Beitelshees, Amber L.

ABSTRACT Ischemic events after contemporary percutaneous coronary intervention (PCI) are uncommon but carry high morbidity/mortality. Identifying patients at highest risk is critical to guide dual antiplatelet therapy (DAPT) intensity while minimizing bleeding. Current machine learning (ML) model‐developed risk scores do not incorporate pharmacogenetic data. To develop and externally validate ML models that integrate clinical, demographic, and CYP2C19 genetic information to predict 1‐year ischemic outcomes post‐PCI to refine clinical decision making. We analyzed 8317 patients from TAILOR‐PCI trial ( n = 4572) and the Precision PCI registry ( n = 3745). The outcome was a composite of cardiovascular death, myocardial infarction, stroke, and stent thrombosis at 1 year. Boruta feature selection identified 11 predictors. Multiple ML algorithms were trained in TAILOR‐PCI and validated in Precision PCI using cross‐validation and synthetic minority oversampling (SMOTE). Model performance was assessed by area under the receiver operating characteristic curve (AUC), sensitivity, and specificity. The best external performance was achieved with a support vector machine (SVM, polynomial kernel) (AUC 0.667; sensitivity 0.871; specificity 0.282), while XGBoost provided a more balanced profile (AUC 0.619; sensitivity 0.442; specificity 0.688). Variable importance for the SVM polynomial model demonstrated that all 11 included Boruta feature selected predictors had relatively high importance (> 75). ML models trained on large clinical trial and real‐world registry datasets can help identify the small subset of patients at high ischemic risk after PCI. This distinction is of relevance because ischemic events are rare, and most patients may safely de‐escalate DAPT to reduce bleeding risk while maintaining ischemic protection through a multimodal approach to risk stratification. Trial Registration: TAILOR‐PCI URL: https://clinicaltrials.gov/ct2/show/NCT01742117

The 2025 American College of Cardiology/American Heart Association/American College of Emergency Physicians/National Association of Emergency Medical Services Physicians/Society for Cardiovascular Angiography & Interventions acute coronary syndrome guideline focuses on strategies to reduce bleeding risk with antiplatelet therapy yet lacks any recommendation related to CYP2C19 genotyping. The impact of CYP2C19 loss‐of‐function alleles on the effectiveness of clopidogrel is well documented, and although prasugrel and ticagrelor more effectively reduce the risk for atherothrombotic events compared with clopidogrel in patients with a CYP2C19 loss‐of‐function allele, clopidogrel reduces bleeding risk without an increase in atherothrombotic events compared with prasugrel or ticagrelor in those without a loss‐of‐function allele. Accordingly, an American Heart Association Scientific Statement supports CYP2C19 genetic testing before oral P2Y 12 inhibitors are prescribed. This commentary summarizes the evidence in support of CYP2C19 ‐guided P2Y 12 inhibitor selection in the context of other 2025 acute coronary syndrome guideline recommendations and urges future guidelines to incorporate recommendations for CYP2C19 genotyping, especially for those at high bleeding risk.

Introduction: Identifying patients at risk for ischemic events after percutaneous coronary intervention (PCI) relies on traditional analysis of limited clinical and imaging variables. Machine learning (ML) has shown promise in effectively predicting cardiovascular risk in population studies. While existing ML models mainly predict mortality and incorporate clinical variables, there is a lack of tools that have utilized genetic data and that predict ischemic events. Aims: This study aims to develop and validate a ML model incorporating genotyping and clinical data to enhance prediction of ischemic outcomes for PCI patients utilizing large prospectively derived diverse datasets. Methods: Patients from the TAILOR-PCI trial (n=5302) were utilized for model development. 50% of the sample was utilized for Boruta feature selection and 50% for training and testing using cross validation. Features included demographics, medical history, medications, PCI characteristics, and genetic data (specifically, CYP2C19 *2, *3, *17 alleles). The primary endpoint was a composite of cardiovascular death, myocardial infarction, stroke, stent thrombosis, and severe recurrent ischemia at 12 months. Multiple ML classification algorithms, including Support Vector Machine (SVM) polynomial, Random Forest, Light gradient boost, XG Boost, among others, were benchmarked for their prediction performance on rare events. The top performing classifiers were externally validated on an independent dataset from the PRECISION PCI study (n=3,745). Results: Mean participant age of the training set was 64.2 ± 11.0 years, with 75.4% being male. During follow-up of 12 months, among 4,572 patients in the entire cohort 343 (7.5%) met the primary outcome. The SVM polynomial model demonstrated the highest area under the curve (AUC) of 0.67 for predicting the primary outcome with test dataset. The sensitivity, specificity, precision, and recall were 0.87, 0.28, 0.07, and 0.87 respectively (Figure). Peripheral arterial disease, body mass index, and age were among the top variables by feature importance. Conclusion: ML models incorporating both clinical and genetic data are feasible and highly promising in predicting major adverse cardiac events that may help guide use of anti-platelet drug therapy. The AUC values are reasonable given imbalances and misclassifications in datasets, and further model optimization with prospective utilization of the model will be paramount.

Introduction: Hydrochlorothiazide (HCTZ) is among the most commonly prescribed antihypertensives in the US, yet <50% of HCTZ treated patients achieve blood pressure (BP) control. Hypothesis: Integrating metabolomic and genomic profiles of HCTZ treated patients could identify novel pathways/biomarkers for optimizing the use of HCTZ. Methods: Primary analysis included 228 white hypertensives treated with HCTZ from the Pharmacogenomic Evaluation of Antihypertensive Response (PEAR) study. Genotyping was determined via Illumina Omni 1M-Quad Chip and untargeted metabolomics was performed on baseline fasting plasma samples using a GC-TOF MS platform. Pathway analysis was used to integrate metabolites significantly associated with HCTZ BP response (FDR<.05) and genetic signals at p<5x10-5 from HCTZ BP genome-wide analysis. Replication was conducted in another 212 white HCTZ treated patients. Replicated genetic signals were used to create a BP response score by summing their BP lowering alleles. We validated this score by testing it in 196 white HCTZ treated patients in the Genetic Epidemiology of Responses to Antihypertensives (GERA) study. Results: Thirteen metabolites were significantly associated with HCTZ systolic BP (SBP) and diastolic BP (DBP) responses (FDR<.05). Metabolomics-genomics integration revealed significant metabolites along with rs2727563 PRKAG2, rs12604940 DCC, and rs13262930 EPHX2, in the netrin signalling pathway (p=1x10-5), as potential markers significantly influencing HCTZ BP response. We successfully replicated those 3 genetic signals and added their alleles to create a response score which explained 11.3% and 11.9% of the variability to HCTZ SBP and DBP responses, respectively. We found that patients carrying one “response” allele had a significantly worse response than carriers of six alleles (ΔSBP/ΔDBP: -1.5/1.2 vs -16.3/-10.4 mmHg, respectively, SBP score p=1x10-8 and DBP score p=3x10-9). We further validated this response score by testing it in the GERA (DBP score p=0.03, SBP score p=0.07). Conclusion: This study highlights the power of using different omics to identify novel pathways/biomarkers of drug response, and suggests that PRKAG2, DCC and EPHX2 might be important determinants of HCTZ BP response.

Introduction: β-blockers, including atenolol, are among the most widely prescribed antihypertensive drug classes. However, their use has been associated with hyperglycemia and incident diabetes, which poses concern for prescribers and patients. The Pharmacogenomic Evaluation of Antihypertensive Responses (PEAR) study is a multicenter clinical trial with one of its objectives to identify genetic influences on adverse metabolic events in patients treated with atenolol. Here we report a genomics-metabolomics integrative approach to identify biomarkers associated with atenolol-induced glucose change. Methods: PEAR participants were genotyped using the Illumina Omni 1M-Quad Chip. Metabolite profiles were acquired through GC-TOF-MS. Association between baseline levels of metabolites and glucose change was conducted using a multivariate linear regression model adjusted for age, gender and baseline fasting glucose. Genomics-metabolomics integration included selecting SNPs within genes involved in the synthesis and degradation of top metabolites and testing their effect on glucose change in 234 atenolol-treated Caucasians in PEAR, adjusting for age, gender, baseline glucose and principal components 1 and 2. A false discovery rate (FDR) of .05 was used to adjust for multiple comparisons. Replication of significant SNPs was conducted in an additional 228 participants treated with atenolol in PEAR. Results: Metabolomics analysis revealed 3 metabolites associated with glucose change that passed FDR adjustment. The top metabolite signal was beta-alanine. Increasing levels of baseline beta-alanine was significantly correlated with greater glucose increases after atenolol treatment (p=.0003). A total of 79 SNPs in 12 genes in the beta-alanine pathway were explored. SNP rs2669429 in DPYS was significantly associated with glucose change, after FDR adjustment. We found that G allele carriers had significantly greater glucose change compared with non-carriers (p=.0006; β=+2.76 mg/dL). This finding was successfully replicated in the other 228 patients who received atenolol (p=.04; β=+1.86 mg/dL). Conclusions: These results suggest that beta alanine and DPYS rs2669429 SNP might be an important predictor of atenolol-induced dysglycemia.

Abstract Acute renal depletion of sorting nexin 1 (SNX1) in mice results in blunted natriuretic response and hypertension due to impaired dopamine D 5 receptor (D 5 R) activity. We elucidated the molecular mechanisms for these phenotypes in Snx1 −/− mice. These mice had increased renal expressions of angiotensin II type 1 receptor (AT 1 R), NADPH oxidase (NOX) subunits, D 5 R, and NaCl cotransporter. Basal reactive oxygen species (ROS), NOX activity, and blood pressure (BP) were also higher in Snx1 ‐/‐ mice, which were normalized by apocynin, a drug that prevents NOX assembly. Renal proximal tubule (RPT) cells from hypertensive (HT) Euro‐American males had deficient SNX1 activity, impaired D 5 R endocytosis, and increased ROS compared with cells from normotensive (NT) Euro‐American males. siRNA‐mediated depletion of SNX1 in RPT cells from NT subjects led to a blunting of D 5 R agonist‐induced increase in cAMP production and decrease in Na + transport, effects that were normalized by over‐expression of SNX1. Among HT African‐Americans, three of the 12 single nucleotide polymorphisms interrogated for the SNX1 gene were associated with a decrease in systolic BP in response to hydrochlorothiazide (HCTZ). The results illustrate a new paradigm for the development of hypertension and imply that the trafficking protein SNX1 may be a crucial determinant for hypertension and response to antihypertensive therapy.

Abstractβ‐Blockers’ heart rate (HR)‐lowering effect is an important determinant of the effectiveness for this class of drugs, yet it is variable among β‐blocker–treated patients. To date, genetic studies have revealed several genetic signals associated with HR response to β‐blockers. However, these genetic signals have not been consistently replicated across multiple independent cohorts. Here we sought to use data from 3 hypertension clinical trials to validate single‐nucleotide polymorphisms (SNPs) previously associated with the HR response to β‐blockers. Using linear regression analysis, we investigated the effects of 6 SNPs in 3 genes, including ADRB1, ADRB2, and GNB3, relative to the HR response following β‐blocker used in the PEAR (n = 757), PEAR‐2 (n = 368), and INVEST (n = 1401) trials, adjusting for baseline HR, age, sex, and ancestry. Atenolol was used in PEAR and INVEST, and metoprolol was used in PEAR‐2. We found that rs1042714 and rs1042713 in ADRB2 were significantly associated with HR response to both β‐blockers in whites (rs1042714 C‐allele carriers, meta‐analysis β = –0.95 beats per minute [bpm], meta‐analysis P = 3×10–4; rs1042713 A‐allele carriers, meta‐analysis β = –1.15 bpm, meta‐analysis P = 2×10–3). In conclusion, the results of our analyses provide strong evidence to support the hypothesis that rs1042714 and rs1042713 in the ADRB2 gene are important predictors of HR response to cardioselective β‐blockade in hypertensive patient cohorts.

Abstract BACKGROUND Interindividual variability in blood pressure (BP) response to antihypertensives has been reported. Although plasma renin activity (PRA) is a potential biomarker for personalizing antihypertensive therapy in European American (EA) and African American (AA) hypertensives, clinical utility of PRA-guided prescribing is incompletely understood. METHODS Using systematic-phased approach, PRA’s clinical utility was assessed. After categorizing by baseline PRA, clinic systolic BP (SBP) responses to metoprolol and chlorthalidone were compared in 134 EAs and 102 AAs enrolled in the Pharmacogenomics Evaluation of Antihypertensive Responses-2 (PEAR-2) trial. Receiver operating characteristic (ROC) analysis was conducted in EAs. Data from PEAR-2 AAs were used to estimate an optimal PRA cut point using multivariable linear regression models. The derived cut point in AAs was tested in a meta-analysis of 2 independent AA cohorts, and its sensitivity and specificity were assessed. RESULTS EAs with PRA &lt; 0.65 ng/ml/hour had a greater decrease in SBP to chlorthalidone than metoprolol (by –15.9 mm Hg, adjusted P &lt; 0.0001), whereas those with PRA ≥ 0.65 ng/ml/hour had a greater decrease in SBP to metoprolol than chlorthalidone (by 3.3 mm Hg, adjusted P = 0.04). Area under ROC curve (0.69, P = 0.0001) showed that PRA can predict SBP response among EAs. However, we observed no association between PRA and SBP response in PEAR-2 AAs. Among independent AA cohorts, those with PRA ≥ 1.3 ng/ml/hour (PEAR-2-derived cut point) responded better to atenolol/candesartan than hydrochlorothiazide (meta-analysis P = 0.01). However, sensitivity of the derived cut point was 10%. CONCLUSIONS PRA at the previously established 0.60–0.65 ng/ml/hour cut point is an effective predictive biomarker of BP response in EAs. However, we were unable to identify PRA cut point that could be used to guide antihypertensive selection in AAs. TRIAL REGISTRATION NCT01203852, NCT00246519, NCT00005520.

Background Thiazide and thiazide‐like diuretics are first‐line medications for treating uncomplicated hypertension. However, their use has been associated with adverse metabolic events, including hyperglycemia and incident diabetes mellitus, with incompletely understood mechanisms. Our goal was to identify genomic variants associated with thiazide‐like diuretic/chlorthalidone‐induced glucose change. Methods and Results Genome‐wide analysis of glucose change after treatment with chlorthalidone was performed by race among the white (n=175) and black (n=135) participants from the PEAR ‐2 (Pharmacogenomic Evaluation of Antihypertensive Responses‐2). Single‐nucleotide polymorphisms with P <5×10 −8 were further prioritized using in silico analysis based on their expression quantitative trait loci function. Among blacks, an intronic single‐nucleotide polymorphism (rs9943291) in the HMGCS 2 was associated with increase in glucose levels following chlorthalidone treatment (ß=12.5; P =4.17×10 −8 ). G‐allele carriers of HMGCS 2 had higher glucose levels (glucose change=+16.29 mg/dL) post chlorthalidone treatment compared with noncarriers of G allele (glucose change=+2.80 mg/dL). This association was successfully replicated in an independent replication cohort of hydrochlorothiazide‐treated participants from the PEAR study (ß=5.54; P =0.023). A meta‐analysis of the 2 studies was performed by race in Meta‐Analysis Helper, where this single‐nucleotide polymorphism, rs9943291, was genome‐wide significant with a meta‐analysis P value of 3.71×10 −8 . HMGCS 2 , a part of the HMG ‐CoA synthase family, is important for ketogenesis and cholesterol synthesis pathways that are essential in glucose homeostasis. Conclusions These results suggest that HMGCS 2 is a promising candidate gene involved in chlorthalidone and Hydrochlorothiazide (HCTZ)‐induced glucose change. This may provide insights into the mechanisms involved in thiazide‐induced hyperglycemia that may ultimately facilitate personalized approaches to antihypertensive selection for hypertension treatment. Clinical Trial Registration URL : http://www.clinicaltrials.gov . Unique identifiers: NCT 00246519 and NCT 01203852.