Citations

Despite advances in pharmacogenomics that have led to innovative therapeutic strategies, substantial interindividual variability in drug efficacy and toxicity persists. Both endogenous and exogenous factors contribute to this unexplained variation, resulting in suboptimal treatment effectiveness, unnecessary healthcare costs, preventable mortality, and significant human suffering. The emerging field of exposomics is aimed at addressing the complex exposures that impact human health and offers an opportunity to interrogate how these exposures influence drug actions and outcomes. Here, we introduce pharmacoexposomics as a complementary framework that systematically quantifies the body's metabolic response to drug and environmental exposures using high-resolution mass spectrometry. We then present IndiPHARM (Individualized Pharmacology), a multi-institution, Advanced Research Projects Agency for Health–supported platform designed to integrate population-scale pharmacoexposomics with pharmacogenomics and clinical data to elucidate interindividual variability in drug response, identify adverse reactions, and inform tailored, effective treatment strategies that advance precision medicine.

Platelets are central to hemostasis and thrombosis. Excessive platelet activation contributes to arterial thrombotic events, including myocardial infarction, ischemic stroke, and complications of peripheral artery disease, whereas excessive platelet inhibition increases bleeding risk. Antiplatelet therapy remains a cornerstone of secondary prevention in atherosclerotic and thrombotic cardiovascular diseases, yet current treatment paradigms do not fully account for biological heterogeneity in platelet function, including differences related to sex, age, hormonal status, and disease context. This state-of-the-art review examines current antiplatelet strategies, fundamental mechanisms of signal transduction, clinical indications, and limitations of preclinical and clinical studies, with a focus on sex-specific considerations and opportunities for personalized antiplatelet therapy. Aspirin and P2Y 12 receptor antagonists are widely used for secondary prevention. However, women have been underrepresented in many pivotal clinical trials, limiting the precision of sex-specific estimates of efficacy and bleeding risk. In parallel, commonly used preclinical models often fail to recapitulate the physiological conditions in which platelets interact with the vasculature, leukocytes, and soluble factors. Emerging therapeutic approaches seek to refine platelet inhibition by targeting pathways that reduce thrombotic risk while preserving hemostasis. Advancing antiplatelet therapy will require integration of mechanistic platelet biology with diverse clinical trial populations, standardized platelet phenotyping, and disease-specific approaches that account for how platelet function is altered across health and vascular disease.

Breast biopsy markers have evolved from simple metallic devices to sophisticated multimaterial tools that have become essential for breast cancer care, demonstrating their clinical benefits in surgical planning, radiation targeting, and cost reduction.

Rotating magnetic fields (RMFs) are widely used to actuate untethered microrobots, yet delivering clinically relevant RMFs at human scale remains challenging. Here, we create a quantitative framework for human‐scale RMF delivery using robot‐arm‐mounted permanent magnets. It links anatomical separation, surface‐clearance constraints, magnet geometry, rotational inertia, and robot‐arm limitations to define a feasible RMF operating envelope. Magnet geometry is most important when target separation is comparable to magnet size; in this regime, diametrically magnetized cylinders provide the most efficient field delivery at constant magnet volume. At larger separations, field delivery converges toward dipole‐like scaling and becomes primarily governed by total magnet volume. Across the dominant microrobot operating envelope (5–30 mT, 1–30 Hz), required magnet mass ranges from sub‐kilogram values for limb and intracranial targets to more than 30 kg for deep thoracic access. As magnet size increases, rotational inertia rapidly emerges as the dominant mechanical constraint, with limb and intracranial targets remaining feasible whereas some thoracic scenarios exceed practical mechanical limits. We demonstrate a representative actuator designed within the proposed framework tested on a human‐scale vascular phantom under flow. This work provides a foundation for codesigning translational microrobots and their magnetic drive platforms tailored for relevant anatomical targets.

Abstract Anthracycline-induced cardiotoxicity (AIC) remains a dose-limiting complication of cancer treatment, often leading to irreversible heart failure, yet effective cardioprotective therapies remain limited. Graph-driven drug repurposing offers a rapid pathway to identify cardioprotective therapies, but different computational models often generate divergent candidate lists that exceed experimental validation capacity. We applied a novel "Union-then-Filter" framework to prioritize therapeutics for AIC using AI-based approaches and evaluated candidates in zebrafish models. Three state-of-the-art graph learning models (TxGNN, CompGCN, and RLR) were trained on the PrimeKG to generate an initial candidate pool. A Large Language Model (LLM) agent automated evidence synthesis across 4,286 PubMed abstracts for the top 50 drugs from each model. Candidates were stratified using dual decision criteria and filtered against existing clinical trials. The pipeline narrowed 150 initial candidates to four novel candidates: Cysteine, Dasatinib, Tranilast, and Tretinoin. Prospective in vivo validation was subsequently conducted using an adult AIC (aAIC) zebrafish model. Phenotypic screening via echocardiography revealed that Cysteine, Tranilast, and Tretinoin significantly restored ejection fraction (EF) (overall one-way ANOVA: \(\:F\left(\text{4,95}\right)=47.66,\:\:p<0.0001\); Tukey's post hoc: \(\:\:\:p<0.01\) for all three treatments vs. DOX+ control) and normalized cardiac remodeling markers in doxorubicin-treated subjects (Kruskal-Wallis: \(\:{\chi\:}^{2}\left(4\right)=17.21,\:\:p=0.0018\); Dunn’s post hoc: \(\:\:\:p<0.01\) for all three comparisons vs. DOX control). This study demonstrates the translational utility of an LLM-integrated AI pipeline for drug repurposing and highlights zebrafish AIC as an efficient bridge between in silico prediction and in vivo phenotypic validation in cardio-oncology.

Importance Thirty-day hospital readmission is used to guide postdischarge care; however, a single binary 30-day estimate does not distinguish when risk occurs from when intervention is most actionable, with implications for allocation of postdischarge resources. Objective To evaluate time-structured hospital readmission risk within 30 days after discharge and implications for clinical utility. Design, Setting, and Participants Retrospective prognostic study of adults with hospital discharges from 2017 to 2024 (from 19 hospitals within a US health system for the development cohort) and 2008 to 2019 (from Medical Information Mart for Intensive Care IV for the external cohort) with 30-day follow-up. Analyses were conducted from November 2025 through June 2026. Exposure Postdischarge prediction window, including landmark intervals of 0 to 7, 8 to 14, 15 to 21, and 22 to 30 days and cumulative horizons through days 7, 14, 21, and 30 after discharge. Main Outcomes and Measures First all-cause hospital readmission before death during each landmark interval and by each cumulative horizon. Risk estimates were derived using clinical information available at hospital discharge. Performance was assessed using discrimination, calibration, decision curves, and capacity-constrained analyses. Results Among 617 841 discharges in the development cohort, the median (IQR) patient age was 65 (50-76) years, and 315 220 discharges (51.0%) occurred among female patients. The 30-day readmission rate was 16.3% (95% CI, 16.2%-16.5%), with 68 797 readmissions (68.2%) occurring within 14 days. Cumulative-horizon analyses characterized readmission risk through days 7, 14, 21, and 30, whereas landmark interval-specific analyses showed lower positive predictive values (PPVs) beyond day 7. At 5% intervention capacity, landmark PPVs were 58.3% (95% CI, 57.4%-59.1%) for days 0 to 7, 16.0% (95% CI, 15.5%-16.6%) for days 8 to 14, 14.5% (95% CI, 13.9%-15.0%) for days 15 to 21, and 12.9% (95% CI, 12.4% to 13.5%) for days 22 to 30; numbers needed to prevent 1 readmission were 8.6, 31.3, 34.6, and 38.6, respectively. Conclusions and Relevance In this prognostic study of adult hospital discharges, hospital readmission risk varied over time and was not captured by a single binary 30-day estimate. These findings suggest that time-structured readmission modeling can characterize when readmission risk occurs, but additional work is needed before such models can reliably guide interventions during specific postdischarge intervals.

ABSTRACT Background Percutaneous left atrial appendage occlusion (LAAO) is an alternative strategy for stroke prevention in patients with nonvalvular atrial fibrillation (AF) with contraindications to long‐term oral anticoagulation therapy. However, evidence comparing the long‐term outcomes of LAAO with those of direct oral anticoagulants (DOAC) remains limited. Methods A systematic search was conducted across PubMed, Scopus, and the Cochrane databases through April 2026. Co‐primary outcomes were major adverse cardiovascular and cerebrovascular events (MACCE; composite of cardiac death, stroke, or systemic embolism) and net adverse clinical events (NACE; composite of MACCE and non‐procedure‐related bleeding). Secondary endpoints included individual components of co‐primary endpoints, ischemic and hemorrhagic stroke, all‐cause death, major or clinically significant bleeding (procedure + non‐procedure related), and stroke or systemic embolism. The random‐effects model was used to generate risk ratios (RRs) and 95% CIs. Results Four randomized controlled trials comprising 5890 AF patients and a median follow‐up of 3 years were included. There was no significant difference between LAAO and DOAC therapy in the risk of MACCE (RR 1.17; 95%CI 0.96–1.41; I 2 = 0%) or NACE (RR 0.89; 95%CI 0.65–1.24; I 2 = 85%). Both LAAO and DOAC therapy were similar in risk of cardiac death, stroke, systemic embolism, hemorrhagic stroke, ischemic stroke, all‐cause mortality, stroke or systemic embolism, and pericardial effusion. LAAO was associated with a lower risk of non‐procedure‐related bleeding (RR 0.60; 95%CI 0.47–0.76; I 2 = 59%). Conclusion LAAO demonstrated efficacy comparable to DOAC therapy in reducing the risk of MACCE, NACE, and ischemic events, while significantly reducing the risk of non‐procedure‐related bleeding. Our findings support a tailored approach to stroke prevention in AF, wherein LAAO serves as an effective alternative for patients with contraindications to DOAC therapy.

Financial toxicity is a significant but often overlooked consequence of cancer for adolescents and young adults (AYAs, 15–39 years old), who face limited savings, unstable employment, and insurance gaps that heighten financial stress. As part of a larger initiative to determine whether our cancer center can use an existing social drivers of health (SDOH) screening process to identify patients who may benefit from targeted supportive interventions, we sought to characterize financial strain, toxicity, and SDOH needs in AYAs treated in our cancer center and compare them to both younger and older cancer patients. This is a cross-sectional observational study at Mayo Clinic of patients with an oncology-related department visit between 8 September 2023–4 June 2024 (N = 81,731) who completed routine institutional electronic screenings for financial strain and SDOH. Survey items assessed financial strain, transportation barriers, and food insecurity; institutional billing data captured financial toxicity (unresolved balances, collection referrals, write-offs). AYAs (n = 5655) were more likely than non-AYAs to report medium/high financial strain, have unpaid debt, or have an unresolved balance, referral to a collection agency, or bad debt write-off, with nearly double the odds of medium/high financial strain compared to patients over 40 (OR 2.02, 95% CI 1.84–2.23, p < 0.001). Within AYAs, ages 26–32 bore the greatest burden: 24.2% in the medium/high-risk category, 9.4% with unresolved billing issues or collections, and 11.6% reporting food insecurity. These findings identify AYAs—particularly those 26–32—as a distinct high-risk group for financial toxicity and health-related social needs, underscoring the need for targeted interventions, tailored financial navigation, and policy efforts to reduce economic barriers to care. Future research should evaluate the effectiveness of these interventions in mitigating financial toxicity, addressing health-related social needs, and improving patient outcomes in this population.

Objective To compare the diagnostic accuracy of four available automated electronic medical record (EMR) retrieval methods, including a large language model (LLM)-assisted workflow, against manual chart adjudication for identifying cardiovascular events. Design Retrospective diagnostic accuracy study. Setting Three sites within a single US tertiary health system. Participants Two adult cohorts with previously adjudicated cardiovascular outcomes were included. Cohort 1 included 2258 patients treated with immune checkpoint inhibitors, and Cohort 2 included 1426 patients who underwent transcatheter aortic valve replacement. Primary and secondary outcome measures The reference standard was clinician manual chart adjudication. Outcomes included ischaemic stroke or transient ischaemic attack, myocardial infarction (MI), heart failure (HF) exacerbation or hospitalisation and a composite major adverse cardiovascular events (MACE) outcome. Automated retrieval methods included International Classification of Diseases (ICD) codes, primary diagnosis, problem list and a zero-shot LLM workflow. Area under the (receiver operating characteristic) curve (AUC), sensitivity, specificity and net reclassification improvement were assessed. Results In Cohort 1, the LLM achieved the highest AUC for stroke (0.920; 95% CI 0.881 to 0.958), MI (0.938; 95% CI 0.905 to 0.971) and composite MACE (0.880; 95% CI 0.854 to 0.907), whereas ICD-based retrieval had a higher AUC for HF (0.882; 95% CI 0.845 to 0.918 vs 0.873; 95% CI 0.831 to 0.914). In Cohort 2, the LLM achieved the highest AUC for all evaluated outcomes: stroke (0.915; 95% CI 0.862 to 0.968), MI (0.928; 95% CI 0.839 to 1.000), HF (0.844; 95% CI 0.803 to 0.884) and composite MACE (0.862; 95% CI 0.829 to 0.895). In Cohort 1, differences in AUC between the LLM and ICD methods were not statistically significant across outcomes, whereas in Cohort 2 the LLM showed significantly higher AUC for stroke and composite MACE. Conclusion In this multisite retrospective validation study, the LLM-assisted workflow showed strong but context-dependent performance for identifying cardiovascular events from the EMR. Performance varied by outcome and cohort, and ICD-based retrieval remained competitive for some use cases. These findings support a complementary role for LLM-assisted extraction in retrospective cardiovascular outcomes research.