The Pharmacogenomics Journal

NAT2 is an important pharmacogene which encodes the N-acetyltransferase 2 enzyme that is involved in the metabolism of multiple medications, and variants in this gene can affect patient response to these medications. CPIC has published a clinical guideline for prescribing hydralazine using NAT2 genotypes. Just prior to the guideline, updated NAT2 star allele numbering and definitions were released, differing somewhat from the historical nomenclature. Clinical pharmacogenomic testing panels often test for the most common star alleles, so knowledge of the most common updated NAT2 star alleles is critical for the implementation of the CPIC NAT2/hydralazine guideline. We first determine NAT2 diplotype frequencies from UK Biobank (UKBB) 200k phased genomes, then analyzed allele, diplotype, and phenotype population frequencies from the All of Us Research program, PennMedicine BioBank (PMBB) and UKBB 500k datasets. We found that analyzing NAT2 diplotypes from phased data provides critical information for algorithms designed to predict diplotypes from unphased data. We observed that NAT2*5, *6, and *4 were the most common star alleles in that order, and the top 11 most frequent NAT2 star alleles were the same across all biobanks. However, differences in star allele frequencies across biogeographical populations were observed. The largest difference led to a higher frequency of NAT2 poor metabolizer phenotypes as compared to rapid and intermediate metabolizer phenotypes in all global populations except in the EAS population, where NAT2 poor metabolizers were in the minority.

BackgroundHigh grade serous ovarian cancer (HGSC) is initially a responsive tumor to platinum (Pt)-based therapy. Pt resistance in HGSC is associated with epigenetic modifications and hypomethylating agents (HMAs) have been studied as carboplatin resensitizing agents. As DNA methylation is detectable in cancer cells and in blood, here we aimed to develop a blood-based methylation signature associated with cancer and cancer recurrence in HGSC. ResultsWe evaluated genome-wide DNA methylation in de-identified peripheral blood mononuclear cells (PBMCs) from women 1) without cancer (controls, n=20); 2) newly diagnosed HGSC (prior to treatment, Pt-naive, n=60) 3) Pt-resistant recurrent HGSC before and after treatment with the novel HMA/DNA methyltransferase inhibitor (DNMTI) guadecitabine (Pt-resistant, n=30). The Pt-resistant patients were enrolled in NCT02901899 clinical trial testing guadecitabine and the PD-1 inhibitor pembrolizumab. DNA extracted from PBMCs was analyzed by using Infinium MethylationEPIC BeadChips. There were 30,369 differentially methylated loci (DMLs) in Pt-naive patients vs. controls (adj. p < 0.05, {beta} >10%), with most loci being demethylated. Enriched pathways in PBMCs from cancer patients included mechanisms of cancer, neutrophil degranulation, and cancer-related signaling pathways (PI3K/AKT, STAT3, HGF, interleukins). The number of DMLs was greater (880 DMLs; adj. p<0.05, {beta}>10%) in Pt-resistant vs. Pt-naive patients, and top enriched pathways associated with Pt-resistant HGSC included pathways in cancer, metabolic pathways, platelet activation, ABC transporters and signaling pathways (calcium, PI3K/AKT, MAPK, Ras, ErbB, Hippo, Wnt). Massive genomewide hypomethylation 5 days after treatment with guadecitabine was observed (13,742 DMLs; adj. p<0.05, {beta}>10%), which persisted 30 days after discontinuation of treatment. Pathways enriched by hypomethylated genes in PBMCs following guadecitabine treatment interestingly included pathways related to neuronal signaling, such as glutaminergic receptor signaling, axonal guidance signaling, synaptic long-term depression, synaptogenesis signaling and serotonin receptor signaling. Deconvolution analysis of the methylome data of PBMCs from Pt-resistant recurrent HGSC before versus after HMA treatment predicted increased naive B cells, memory and naive CD+ T cells, naive CD4+ T cells, and neutrophils and decreased monocytes. ConclusionsWe propose new DMLs associated with Pt-naive versus Pt-resistant HGSC. These findings can lead to new biomarkers for HGSC.

Pharmacogenetics (PGx) has traditionally focused on a small number of high-impact variants affecting drug response due to the fact that PGx studies are labor-intensive and therefore low-throughput. Population biobanks linked to electronic health records (EHRs), including the UK Biobank (UKB) with prescription data for [~]230,000 individuals offer opportunities to scale PGx research. This, however, comes with a challenge as EHRs do not provide direct treatment response outcomes. One way to overcome this is to draw indirect drug response phenotypes from prescription records. Here, we propose preSCRIPT, a framework to filter and annotate raw prescriptions from the UKB to derive phenotypes for analyses which includes an algorithm to distinguish short prescription gaps from true dose changes. As a proof of concept, we applied preSCRIPT to warfarin, paracetamol, codeine, amitriptyline, simvastatin, aspirin, and amlodipine and derived therapy length and median daily doses. We tested associations for those seven drugs and two phenotypes across SNPs, cytochrome P450 (CYP) genes, and HLA alleles. We replicated known associations such as CYP2D6 variants with amitriptyline therapy length and dose, CYP2C9/CYP4F2/CYP2C19 with warfarin dose, and CYP2D6 with codeine dose. For drugs without formal PGx guidelines, we identified an association between CYP2D6 activity and aspirin therapy length and several SNPs, including rs62471929 (CYP3A5), a variant for amlodipine dose, replicated in an independent hold-out set. Overall, our study shows that preSCRIPT can recover established PGx associations, prioritize exploratory novel candidate loci, and may serve as a tool for large-scale pharmacogenomics.

Background The use of oral retinoids and valproate during pregnancy can cause birth defects. In 2018, the EMA revised Pregnancy Prevention Programs (PPPs) for these medications. Pharmacy technicians in Denmark dispense prescription medications and must counsel customers. Aims This study aimed to examine knowledge of the teratogenicity of oral retinoids and valproate and use of the relevant PPPs among pharmacy technicians in Denmark. Methods A cross-sectional survey was conducted in spring 2025 using questionnaires developed for and tested in an international project. Data was collected via relevant Facebook groups and email invitations. Descriptive statistics were used for analyses. Results For oral retinoids, 80 respondents were analyzed; 95% were women, 86% were pharmacy technicians, the mean age was 37.2 years. Most dispensed oral retinoids several times per month. Two respondents did not know retinoids were teratogenic. The most used PPP measure was the outer packaging warning (54%). Informing women about teratogenic effects was the most common practice. For valproate, 41 respondents were analyzed. Their characteristics were similar to those of respondents in the oral retinoid survey. Most dispensed valproate once per month. One-third did not know valproate was teratogenic. The outer packaging warning was used by 19%. The most common practice was referring to the prescribing physician if pregnancy was suspected. Conclusion Danish pharmacy technicians knowledge about teratogenic drugs and the PPP was poorer than that of pharmacists, especially regarding valproate, and requires attention in educational programs. The feasibility of PPP measures for both oral retinoids and valproate should be optimized.

BackgroundPharmacogenomic testing (PGx) can optimise drug efficacy and minimise toxicity, but the extent of prescriber adherence to PGx recommendations remains unclear. We aimed to quantify clinician adherence to international genotype-guided prescribing recommendations in a cohort of paediatric oncology patients. MethodsWe reviewed files of children enrolled in the MARVEL-PIC (NCT05667766) randomised control trial, who had PGx recommendations available. Patients were included if 12 weeks had passed since their PGx report was released to clinicians. Prescribing events were identified for actionable PGx recommendations, and classified as "explicitly followed", "inadvertently followed", or "not followed". Adherence was assessed by patient, drug, and recommendation. Results2,063 PGx recommendations were available for 216 patients. 64 (3.1%) recommendations were actionable for 44 patients and 10 drugs within the 12-week study period. Recommendations were explicitly followed in 57/288 (19.8%) of prescribing events, inadvertently followed in 145 (50.3%), and not followed in 86 (29.9%). Mercaptopurine demonstrated the highest rate of explicit adherence (87.5%). No significant associations were observed between adherence and age group, cancer type, drug type, or strength of recommendation. ConclusionAdherence to pharmacogenomic recommendations was very low, highlighting the need to understand barriers to PGx implementation, and consideration of clinical decision supports to facilitate adherence. Plain Language SummaryPharmacogenomic medicine (PGx) looks at how our genes affect our response to drugs, including their effectiveness and toxicity. Through genetic analysis we can create recommendations for drug dosing, avoidance, and monitoring. The MARVEL-PIC study aims to understand if having PGx recommendations decreases the rate of adverse events in children with cancer. We aimed to understand how often prescribers follow PGx recommendations after they are made available, in the MARVEL-PIC trial. To do this, we reviewed medical records and identified relevant prescribing events. We marked these as "recommendation explicitly followed", "recommendation not followed", or "recommendation inadvertently followed" (where the recommendation was followed, but it wasnt clear if this due to PGx). We found that when recommendations were available, they were only explicitly followed in around 20% of cases. In 50% of cases, they were followed but it was unclear whether this was due to PGx. In the remaining 30%, they were not followed. We also found that alerts on our electronic system were fired in about 80% of events where the recommendation was not followed, but did not change the outcome. These findings show that prescriber adherence to PGx recommendations is low. We need to better understand why this is the case and implement more specific tools to assist prescribers in following recommendations. Article HighlightsO_LIPharmacogenomic (PGx) testing can reduce adverse drug reactions by guiding drug choice, dosing, and monitoring. C_LIO_LI!Prescriber to PGx recommendation adherence has not been widely investigated. C_LIO_LIRetrospective analysis showed that explicit adherence to recommendations occurred in only 19.8% of relevant prescribing events. C_LIO_LIIn 50.1% of prescribing events, recommendations were followed, but there was no clear reference to PGx. C_LIO_LIMercaptopurine had the highest explicit adherence (87.5%) from the drugs analysed. C_LIO_LIThere were no statistically significant associations between adherence and age group, cancer type, drug type, or recommendation strength. C_LIO_LIRecommendations were explicitly followed in 29% of events where an interruptive alert was fired, and inadvertently followed in 8%. C_LIO_LITailored interruptive alerts have been shown to increase adherence in other studies, suggesting that the specific design of interruptive alerts may influence adherence. C_LIO_LIWe concluded that explicit prescriber adherence to PGx recommendations is very low (19.8%), and further research needs to be done to understand barriers to implementation. C_LI

BackgroundCardiovascular disease (CVD) risk is managed in primary care using lipid-lowering therapies (LLTs) and antihypertensives (AHTs) for primary (no prior CVD) or secondary (with prior CVD) prevention, but patients may discontinue treatment. Little synthesised real-world data for LLT/AHT discontinuation exists. MethodsWe systematically reviewed English language reports of observational studies from PubMed, EMBASE, Web of Science, and CINAHL published from 2010-2025 describing discontinuation/restarting prevalence for first-to-third line LLTs/AHTs used for CVD prevention in primary care (PROSPERO: CRD420250599340). Data were extracted on discontinuation/restarting prevalences and associations between discontinuation and sociodemographic factors. FindingsOf 5,756 records, 31 (16 LLT; 15 AHT) reports were included representing 9,146,252 patients. Risk of bias was generally low except for two papers with substantial risk of bias from unmeasured confounding. LLT median (IQR) discontinuation and restarting prevalences were 43% (38%; 54%) and 43% (22%; 64%), respectively. AHT discontinuation and restarting prevalences were 41% (30%; 49%) and 28%, respectively. Discontinuation/restarting prevalence depended on discontinuation definition and indication. Patients aged around 65 years old were less likely to discontinue than younger or older patients, for both LLTs and AHTs. Women discontinued LLTs more often irrespective of indication; men discontinued AHTs more often for primary prevention. Income-based socioeconomic position (SEP) measures were associated with discontinuation, but composite SEP measures were not. Minority ethnic groups were more likely to discontinue LLTs and AHTs. InterpretationThis systematic review of real-world data identified discontinuation inequities in first-to-third line LLTs and AHTs based on age, sex, and ethnicity. Awareness of these patterns and additional research into patient-level drivers of drug discontinuation could improve health equity by addressing LLT/AHT discontinuation in the highest-risk patients. FundingThis work was funded by the NIHR UCLH BRC. No funders had any role in data collection, analysis, manuscript preparation, or the decision to publish. Research in contextWe searched PubMed for reviews, systematic reviews, and meta-analyses examining associations between age, sex, socioeconomic position, ethnicity (search string: "sociodemograph*" OR "age" OR "sex" OR "socioeconomic status" OR "socioeconomic position" OR "ethnic*" OR "race" OR "racial" OR "Asian*" OR "India*" OR "Pakistan*" OR "Bangladesh*" OR "Black" OR "African" OR "Afro*") and discontinuation (search string: persist* or discontinu* or stop*) of antihypertensives/lipid-lowering drugs (search string: "anti$hypertensive" OR "blood*pressure lowering" OR "ACE inhibitor" OR "angiotensin receptor blocker" OR "calcium channel blocker" OR "thiazide-like diuretic" OR "thiazide diuretic" OR "lipid*lowering" OR "lipid*reducing" OR "statin" OR "HMG-CoA reductase inhibitors" OR "proprotein convertase subtilisin/kexin type 9" OR "PCSK9 inhibitor"). We did not restrict reports by date or language. Of 711 results, one relevant article was found. Two more relevant articles were found in the searches performed for this systematic review. One 2017 systematic review of twenty-two real-world studies found that for nineteen studies with a dichotomous discontinuation outcome, 16% to 93% of patients discontinued statins across follow-up time of 0 days (cross-sectional studies) to median 4.1 years follow-up. The authors did not report on the proportion of patients discontinuing by sociodemographic group. A 2018 systematic review and meta-analysis of RCTs and real-world data found that for patients aged [&ge;]65, lower income was associated with discontinuation across seven studies (odds ratio [95% confidence interval] 1{middle dot}20 [1{middle dot}06 to 1{middle dot}36]), though the degree of heterogeneity in the studies used for meta-analysis was high (I2 = 0.89). Female gender (1{middle dot}03 [0{middle dot}98 to 1{middle dot}09]) was not associated with statin discontinuation, and there was a trend towards an association between Black/non-White race and discontinuation (1{middle dot}57 [0{middle dot}92-2.68]). A 2024 systematic review of 52 RCTs and real-world studies found that the prevalence of statin discontinuation ranged from 0.8% to 70.5%, and was higher for primary prevention, and that male sex and non-White ethnicity were associated with statin discontinuation. Our search found no prior systematic reviews or meta-analyses describing differences in discontinuation of AHTs by age, sex, SEP or ethnicity. Added value of this studyThis study is the first systematic review of sociodemographic factors influencing antihypertensive discontinuation in primary care, and complements the findings described above with new evidence on statin discontinuation in real-world settings. With respect to statins and/or ezetimibe, we found that younger (below 60) and older (above 75) patients were more likely to discontinue statins, for both primary and secondary prevention. Female sex was associated with a small but consistent increase in statin discontinuation across our included studies. Individual income appeared to associate with statin discontinuation, but not composite SEP measures such as the Indices of Multiple Deprivation. For antihypertensives, we found that younger and older patients were more likely to discontinue for both primary and secondary prevention, with discontinuation at its lowest around 70 years. Male sex was associated with a small but consistent increase in discontinuation in primary prevention but was associated with marginally reduced discontinuation in a larger study of patients using AHTs for mixed prevention. Lower individual income appeared to positively associate with antihypertensive discontinuation, but composite SEP measures did not. In all studies reporting ethnic differences in discontinuation, non-majority ethnic groups were consistently more prone to discontinuation. Implications of all the available evidenceNon-persistence rates for lipid-lowering medications and antihypertensives are considerable and constitute a possible avenue to reduce CVD. In the first comprehensive evidence synthesis across socio-demographic groups, we show discontinuation rates in real-world settings differ across groups, which may contribute to existing health inequities. For statins, it is unclear how sex associates with discontinuation, given the inconsistency of results across different systematic reviews and meta-analyses. Low income and minority ethnic group membership are associated with statin discontinuation. Our findings suggest that commonly used AHTs are discontinued more often in men than in women, in the youngest and oldest patients, and in minority ethnic groups. Lower individual income may associate with statin discontinuation, but belonging to a lower SEP group such those derived from the Indices of Multiple Deprivation was not associated with statin discontinuation. Future research efforts should address the intersectionality in these patterns, to ascertain whether sociodemographic disadvantages combine to drive higher discontinuation rates in specific patient subgroups. Data on discontinuation by the type of LLT/AHT used (e.g. angiotensin-converting enzyme inhibitor versus for AHTs) should also be collated. Causes for antihypertensive/lipid-lowering medication discontinuation should be investigated using qualitative methods to ascertain reasons for discontinuation in patient groups directly, utilising underrepresented patient populations where possible. Data from such qualitative studies may inform future interventions to reduce discontinuation in the most at-risk patient groups. Clinicians should heighten efforts to maintain or reinitiate therapy in those prone to lipid-lowering therapy and antihypertensive discontinuation, assuming no clinical contraindications.

Drug repurposing offers a practical strategy to identify new therapeutic uses for approved drugs, potentially reducing the time and cost associated with conventional drug development. We present a novel three-stage drug repurposing pipeline that integrates knowledge graph-based gene prediction, network-based drug-disease association analysis, and systematic classification of candidate drugs by therapeutic class. The pipeline integrates DGLinker to predict novel disease-associated genes, SAveRUNNER to identify drug repurposing candidates, and ATC Category Enrichment Analysis (ATCEA) to prioritise candidates by pharmacological class. We benchmarked the pipeline across twelve diseases using DrugBank and MEDI2-HPS as validation resources. Utilising DGLinker-expanded disease-gene sets as input increased the number of predicted repurposed drugs, while overall discriminative performance remained stable across diseases (AUROC 0.71-0.77). Application of ATCEA consistently improved precision, F1-score, and specificity, while reducing recall, reflecting a conservative prioritisation strategy that contracts the candidate space while retaining pharmacologically coherent drug-disease candidates. We further applied the pipeline to amyotrophic lateral sclerosis (ALS), a neurodegenerative disease with limited therapeutic options, and performed a deeper literature-based validation of the results. Incorporation of DGLinker-predicted genes substantially increased the number of significant candidate drugs and uncovered enriched ATC categories not identified using known ALS genes alone, including antidepressants and antipsychotics. Moreover, several drugs with supporting evidence available in the literature were identified only when DGLinker-predicted genes were used. Overall, 77 candidate drugs were prioritised within significantly enriched ATC categories, several of which are supported by previously published studies. To provide exploratory real-world support for these findings, we further evaluated candidate drugs in a longitudinal electronic health record (EHR) dataset of 2361 patients with ALS from King's College Hospital. Although the number of evaluable drugs was limited due to sample size, the EHR analysis provided additional clinically relevant context for selected prioritised drugs and pharmacological classes. Our pipeline demonstrates potential to accelerate drug repurposing by integrating complementary computational approaches to each step of the process, providing an end-to-end framework that showed robust performance across benchmarking experiments and use cases.

BackgroundAddressing the underlying causes of atrial fibrillation (AFib) is critically important. While potential AFib-related genes have been recognized, the impact of modifying these genes in humans remains poorly understood. ObjectiveWe assessed the cellular dependencies of 309 genes previously associated with AFib through genome-wide association studies using data from the Cancer Dependency Map project, aiming to prioritize potential therapeutic targets with minimal off-target effects. MethodsWe analyzed CRISPR-Cas9 knockout (CHRONOS scores) and RNA interference (RNAi) knockdown (DEMETER2 scores) screening data from 1,927 human cell lines across 24 tissue types, focusing on tissues associated with AFib initiation, presentation, and progression: autonomic ganglia, central nervous system (CNS), and soft tissue. We examined the expression and dependency scores of the AFib-associated genes, identifying significant correlations between gene expression and cellular dependency within specific tissues using Pearson correlation coefficients and controlling the false discovery rate (FDR) at 5%. ResultsOut of the 309 AFib-associated genes, 206 genes (66.7%) had CHRONOS dependency scores and 229 (74.1%) had DEMETER2 dependency scores available. Several genes showed significant negative dependency scores (CHRONOS < -0.5) across multiple tissues, indicating potential off-target effects if inhibited. In contrast, we identified 12 genes with significant expression-driven dependencies within AFib-associated tissues. In CNS cell lines, HAND2 (R = -0.456, FDR = 0.002) and VGLL2 (R = -0.434, FDR = 0.005) showed significant negative correlations between gene expression and cellular dependency. In soft tissue cell lines, BEST3 (R = -0.679, FDR = 0.001) and PITX2 (R = -0.679, FDR = 0.001) also demonstrated strong negative correlations. Additionally, ERBB4 in CNS lines showed a significant negative correlation (R = -0.361, FDR = 0.048). These findings suggest that inhibiting these genes may selectively affect high-expressing cells in AFib-associated tissues while minimizing effects on other tissues. ConclusionOur analysis identified HAND2, VGLL2, BEST3, and ERBB4 as potential therapeutic targets for AFib, demonstrating significant expression-driven dependencies in AFib-associated tissues with no pan-tissue essentiality. These results provide a quantitative basis for developing targeted therapies with reduced off-target effects. CONDENSED ABSTRACTAtrial fibrillation (AFIB) is one of the most common cardiac arrhythmias with numerous known risk factors. Although many AFIB-associated genes have been identified, the impact of screening or the effects of modifying these genes in humans remain poorly understood. We examined CRISPR knockout and RNAi knockdown screen data from nearly 2,000 human cell lines to assess the cellular dependencies of 309 genes associated with AFIB, previously identified through genome-wide association studies. Some genes demonstrate broad cell dependencies across various tissue types, indicating potential off-target effects if inhibited. Conversely, HAND2, VGLL2, BEST3, and ERBB4 were identified as genes of interest because their genetic knockouts specifically impacted high-expressing cells from tissue lineages pertinent to AFIB and/or were not pan-dependent. Overall, analyses of genetic screen data identified AFIB-associated genes whose knockout or knockdown selectively affected cell lines of relevant tissue lineages, prioritizing targets for potential AFIB treatments.

Amitriptyline is commonly prescribed for chronic pain, yet treatment response and tolerability vary substantially. Genetic variation in CYP2C19 and CYP2D6 influences amitriptyline metabolism, but evidence linking pharmacogene status to clinical outcomes in chronic pain is limited. Amitriptyline is typically prescribed for chronic pain at lower doses than for depression, which may reduce pharmacogenomic effects on clinical outcomes. We analysed 1,146 participants with chronic pain from the Australian Genetics of Depression Study who reported amitriptyline use, treatment outcomes, and genotype data. Metaboliser phenotypes were assigned using PharmCAT. Associations with self-reported effectiveness and discontinuation due to side effects were examined using regression models adjusted for age and sex. Only CYP2C19 intermediate metabolisers showed nominally lower odds of discontinuation and reduced likelihood of reporting moderate effectiveness. Overall, pharmacogenetic phenotypes were not significantly associated with patient-reported amitriptyline outcomes in chronic pain, potentially reflecting the lower doses typically prescribed for pain management.

Motivation: Human liver microsome (HLM)-based metabolic stability assays are fundamental in early drug discovery, shaping pharmacokinetic profiles and oral bioavailability. However, these experimental assays are labor-intensive and time-consuming, limiting their application in large-scale virtual screening. Computational models can prioritize compounds at scale, yet most are classification-based, leaving quantitative and interpretable prediction of HLM half-life limited. Results: In this study, we developed a quantitative machine learning model for the direct prediction of HLM half-life (T1/2) by integrating 11,790 compounds combining in-house and curated public data. Among various combinations of molecular features and learning algorithms, the XGBoost model with RDKit 2D descriptors achieved the best predictive performance, with an RMSE of 0.507 and an R2 of 0.431 on an independent test set. Shapley Additive Explanations (SHAP) analysis identified lipophilicity and known metabolic soft-spot features as the primary contributors to the predictions. These results suggest that this quantitative approach provides a practical framework for defining metabolic stability margins, thereby supporting rapid Go/No-go decisions in preclinical drug discovery. Availability: The source code, data, and trained model are available at https://github.com/joshua-416/PredHLM.

Long COVID (LC) is a complex condition characterized by persistent, chronic multisystem manifestations, with a significant proportion of patients exhibiting neurological symptoms. Human ion channels (HICs), particularly potassium channels, are abundantly expressed in the nervous system and linked to key metabolic processes, making them potential candidates for understanding LC pathophysiology and drug repurposing. Meta-analysis of RNA-Seq datasets from COVID-19 recovered and LC patients was performed to identify altered HICs in LC. Differential gene expression analysis, functional enrichment analysis, and weighted gene co-expression network analysis (WGCNA) were performed to uncover key genes, pathways, and co-expression modules consisting of HICs, lipid metabolism-, and immune signaling-related genes. Drug-gene interaction analysis was performed to identify approved drugs targeting potential HICs. A total of 715 dysregulated genes, including eighteen HICs were identified, among which seven were potassium channels. Three significant modules containing HICs, lipid metabolism-, and immune signaling-related genes were identified and found to be associated with antigen processing and presentation, complement and coagulation cascades, and cytokine-related pathways. Approved drugs targeting KCNA6, KCNJ10, KCNN3, and KCNH4 were identified. With further experimental validation, these dysregulated potassium channels, supported by their co-expression networks and pathway associations, may act as potential candidates for drug repurposing in LC patients.

Drug repurposing -- the application of approved or shelved compounds to new therapeutic indications -- offers a cost- and time-efficient alternative to de novo drug discovery. However, the systematic identification of repurposing candidates from the rapidly expanding body of clinical trial data remains a significant challenge. Here I present a publicly accessible AI-powered tool that mines the ClinicalTrials.gov registry to identify approved drugs with under-explored therapeutic potential in high-value disease areas. The tool integrates natural language processing, mechanism-of-action pathway analysis, and trial density scoring to surface candidates where biological plausibility is high and clinical trial coverage is sparse. I demonstrate the tools utility across six cross-therapeutic case studies spanning oncology, cardiology, neurology, rare diseases, immunology, and infectious disease. Key findings include: the identification of Zonisamide as an under-explored combination candidate for obesity alongside GLP-1 receptor agonists; mechanistic validation of SGLT2 inhibitors in heart failure with preserved ejection fraction (HFpEF); and a novel cross-domain mapping of anti-TNF biologics to early-stage neurodegeneration via shared neuroinflammatory pathways. The tool is freely accessible and designed to lower the barrier for academic and industry researchers to systematically pursue repurposing opportunities.

Prader-Willi syndrome (PWS) is a rare multisystemic disorder characterized by obesity, endocrine dysfunctions, and psychiatric comorbidities, which imply frequent use of psychotropic medications. They account for atypical responses to standard dosages of psychiatric drugs. Pharmacogenetics could be part of the reason for this situation, potentially offering a valuable tool for individualized treatment. This study analyzed allelic and phenotypic frequency distributions of five of the main cytochrome P450 enzymes (CYP2D6, CYP2B6, CYP2C19, CYP2C9, CYP3A4) involved in psychiatric drug metabolism in 47 patients with genetically confirmed diagnosis of PWS and compared them to reference frequencies in the general European population. Allelic frequency comparisons between the European reference population and the overall PWS cohort revealed a significant global difference for CYP2B6, with CYP2C19 and CYP2D6 showing trends toward significance. Although no global allelic differences remained significant after false discovery rate correction, post-hoc analyses consistently identified an enrichment of reduced- or non-functional alleles CYP2B619 and CYP2D610 in patients with PWS. Predicted metabolizer phenotype analyses showed a significant shift toward intermediate metabolizers of CYP3A4 in the PWS cohort, with corresponding depletion of normal metabolizers. Subgroup analyses indicated that allelic differences were more pronounced in maternal uniparental disomy and non-deletion subtypes, particularly for CYP2B6, although no significant differences were observed between PWS genetic subtypes. Overall, results imply potential differences in metabolizing activity in PWS patients, and subsequent implications in drug efficacy and tolerability. These results support the idea that pharmacogenetic testing may improve therapeutic decision-making in PWS for psychiatric treatment. Larger studies are needed to confirm these preliminary results.

Selective inhibition of phosphodiesterase 4B (PDE4B) remains a promising strategy for preserving the anti-inflammatory benefit of PDE4 inhibition in chronic obstructive pulmonary disease while reducing PDE4D-associated tolerability liabilities. This study integrated SHAP-interpretable machine learning, natural product virtual screening, hierarchical docking, post-docking MM-GBSA, isoform cross-docking, binding-pocket comparison, ADMET prediction, and 100 ns molecular dynamics simulations to identify PDE4B-selective inhibitors from the LOTUS natural product database. A Random Forest classifier trained on curated ChEMBL PDE4B bioactivity data achieved an external performance with AUC-ROC = 0.955, accuracy = 0.893, F1-score = 0.896, MCC = 0.785, and prioritized 119,698 predicted actives from 276,518 LOTUS compounds. SHAP analysis identified BertzCT and TPSA as major contributors to predicted activity. Sequential Lipinski, PAINS, and QED filtering retained 14,210 candidates for structure-based evaluation. Extra precision docking identified four leads with PDE4B docking scores of -9.123 to -12.080 kcal/mol, all outperforming roflumilast (-7.658 kcal/mol). Cross-docking and post-docking MM-GBSA supported preferential PDE4B binding for three candidates. The top lead, LTS0048837, maintained a stable PDE4B-bound pose during simulation, with comparatively stronger interaction persistence than its PDE4D complex and the roflumilast reference. These findings nominate LTS0048837 as a computationally prioritized PDE4B-selective natural product lead requiring experimental enzyme, cellular, and pharmacokinetic validation.

BackgroundType 2 diabetes (T2D) represents a major global health burden, with over 700 GWAS loci identified. Translation to biological mechanisms remains challenging. This study employs systematic post-GWAS functional annotation to characterize the RPTOR locus, encoding Raptor, a scaffold protein critical for mTORC1 signaling and beta-cell function. MethodsWe analyzed 31 GWAS credible sets containing rs12950541 (chr17:80760693 G>A) using Open Targets Platform v24.12, encompassing 20 metabolic traits. L2G scoring, colocalization analysis, and QTL mapping in GTEx v8 were performed. Independent Variant Effect Predictor (VEP) analysis of the linkage disequilibrium (LD) block was conducted to characterize all variants in LD (D [&ge;] 0.7) with rs12950541. RNA-protein interaction networks were predicted using RNAct/catRAPID for key RPTOR transcripts and functionally enriched using ToppGene. Drug target and novelty analyses were performed using ChEMBL, PubMed, and ClinicalTrials.gov databases. Phenome-wide associations and regulatory annotations were obtained from the T2D Knowledge Portal. ResultsRPTOR was consistently ranked #1 L2G gene across all 31 credible sets (mean score 0.428, range 0.383-0.503). T2D showed strong GWAS-GWAS colocalizations (H4>0.8) with adiposity traits. Skeletal muscle demonstrated strongest QTL evidence with sQTL at P=1.21x10-16 and multiple eQTLs/tuQTLs. Critically, zero GWAS-QTL colocalizations and zero QTL in pancreatic islets, adipose, or liver highlight an "eQTL gap." VEP analysis of 140 LD partners revealed exclusively non-coding variants (100% MODIFIER impact), including 24 regulatory region variants and 2 transcription factor binding site variants. RNAct analysis revealed that the NMD transcript RPTOR-208 shows stronger RNA-protein interactions than the canonical transcript, with predicted binding partners including sulfonylurea receptors (ABCC8/ABCC9), IGF1R, and chromatin remodelers, enriched for glucose-mediated signaling and SWI/SNF complex pathways. ABCC8 is confirmed as the molecular target of sulfonylurea drugs (ChEMBL: CHEMBL2071), and literature analysis confirms that the RPTOR-ABCC8 RNA-protein interaction is completely novel, with no prior publications linking RPTOR transcript biology to sulfonylurea receptor function. T2DKP PheWAS confirmed 78 significant associations across 18 phenotype groups, revealing effects on acute insulin response, insulin sensitivity, HDL cholesterol, hepatic enzymes, and sleep traits, with transcription factor binding analysis showing that rs12950541 directly enhances p300 enhancer marking while reducing CTCF insulator binding. ConclusionsSeven convergent lines of evidence support rs12950541 as a strong candidate regulatory variant at RPTOR. Integration of post-GWAS annotation, VEP characterization, RNA-protein interaction networks, and translational drug target analysis converges on a regulatory mechanism involving splicing, chromatin remodeling, and metabolic signaling pathways. The novel predicted interaction between RPTOR-208 and ABCC8/ABCC9 suggests a previously unrecognized molecular bridge between mTORC1 signaling and KATP channel-mediated insulin secretion, with potential implications for understanding sulfonylurea-mTOR pathway crosstalk in T2D.

Drug-induced arrhythmias, particularly Torsades de Pointes (TdP), pose a significant risk to patient safety and can sometimes have life-threatening outcomes. They remain a major concern in drug development and regulation. Machine learning (ML) has become a powerful tool for analyzing complex biological and chemical datasets, enabling researchers to identify subtle patterns that differentiate safe compounds from those likely to cause dangerous cardiac effects. However, most existing in silico approaches do not sufficiently incorporate biological elements, relying heavily on chemical and structural properties or on computationally expensive simulations. Here, we introduce BioMADE, a novel ML framework that harnesses small-molecule-protein activity profiles from publicly available datasets to predict TdP risk without requiring exhaustive mechanistic annotation. Activity data from ChEMBL were used to train individual models for each gene, which predict activity values for any given compound. A curated set of arrhythmia-relevant genes was then used to construct a latent biological embedding (BioMADE embedding) for each molecule. We validated the performance of these features in distinguishing biological elements such as ATC3 class, showing superior classification performance compared with representations such as Molformer (lacks biological information) and MACCS (limited chemical properties) (0.85 AUROC vs 0.81 and 0.73, respectively). BioMADE representations served as input to a support vector machine classifier to discriminate TdP-inducing drugs from safe compounds. BioMADE achieved an AUROC of 0.89 in internal validation, indicating strong predictive performance. Against state-of-the-art models such as ADMEThyst, BioMADE achieved an AUROC of 0.74 on ADMEThysts validation set (vs. 0.72 for ADMEThyst). When we combined both approaches, the AUROC reached 0.77. These results demonstrate that BioMADE provides a scalable, biology-informed, and generalizable approach for predicting drug-induced toxicities. By integrating protein activity profiles into toxicology modeling, our framework highlights the critical role of human biology in adverse drug reaction prediction, an aspect often overshadowed by purely chemical or structural descriptors.

BackgroundIntensive systolic blood pressure (SBP) reduction is routinely employed to limit hematoma expansion in spontaneous intracerebral hemorrhage (ICH). However, the renal consequences of sustained aggressive SBP lowering in real-world clinical practice remain incompletely characterized. MethodsWe conducted a retrospective cohort study of adults admitted to the intensive care unit with spontaneous ICH between 2011 and 2023. Hourly SBP measurements over the first 7 days were standardized and clustered using k-Shape time-series clustering to identify distinct shape-based SBP trajectories. Acute kidney injury (AKI) was defined using Kidney Disease: Improving Global Outcomes (KDIGO) criteria. Multivariable logistic regression assessed associations between SBP trajectory cluster and AKI, adjusting for demographics, baseline illness severity, renal function, and nephrotoxic medication exposure. ResultsAmong 233 patients (mean age 61.2{+/-}14.1 years), two distinct SBP trajectories were identified: Cluster 1 (rebound SBP trajectory), a progressive upward SBP trajectory with gradual rebound, and Cluster 2 (rapid-drop SBP trajectory), a lower SBP trajectory characterized by rapid early reduction and sustained levels below 140 mm Hg. Overall, 70.4% developed AKI of any stage. Patients of Cluster 1 (rebound SBP trajectory) had significantly higher odds of AKI compared to those of Cluster 2 (rapid-drop SBP trajectory) (adjusted OR 1.97; 95% CI, 1.03-3.78). Higher maximum nicardipine dose was independently associated with AKI (OR 1.14 per mg/h; 95% CI, 1.03-1.26). SBP trajectory cluster was not significantly associated with hematoma expansion (defined as a binary outcome based on physician-documented expansion vs. no expansion), neurological outcomes, or 1-year mortality. ConclusionsIn ICH patients, rapid early decline in SBP followed by relative stabilization at lower levels (<140 mm Hg) is associated with increased risk of AKI without clear neurological benefit. These findings highlight the importance of balancing cerebral hemorrhage control with renal perfusion and support cautious implementation of intensive BP targets in clinical practice.

Background CLN2 disease, Neuronal Ceroid Lipofuscinosis (NCL) type 2, is a rare, genetic neurodegenerative condition predominantly affecting children. CLN2 disease is characterized by seizures, language and motor decline, vision loss, and premature death. Currently, the only regulatory-approved therapy is the enzyme replacement therapy (ERT) Cerliponase alfa, administered fortnightly via intracerebroventricular infusion as a lifelong treatment. While ERT has been shown to slow motor and language decline, it is not curative and does not fully address disease progression, including retinal degeneration. To better understand the lived experience of affected families, and perspectives on current and emerging treatments, we conducted a community survey of parents and caregivers of individuals with CLN2 disease. Methods A 25-question anonymous, voluntary survey was distributed through the BDSRA Foundation and international partner patient advocacy organisations via email and social media. Eligible participants included current and bereaved parents or primary caregivers of individuals with CLN2 disease, regardless of treatment history. The survey explored treatment experiences, unmet needs, and knowledge of and attitudes toward emerging therapeutic approaches, particularly gene-based therapies. Results Ninety-eight respondents from 19 countries completed the survey. Fifty-seven respondents reported current or prior use of ERT, with 94.7% (n=54/57) actively receiving treatment at the time of survey. ERT was perceived to provide greatest benefit for motor function and seizure control; however, respondents reported substantial treatment burden (mean burden score 4.8/7, n=66). Despite treatment availability, 94.9% of respondents (n=75/79) indicated a need for alternative therapeutic options and 94.8% (73/77) expressed interest in learning more about gene therapy. Overall, 72.4% (n=55/76) reported they were likely or very likely to consider participation in an investigational gene therapy trial. Key factors influencing decision-making included potential safety risks (57.9%, n=44/76), preclinical safety and efficacy evidence (54.0%, n=41/76), and whether ERT discontinuation would be required to participate (54.0%, n=44/76). Conclusion While ERT has altered the treatment landscape for CLN2 disease, this survey highlights the ongoing disease burden and treatment challenges experienced by families. Findings demonstrate strong community interest in next-generation therapies that may reduce treatment burden and provide more comprehensive disease modification, including effects on both central nervous system (CNS) and ocular manifestations.

Background: The Cilostazol Albumin Treatment in Subarachnoid Hemorrhage (CATS) trial evaluates eight active cilostazol-human albumin regimens plus control in patients with aneurysmal subarachnoid hemorrhage. We summarized the rationale for the primary statistical design, compared alternative Phase II methodologies, and evaluated reduced-arm sensitivity scenarios. Methods: The binary primary endpoint is Common Data Elements-defined delayed cerebral ischemia within 14 days after randomization. The selected design is Bayesian adaptive, with a burn-in phase, response-adaptive randomization among active arms while maintaining fixed control allocation, four interim analyses, early stopping for expected success or futility, and a two-dimensional normal dynamic linear model. Primary operating characteristics were obtained from 1,000 virtual trials per scenario using Fixed and Adaptive Clinical Trial Simulator version 7.0.0. Exploratory simulations evaluated six-, four-, and two-active-arm configurations and simplified alternative designs. Results: Compared with fixed equal allocation, the Bayesian adaptive design preserved an approximately 10% false-success probability under the global null while improving probability of success and efficiency in clinically relevant scenarios. Under the Realistic scenario, probability of success increased from 0.61 to 0.86, expected sample size decreased from 400 to 308, and expected duration decreased from 235 to 187 weeks. Under common thresholds, null probability of success was 0.098 for the full anchor and 0.073 for Reduced-6; Reduced-6 probabilities of success were 0.774 and 0.765 in the Realistic and Realistic2 scenarios. However, Reduced-6 omitted two monotherapy anchors and was less robust in Backwards2. In the comparator simulation, the selected design had probability of success of 0.858 and expected sample size of 308.3 under the Realistic scenario, compared with 0.624 to 0.845 and approximately 352 to 400 for simplified comparators. Conclusions: For identifying the most promising cilostazol-human albumin regimen for Phase III rather than confirming efficacy, the Bayesian response-adaptive design with two-dimensional normal dynamic linear model borrowing is more efficient and better aligned than simplified comparators. The full nine-arm design remains preferable because it preserves the complete therapeutic discovery space and is more robust to misspecified or non-smooth response surfaces.

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 x 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.