stress disorder, and schizophrenia. This study represents the largest trans-ancestry GWAS of SSRI response, highlighting emerging biological signals. Limited power emphasises the need for larger and ancestrally diverse cohorts to better characterise the genetic architecture of antidepressant response.

Low self-esteem (LoST) is a latent yet critical psychosocial risk factor that predisposes individuals to depressive disorders. Although structured tools exist to assess self-esteem, their limited clinical adoption suggests that relevant indicators of LoST remain buried within unstructured clinical narratives. The scarcity of annotated clinical notes impedes the development of natural language processing (NLP) models for its detection. Manual chart reviews are labor-intensive and large language model (LLM)-driven (weak) labeling raises privacy concerns. Past studies demonstrate that NLP models trained on LLM-generated synthetic clinical notes achieve performance comparable to, and sometimes better than those trained on real notes. This highlights synthetic data’s utility for augmenting scarce clinical corpora while reducing privacy concerns. Prior efforts have leveraged social media data, such as Reddit, to identify linguistic markers of low self-esteem; however, the linguistic and contextual divergence between social media and clinical text limits the generalizability of these models. To address this gap, we present a novel framework that generates context-enhanced synthetic clinical notes from social media narratives and evaluates the utility of small language models for identifying expressions of low self-esteem. Our approach includes a mixed-method evaluation framework: (i) structure analysis, (ii) readability analysis, (iii) linguistic diversity, and (iv) contextual fidelity of LoST cues in source Reddit posts and synthetic notes. This work offers a scalable, privacy-preserving solution for synthetic data generation for early detection of psychosocial risks such as LoST and demonstrates a pathway for translating mental health signals in clinical notes into clinically actionable insights, thereby identifying patients at risk.

Abstract Background Anxiety disorders (ANX) affect 30–60% of individuals with bipolar disorder (BD), yet limited research has systematically examined clinical characteristics and treatment patterns in this comorbid population. This study investigated demographic, clinical, and pharmacotherapeutic differences between individuals with BD with and without comorbid ANX. Methods Cross-sectional data from 2,225 adults with BD enrolled in the Mayo Clinic Bipolar Disorder Biobank were analyzed. Participants were assessed for comorbid ANX, demographics, clinical characteristics, medication use, and treatment response using the Alda-A scale. Results Overall, 61% (n = 1,366) had comorbid ANX. Individuals with BD + ANX were younger (40.4 vs. 43.6 years, p < 0.001), more likely female (66.6% vs. 54.8%, p < 0.001), and exhibited higher rates of rapid cycling (64.2% vs. 45.2%, p < 0.001), suicide attempts (40.4% vs. 24.8%, p < 0.001), substance use disorders (63.5% vs. 54.8%, p < 0.001), and somatic comorbidities (MCIRS: 6.68 vs. 5.42, p < 0.001). Pharmacotherapeutically, BD + ANX individuals were less likely to receive lithium (37.1% vs. 47.8%, p = 0.005) and valproic acid (21.7% vs. 29.6%, p = 0.047), but more likely to receive antidepressants (53.8% vs. 39.5%, p < 0.001), benzodiazepines (39.9% vs. 26.6%, p < 0.001), and gabapentinoids (8.5% vs. 4.5%, p < 0.001). Notably, 17.3% of BD + ANX individuals received antidepressants without mood stabilizer coverage. Treatment response (Alda-A) scores were significantly lower in BD + ANX for lithium (4.91 vs. 6.05, p < 0.001), mood-stabilizing anticonvulsants (5.09 vs. 6.22, p < 0.001), and second-generation antipsychotics (4.67 vs. 5.73, p < 0.001). Similar patterns were observed in both BD-I and BD-II subtypes. Conclusions Individuals with BD + ANX represent a more severely affected subgroup with distinct prescribing patterns favoring antidepressants over mood stabilizers and attenuated mood stabilizers response. These findings highlight the need for anxiety-informed treatment algorithms recognizing anxiety comorbidity as a negative prognostic factor.

ABSTRACT Background Mood episodes in bipolar disorder (BD) follow a seasonal pattern in approximately 25% of patients. We aimed to examine the associations of specific seasonality patterns in BD with various clinical phenotypes and treatment‐related outcomes. Methods Patients from the Mayo Clinic Bipolar Disorder Biobank with available seasonality data were included in the study. Among those reporting any seasonality, participants were further categorized into four seasonality groups: fall/winter depression, spring/summer (hypo)mania, biphasic (fall/winter depression with spring/summer (hypo)mania), and equinoctial seasonality (any mood episode occurring in spring or fall). Regression models were used, treating seasonality patterns as the predictor of clinical phenotypes (BD subtype, early onset, history of psychosis, suicide attempt, rapid cycling), lifetime medication exposure (number of unique antidepressants, antipsychotics, and any psychotropics) and treatment response to mood stabilizers (Alda A score). Results Among 1702 patients with BD (61.5% female, mean age = 41.43 ± 14.73 years), 44.6% reported seasonal mood episodes. Individuals with any seasonality were more likely to have BD‐I (OR = 1.29, p = 0.024), early onset (OR = 1.29, p = 0.026), exposure to higher lifetime number of antipsychotics (IRR = 1.20, p < 0.001) and any psychotropics (IRR = 1.13, p < 0.001). Fall/winter depression was associated with a higher lifetime number of antidepressants (IRR = 1.18, p = 0.021), any psychotropics (IRR = 1.14, p = 0.013), worse response to lithium ( β = −0.82, p = 0.029) and all mood stabilizers ( β = −0.69, p = 0.014). Spring/summer (hypo)mania showed negative associations with rapid cycling (OR = 0.36, p < 0.001) and lifetime number of antidepressants (IRR = 0.70, p = 0.013). Equinoctial seasonality was linked to a history of psychosis (OR = 1.76, p = 0.004). Conclusion Seasonality in BD is associated with distinct clinical features and treatment‐related outcomes. Specific seasonality patterns, particularly fall/winter depression, may reflect a more complex and difficult‐to‐treat illness course. These findings highlight the need for research to better characterize seasonality subtypes beyond a binary “seasonal” vs. “non‐seasonal” classification and to explore how these patterns influence the illness course and management of BD.

IntroductionAlcohol withdrawal syndrome (AWS) is among the most severe components of alcohol dependence (AD). Severe AWS, especially complicated with seizures and delirium, was a common reason for medical complications and death. The introduction of benzodiazepines decreased but did not eliminate those risks. The definition of AWS evolved in different versions of Diagnostic and Statistical Manual for Mental Disorders (DSM) from having hand tremor plus one or more other symptoms in DSM-III-R to any two or more symptoms in DSM-IV, which was retained in the current version of DSM-5.ObjectivesComparing subgroups of individuals with AUD based on AWS symptoms profile to define phenotypes for investigation of biological underpinnings of AWS phenomenology and treatment response.MethodsTreatment-seeking individuals diagnosed with DSM-IV alcohol dependence (AD; n=473; 35.5% females) were assessed with Psychiatric Research Interview for Substance and Mental Disorders (PRISM), Timeline Follow back (past 90 days) alcohol consumption, Penn Alcohol Craving Scale (PACS), Pittsburgh Sleep Quality Index (PSQI), Patient Health Questionnaire-9 (PHQ9), and General Anxiety Disorder-7 (GAD-7). Latent class analysis was used to classify subjects according to lifetime prevalence of AWS symptoms according to DSM-III-R or DSM-IV/5. Demographic and other clinical variables were compared among clusters by linear model ANOVA and chi-squared test.ResultsFour clusters were identified: subjects with (1) no history of meeting any AWS criteria (n=50); (4) with complete number of AWS symptoms, meeting both DSM definitions (n=259); and two clusters of subjects with smaller number of AWS symptoms: (3) those meeting only DSM-IV/5 criteria (n=94) and (2) those also meeting DSM-III-R criteria (n=70). Compared to cluster 2, the Cluster 3 members had higher frequency of anxiety, insomnia, and restlessness during withdrawal. The clusters also differed in the total number of drinks (p=0.021), average drinks per drinking days (p=0.013), history of anxiety disorder (p=0.006), substance-induced depression (p=0.019), and higher scores of PSQI (p<0.001), PHQ9 (p<0.001), and GAD-7 (p<0.001) with a relative increase in frequency between clusters as follows 1<2<3<4. Group 4 also had the highest percentages of subjects with history of seizure, delirium or hallucinations.ConclusionsDifferent definitions of AWS capture clusters of AD patients with different symptomatology, comorbidity and consumption patterns. Understanding of biological underpinnings behind those differences may guide improvement in personalized treatment selection.Disclosure of InterestNone Declared

ABSTRACT Objectives To explore the clinical impact of social determinants of health (SDoH) stress in bipolar disorder (BD) course of illness and to explore potential gender differences in SDoH stress endorsement. Methods 2280 individuals living with BD type I or II (62% women; 67.5% BD‐I) participating in the Mayo Clinic Bipolar Biobank who completed a questionnaire assessing positive and negative‐valent stressors in the 12 months prior to study enrollment were included for this analysis. Six negative stressors were used as proxies of SDoH. Generalized linear models were used to test the association between SDoH stressors and markers of illness severity as well as potential differences by sex by including an interaction term. Results After adjusting for age, sex, and recruitment site, SDoH‐related stressors were significantly associated with a more severe course of illness in both men and women. Except for unemployment rates, women reported more frequently SDoH‐related stress (i.e., lack of family support, financial problems, health coverage and access problems). Conclusions The study found significant gender differences in both the endorsement of specific SDoH stressors and in the prevalence of specific illness severity markers; however, it did not directly test the predictive relationship between gender differences in SDoH and illness severity markers. Future studies aiming to achieve health equity in mental health care should continue assessing social determinants of health stressors, incorporating a gender perspective to better understand their influence on bipolar disorder.

Abstract Background Acamprosate is an FDA-approved medication for the treatment of alcohol use disorder (AUD), however, only about 50% of patients will respond to acamprosate. Previously, we conducted an acamprosate trial (n=442), which is one of the largest acamprosate clinical studies ever conducted and currently the only study with multi-omics data. To date, no biological measures are utilized to predict response to acamprosate treatment. Aims & Objectives We set out to apply our established pharmaco-omics informed genomics strategy to identify potential biomarkers associated with acamprosate treatment response. Method Our open-label acamprosate clinical trial recruited 442 patients with AUD, all of whom were treated with acamprosate for three months. The primary outcomes were 1) relapse to alcohol use and 2) relapse to heavy drinking. Alcohol consumption was measured using the self-reported timeline follow back method. Plasma protein levels were measured using Olink proximity extension immunoassays. Results Our established “Pharmaco-Omics-informed genome-wide association study (GWAS)” research strategy identified 12 proteins, including interleukin-17 receptor B (IL17RB), that were associated with acamprosate treatment response. A GWAS for IL17RB concentrations identified several genome-wide significant signals. Specifically, the top hit single nucleotide polymorphism (SNP) rs6801605 with a minor allele frequency of 38% in the European American population mapped 4 kilobase (Kb) upstream of IL17RB, and intron 1 of the choline dehydrogenase (CHDH) gene on chromosome 3 (p: 4.8E-20). The variant genotype (AA) for the SNP rs6801605 was associated with lower IL17RB protein expression. In addition, we identified a series of genetic variants in IL17RB that were associated with acamprosate treatment outcomes. Furthermore, the variant genotypes for all of those IL17RB SNPs were protective against alcohol relapse. Finally, we demonstrated that the basal level of mRNA expression of IL17RB was inversely correlated with those of nuclear factor-κB (NF-κB) subunits, and a significantly higher expression of NF-κB subunits was observed in AUD patients who relapsed to alcohol use. Discussion & Conclusions This study illustrates that IL17RB genetic variants might contribute to acamprosate treatment outcomes. This series of studies represents an important step toward generating functional hypotheses that could be tested to gain insight into mechanisms underlying variation in acamprosate treatment response phenotypes.

Abstract Background: Responsiveness to mood-stabilizing pharmacotherapy varies in bipolar disorder (BD). We investigated clinical correlates of second-generation antipsychotic (SGA) treatment response and conducted the first genome-wide association study (GWAS), including exploratory polygenic scores (PGS), of SGA pharmacogenomic treatment response in BD. Methods: Treatment response was quantified using the Alda scale, and GWAS was performed using Alda-A score, controlling for sex, genotyping batch, and the genomic principal components. Results: The cohort included 2,159 adults with BD (1,416 BD-I, 691 BD-II, 51 schizoaffective BD), mean age 41.8 years, 62% female, 84% white, and 14% Hispanic. Nearly half (48%) were treated with SGAs. Current SGA users were younger (41.2±14.7 vs. 42.5±15.3 years, p =0.040), more likely to be Hispanic (14% vs. 11%, p =0.047), had a higher body mass index (BMI; 30.4±7.6 vs. 29.5±7.1 kg/m 2 , p =0.005). Lifetime comorbidity patterns for current SGA users include higher rates of manic psychosis (29% vs. 17%, p <0.001) and eating disorders — Anorexia Nervosa (7% vs. 4%, p =0.003), Bulimia Nervosa (7% vs. 4%, p =0.003), and Binge Eating Disorder (14% vs. 11%, p =0.030). We detected a genome-wide significant association between SGA Alda-A scores and GAS7 variants (top variant: rs202127418, β =-2.998, p =4.96E-08). However, SGA response was not significantly associated with PGS for schizophrenia, BD, and major depression (FDR>0.05). Conclusions: SGAs are frequently utilized as mood stabilizers in patients with BD and are associated with manic psychosis and eating disorders. GAS7 variants may predict SGA response, but larger, more diverse cohorts are needed for validation.