Social deprivation is associated with increased morbidity and mortality. However, there is a paucity of data on the effect of social deprivation on outcomes following percutaneous coronary intervention (PCI), stratified by P2Y12 inhibitor therapy. The objective of this study was to evaluate the association between social deprivation and the incidence of major adverse cardiovascular events (MACE) in P2Y12 inhibitor-treated post-PCI patients enrolled at the University of Florida. Composite social determinants of health (SDOH) measures, social deprivation index (SDI) and social vulnerability metric (SVM), at the zip code tabulation area-level were used to define social deprivation. The primary outcome was MACE (composite of cardiovascular death, myocardial infarction, stroke, or stent thrombosis). COX models adjusted for age, race, sex, and CYP2C19 genotype were used to evaluate the association between continuous SDOH scores and risk of MACE. A total of 1801 patients were included. Median SVM percentile was similar between clopidogrel-treated and alternative therapy patients [62 (43-89) and 59 (43-90); p=0.4], but clopidogrel-treated patients tended to have higher SDI scores [61 (43-81) and 60 (42-79); p=0.03]. Among clopidogrel-treated patients, both SDI (adj HR: 1.01; 95%CI: 1.00-1.02; p=0.04) and SVM (adj HR: 1.01; 95%CI: 1.00-1.02; p=0.01) were positively associated with MACE, with risk increasing linearly (e.g., each one-point increase in score resulted in a 1% increase in risk). This association was not observed in patients receiving alternative therapy. These data suggest social deprivation is associated with an increased MACE risk in clopidogrel-treated patients following PCI but may not be associated in those treated with prasugrel or ticagrelor.

Our previous work established that healthy pancreatic acinar cells obtained from deceased organ donors can be classified into subtypes that are used to characterize pancreatic ductal adenocarcinoma (PDAC). Bulk RNA sequencing was performed on the RNA isolated from 55 normal pancreatic acinar tissues. Subsequent spectral clustering and principal component analyses revealed that the RNA sequencing data split into two different groups arbitrarily described as Group 1 (n=31) and Group 2 (n=24). Group 2 aligns with the more aggressive classical and basal subtype (Moffitt) and Group 1 with the less aggressive exocrine (Collisson) subtype. The intent of this study is to determine a priori if a given normal pancreatic acinar tissue aligns with the less aggressive exocrine subtype or the aggressive classical/basal subtype. To advance this investigation further, we used a machine learning framework, MLSeq, to construct a predictive model that categorizes these subtypes and identifies significant gene predictors. Using the RNA-seq data from 55 cohort, count data were preprocessed and split into training (70%) and test (30%) subsets. Feature selection identified 1000 variable genes, refined to four primary predictive genes- KRT7, KRT19, TUBA1A, and MSN. Model performance was evaluated using accuracy, sensitivity, and specificity metrics. This MLSeq-based model achieved an accuracy of 82.35%, with a sensitivity of 72.73% and a specificity of 100%. This study emphasizes the integration of machine learning and transcriptome analysis to interpret the molecular mechanisms driving the acinar cell subtypes. Future research will focus on enhancing the models by expanding the number of samples, optimizing algorithms, and confirming predicting markers in external datasets.

Obstructive sleep apnea (OSA) often co-occurs with type-2 diabetes mellitus (T2DM). Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) and sodium-glucose cotransporter 2 inhibitors (SGLT2is) are both considered first-line options (usually with metformin) for T2DM. While both benefit OSA, evidence for GLP-1 RA in reducing OSA severity is more robust to date. We hypothesized comorbid OSA is associated with preferential treatment with GLP-1 RAs vs SGLT2is (as add-on to metformin) in patients with T2DM. We used Medicare claims (10/1/2016-12/31/2020) to identify patients aged ≥65 years with T2DM (≥2 outpatient or ≥1 inpatient ICD-10-CM claim for T2DM) who newly initiated a GLP-1 RA or SGLT2i. Index date was first date prescribed GLP-1 RA or SGLT2i. Patients were required to have ≥1 year continuous enrollment and ≥1 metformin prescription within 90 days before index date. OSA was defined as ≥2 outpatient or ≥1 inpatient OSA claim. We excluded those with CKD stage 4/5 or end-stage renal disease in 1-year baseline period. We used logistic regression to estimate odds of being prescribed GLP-1 RA vs. SGLT2i, comparing those with and without OSA. We controlled for age, race, sex, acute myocardial infarction, stroke, heart failure, obesity, CKD, insulin use, and index year. Of 60,548 patients with metformin-treated T2DM (mean age=72.6 [±5.33] years, 47.5% female), 45.2% were GLP-1 RA new users and 54.8% SGLT2i new users. Compared to those without OSA, patients with OSA had 17% increased odds of being prescribed GLP-1 RA vs. SGLT2i (aOR=1.17, 95% CI=1.11-1.22) after adjustment. Given potential of GLP-1 RAs to treat OSA, research should incorporate clinicians’ perspectives and prescribing practices when investigating the effectiveness and safety of these agents for OSA treatment.

Tobacco contains many carcinogens that increase cancer risk via forming DNA adducts that could lead to genetic mutations. Of those carcinogens, NNK and its major metabolite NNAL are tobacco-specific potent lung carcinogens. While methods have been developed to quantify NNK- and NNAL-specific DNA adducts to infer exposure and risk, such methods are less feasible to clinical translation due to low abundance of these adducts and limited access to target tissues. Previous efforts have shown that quantifying the hydrolyzed adducts from proteins can potentially serve as an indirect surrogate biomarker and even be measured in clinically available samples like plasma. The NNK-specific hydrolyzed adduct, HPB, has been quantified in protein samples from mice and human plasma while diol, the NNAL-specific hydrolyzed adduct, has not been characterized in physiologically relevant mouse models or clinical samples. We hypothesize that the bioactivation distributions between NNK and NNAL is condition dependent, including NNK dose and route of administration. To test this hypothesis, we quantified the hydrolyzed products in the lung, liver and serum samples in three commonly used mouse models for lung tumorigenesis with NNK administered via i.p. injection, drinking water, or via tobacco smoke exposure in A/J mice. Substantial differences were observed in the ratio of diol to HPB, suggesting that NNK and NNAL bioactivation are highly condition dependent. This reveals that some models may have limited physiological relevance in extrapolating NNK-induced lung carcinogenesis to humans. Altogether, this study provides insight to NNK metabolism and bioactivation as it relates to the clinical relevance of current mouse models, which will be validated in future clinical evaluations.

Opioids with low intrinsic efficacy offer potential for separating analgesia from side effects and can act as opioid-sparing agents. In this study, we identified a derivative of mitragynine pseudoindoxyl (MP), BP-194, using a semi-synthetic approach informed by the cryoEM structure of MP. BP1-94, characterized by extremely low intrinsic efficacy, exhibits properties intermediate between a traditional agonist and an antagonist. Despite its low efficacy, BP1-94 demonstrates mu-opioid receptor dependent antinociceptive effects in pain models (thermal, neuropathic, cancer, and inflammatory), with reduced respiratory depression, locomotor effects, and conditioned place preference (CPP) in mice. Intravenous BP1-94 was not self-administered in rats but reduced morphine self-administration and showed benefits in mitigating physical dependence and withdrawal symptoms. Pharmacokinetic studies revealed BP1-94 achieves superior brain penetration compared to MP. CryoEM structures of BP1-94 in active (G protein-bound) and inactive (Nb6-bound) states suggest a steric clash with Y771.39 in TM1 displaces BP1-94 from conserved residue D1493.32, disrupting an H-bond interaction with a basic nitrogen conserved in traditional mu agonists. This provides a mechanistic basis for modulating receptor efficacy by targeting TM1. Overall, BP1-94 represents a promising low-efficacy opioid candidate that may limit opioid use while maintaining effective analgesia.

Mitragyna speciosa (kratom) is a natural supplement that is popular for its purported pain-relieving and mood-enhancing effects. Reports exist of women using kratom during pregnancy resulting in newborns displaying symptoms of withdrawal. The most abundant psychoactive component in kratom is mitragynine (MTG). This study evaluated the pharmacokinetics of isolated MTG in pregnant Sprague-Dawley rats and withdrawal symptoms in rat neonates after in utero exposure. Pregnant rats were orally dosed twice a day with MTG (20 mg/kg).
For the pharmacokinetic study, samples of maternal plasma and whole fetal tissue were taken on gestational day 14 up to 8 hours post-final-dose. Concentration-time data for maternal plasma and fetal tissue were subjected to non-compartmental analysis using Phoenix WinNonlin. Metabolite-to-parent ratios were used to determine that mitragynine acid, a metabolite, doesn’t cross the placental barrier and that other active metabolites do not preferentially accumulate.
For the pharmacodynamic study, rats were dosed throughout gestation. After birth, neonate behavior was monitored for symptoms of withdrawal after administration of naltrexone. Rat neonates with prenatal exposure to MTG exhibited similar behaviors as neonates exposed to morphine.
These findings highlight the potential risks associated with kratom use during pregnancy, suggesting that it could lead to withdrawal symptoms in newborns, and underscore the need for clinical guidelines regarding its use in expectant mothers. Further research using these rat models can deepen understanding of kratom’s effects and inform safe practices.

Alzheimer’s disease and related dementia (ADRD) patients with hypertension (HTN) have a higher risk for adverse cardiovascular outcomes, compared to those with no HTN. However, the effect of different HTN control levels on these outcomes has not been explored. Validated HTN computable phenotype algorithms were applied to electronic health records (EHR)-based data of ADRD patients from the OneFlorida Data Trust to identify HTN control levels (well-controlled BP >80%, intermediate-controlled BP 80-50%, uncontrolled BP [controlled <50%], and apparent treatment-resistant HTN [aTRH]) over an 18-month exposure period after ADRD index date. The outcome was a composite outcome of myocardial infarction or stroke, captured after the exposure period using validated ICD-9/10 codes. adjusted Cox proportional hazard model was used for analysis. 9,430 patients were included in the study, with a mean age of 74 years, 56% female, and 75% White. The cohort included 3,523 patients without HTN, 1,295 with well-controlled HTN, 986 with intermediate-controlled HTN, 844 with uncontrolled HTN, and 2,782 with aTRH. Patients with uncontrolled HTN showed the highest risk for the composite outcome (hazard ratio [HR]: 1.97; 95% confidence interval [CI]: 1.54–2.51; p < 0.0001), followed by those with well-controlled HTN (HR: 1.53; 95% CI: 1.22–1.93; p = 0.0003), intermediate-controlled HTN (HR: 1.49; 95% CI: 1.16–1.90; p = 0.0018), and aTRH (HR: 1.30; 95% CI: 1.04–1.63; p = 0.0218), compared to ADRD patients without HTN. The results were consistent in subgroups and sensitivity analysis. Our results suggest that in older ADRD patients with HTN, the level of BP control, as opposed to the high-risk HTN phenotype of aTRH, may have a stronger influence on a patient’s cardiovascular outcome.

Carboxylesterase 1 (CES1) is an abundant hepatic drug metabolizing enzyme whose substrates include endogenous compounds and medications from a variety of therapeutic classes. CES1 is a serine hydrolase that hydrolyzes the cleavage of amides, esters, and thioesters. Concerns have been raised over botanical-drug interactions (BDIs) and their potential to impair enzymatic function resulting in therapeutic failures or adverse effects. BDIs have been extensively investigated with the cytochrome P450 family of enzymes; however, CES1 BDIs have remained largely uninvestigated. This study assessed the CES1 in vitro inhibitory effects of commonly used botanical extracts and their major constituents using an established incubation assay and LC-MS/MS analysis. Ashwagandha, cannabis plant material, saw palmetto, St. John’s wort, turmeric, and yohimbe all demonstrated significant reversible inhibition of CES1 metabolism. Cannabigerol, curcumin, lauric acid, linoleic acid, hypericin, hyperforin, kaempferol, and tetrahydrocannabinolic acid (THCA) demonstrated significant reversible inhibition of CES1. THCA was the most potent inhibitor of CES1 and displayed a complete mixed competitive-noncompetitive type inhibition. Cannabis plant material and turmeric all displayed inhibition potencies (Ki) of < 1 µM relative to their most abundant constituent. Additionally, curcumin and turmeric were found to be most stable in 50:50 methanol:water (1% formic acid) and least stable in a phosphate buffer solution (pH = 7.4). Using available pharmacokinetic data, it is anticipated that most extracts and constituents will impact the AUC of CES1 substrate medications. However, further in vitro and clinical studies must be conducted to fully elucidate BDI risk through CES1 impairment.

Carfilzomib(CFZ) has great treatment efficiency in multiple myeloma (MM) but has been associated with cardiovascular adverse events (CVAE).We aim to assess the causal inference of previously identified bile acids tauro & glycol ursodeoxycholic(TUDCA, GUDCA) with CFZ-CVAE using a Mendelian randomization (MR) approach.We analyzed data of 49 MM patients from the ‘Prospective study of cardiac events during proteasome inhibitor therapy’ study by conducting genotyping using Illumina’s multi-ethnic array & imputation by TOPMED panel.Metabolome-wide association studies (mGWAS) on TUDCA & GUDCA were performed in PLINK. To infer causality, two-sample MR by Inverse variance weighted method (IVW) was performed in MR Base using top SNPs (p<5*e-6) from the mGWAS as instrumental variables. Sensitivity analyses was conducted by MR-Egger, weighted median & weighted mode, while pleiotropy & heterogeneity were evaluated by MR-Egger intercept & Cochran’s Q test, respectively.For outcome analysis, we used one of the largest UK Biobank data (cardiovascular disease case= 177923) available in MR-Base.In mGWAS, 97 SNPs were linked to TUDCA & 11 SNPs to GUDCA at p<5e-6.IVW analysis showed significant associations for both TUDCA (p=0.03) & GUDCA (p=0.0038). Sensitivity analyses supported these findings, with no pleiotropy (p=0.087) or heterogeneity (p=0.68) observed for GUDCA. TUDCA, however, showed significant pleiotropy (p=0.0052) & heterogeneity (p=3.81e-13), suggesting involvement of alternative pathways. Key genes associated with TUDCA (MYOCD & ZFHX3) indicate potential cardiac involvement & possible MR assumption violations. MR analysis identified a significant causal link between GUDCA & CVAE. However, validation in larger cohorts is required to confirm this causal relevance.

Apratoxin A is a potent anticancer agent isolated from marine cyanobacteria, with a novel mechanism of action (MOA), inhibiting cotranslational translocation. It holds potential for treating cancers with active secretory pathways and high vascularization. This study aims to explore its MOA in colorectal cancer (CRC) and improve therapeutic efficacy and window by (a) using medicinal chemistry and (b) identifying targets for combination therapy. Apratoxins induced differential gene expression related to cancer processes, including the cell cycle, DNA repair, replication, secretory pathways, and growth factor signaling. Structure-activity relationship studies led to apratoxin S9, the most potent analogue with improved selectivity. In parallel, global RNAi screening of the druggable genome identified CRC vulnerabilities that increase susceptibility to apratoxin A. siRNAs that increased sensitivity to a low dose of apratoxin A were considered hits, with pathway analysis revealing key networks involved in angiogenesis and DNA repair. In vitro and in vivo results validate a combination therapy with PARP1 inhibitors, suggesting enhanced CRC treatment. To explore the potential synergy, data from medicinal chemistry and chemogenomic and transcriptomic analyses were integrated. Apratoxins inhibit RTK secretion via SEC61 blockade, indirectly suppressing PARP1 activation, creating a cooperative effect when combined with PARP1 inhibitors. Cotreatment of mice with apratoxin S9 with an FDA-approved PARP1 inhibitor strongly inhibited tumor growth in a xenograft model. Our study supports the development of an effective combination therapy with apratoxins and demonstrates a strategy to improve chemotherapeutic outcomes.

Duchenne muscular dystrophy (DMD) is a rare pediatric disease causing muscle weakness and disability, with bone fragility as a major complication. Quantitative magnetic resonance (qMR) biomarkers strongly correlate with functional endpoints in DMD clinical trials. This study aimed to extract clinically meaningful qMR biomarkers to quantify longitudinal changes in bone fragility during DMD progression using deep segmentation and voxel-wise feature extraction.
The qMR dataset included longitudinal data from 13 individuals with DMD (NCT01484678), with each scan consisting of 3 slices at the mid-femur. Cortical bone and marrow were segmented using a trained DynUNET, and PyRadiomics extracted a broad range of features, focusing on shape. Top shape features were selected based on Pearson correlation coefficients comparing features with MR spectroscopy fat fraction measures from the vastus lateralis (FFVL) muscle as an index of disease severity. Subgroup analysis examined changes in shape features from cortical bone and marrow.
The trained segmentation model achieved a Dice score of 97.49%, indicating high segmentation accuracy. As shown in Figure, shape features and cortical thickness from bone images gradually changed with DMD progression, reflecting FFVL. Subgroup analysis of 8 individuals revealed a significant difference in cortical thickness among steroid users (p=0.057, t-test).
This study demonstrates that AI-assisted radiomic feature extraction effectively identifies key features from MRI data, with the extracted features serving as qMR biomarkers for assessing bone fragility. Furthermore, the features of cortical bone correlated with FFVL, indicating that they capture worsening bone health with DMD disease progression.

Acute Myeloid Leukemia (AML) is an aggressive cancer with poor treatment outcomes, especially in pediatric patients, due to high relapse rates and drug resistance. The standard treatment involves cytarabine (Ara-C), daunorubicin, and etoposide (ADE), but resistance remains a challenge. To investigate genetic factors affecting drug resistance and treatment outcomes, we conducted a pharmacogenomic analysis of genetic variants in AML-related and drug-metabolizing genes in patients treated with ADE in the AML02 and AML08 clinical trials. We analyzed SNPs in genes related to ADE pharmacology and a myeloid gene panel from genome-wide Illumina 2.5 Omni microarray data of 400 patients. Associations were examined with clinical endpoints like minimal residual disease after induction 1 (MRD1), event-free survival (EFS), and overall survival (OS), adjusting for genetic ancestry. At a p-value of <0.01, 215 SNPs in 52 genes were linked to at least one clinical outcome. SNPs in solute carrier (SLC) transporters, such as SLC22A1, SLC28A1, and SLC28A3, were associated with increased MRD1 positivity, indicating poor prognosis, while SNPs in SLCO1B1 predicted better EFS and OS. Five SNPs in cytidine deaminase (CDA), which inactivate Ara-C, were linked to poorer EFS and OS and showed regulatory potential in chromatin states. SNP rs1048977 was associated with reduced CDA enzyme activity. Among myeloid genes, SNPs in RUNX1, CSF3R, and CDKN2A were linked to poor clinical outcomes. Dual mutations in CSF3R and CEBPA were associated with high relapse rates. In conclusion, this pharmacogenomic analysis identifies key variants influencing treatment outcomes, offering potential for personalized chemotherapy and better prognosis prediction in pediatric AML.