Risk Model Research Articles

The correlation between uveal melanoma and iris nevus

PurposeTo investigate the impact of iris nevus on the incidence and prognosis of uveal melanoma (UM).MethodsA retrospective cohort study was conducted involving 1128 UM cases and 4356 healthy controls. Participants were categorized based on iris nevus presence and grade: grade 0 (no iris nevus), grade 1 (single iris nevus), grade 2 (multiple iris nevi), and grade 3 (partial or complete darkening of the iris). Propensity matching score method was employed to control for age and sex differences, while the χ 2 test was used to compare the existence rate and grade of iris nevus between groups. Univariate ANOVA evaluated differences among various iris nevus grades, the Kaplan–Meier method analyzed the prognosis of patients with different iris nevus grades, and multivariate Cox proportional risk regression analysis was conducted to evaluate the relationship between clinical data and prognosis.ResultsA total of 5484 subjects were analyzed. UM patients exhibited a higher prevalence and grade of iris nevus (all P < 0.001) after propensity matching. Patients with iris nevus in the affected eye did not show a worse prognosis (P = 0.414). However, those with partial or complete iris darkening or multiple nevi had a poorer prognosis compared to those with a single or no iris nevus (all P < 0.05). Iris nevus presence and grade in the healthy eye had no prognostic impact (P = 0.726 and P = 0.825, respectively). The multivariate COX proportional risk model showed that tumor diameter (P < 0.001), age (P = 0.020), and grade of iris nevus in the affected eye (P = 0.009) were independent risk factors for a worse prognosis. LSD analysis revealed that patients with partial or complete darkening of the iris had larger tumor diameters than those without iris nevus (P = 0.013), single nevus (P = 0.015), and multiple nevus (P = 0.023).DiscussionOur findings indicate a higher proportion and grade of iris nevus in UM patients compared to controls, and a worse prognosis for UM patients with higher-grade iris nevi in the affected eye.

The landscape in telomere related gene prognostic signature for survival and medication treatment effectiveness prediction in hepatocellular carcinoma

ObjectiveTelomeres, made of repetitive DNA sequences and shelterin complexes, which were found at the ends of chromosomes and had been extensively studied in cancer research. However, in hepatocellular carcinoma (HCC) was still relatively scarce. In this study, we investigated the correlation between telomerase-related genes (TRGs) and the prognosis and immunotherapy of HCC patients to enhance clinical outcomes.MethodsIn this work, TRGs were gathered using TelNet, while clinical information and gene expression data for HCC patients were retrieved from the Cancer Genome Atlas (TCGA) database. A risk prediction model based on TRGs was created using COX and Lasso regression analyses, with ROC curves used to assess prognostic efficacy. Univariate and multifactorial COX regression analyses were used to determine if the risk model had an independent impact on prognosis. Nomograms were created to enhance clinical usability, and calibration curves were used to assess predictive ability at various time points. The Tumor Immune Dysfunction and Exclusion (TIDE) score was used to analyze differences in immune infiltrating cells between risk groups. The study analyzed the relationship between risk ratings and drug treatment effectiveness using data from the CellMiner database. The hub gene was identified and its relationship to prognostic markers of HCC patients was examined. The expression of hub genes in immune cell subpopulations was also investigated by single-cell data.Results2093 TRGs were identified, with 949 showing significant differences in expression between HCC and paracancerous tissues. Seven risk genes were overexpressed in tumor tissues, leading to lower survival rates in high-risk patients. Risk model could independently predict the prognosis of HCC patients. Analysis of tumor immune infiltrating cells revealed significant differences in cell abundance between risk groups, with notable variations in immune subset enrichment between subgroups. Higher risk scores correlated with increased sensitivity to sorafenib, mitoxantrone, oxaliplatin, gemcitabine, and entinostat, while sensitivity decreased for vincristine, etc. CDCA8 was identified as a key gene in the Protein Interaction Network, while high expression associated with poorer overall survival, tumor proliferation and metastasis. The results of single-cell data analysis suggest that CDCA8 may promote the development of HCC by affecting T lymphocytes.ConclusionThe TRG-based risk model could predict HCC patient prognosis and closely linked to tumor immune environment, which could offer new possibilities for clinical treatment.

Efficacy, safety, and biomarker analysis of first-line immune checkpoint inhibitors with chemotherapy versus chemotherapy for advanced gastric cancer: a multicenter, retrospective cohort study

BackgroundRecent phase III randomized controlled trials have demonstrated that first-line immune checkpoint inhibitors (ICIs) improve prognosis in advanced HER-2-negative gastric cancer patients with programmed death ligand 1 (PD-L1) combined positive score (CPS) higher than 5. However, these findings are not confirmed in real-world settings, and the benefits in PD-L1 CPS < 5 patients remain controversial.MethodsIn this multicenter, retrospective cohort study, data from across thirteen medical centers were analyzed by inverse probability of treatment weighting for matching, alongside univariate and multivariate COX proportional hazard regression models. Genomic and transcriptomic analyses were conducted to identify efficacy prognostic models and resistance mechanisms.ResultsThis study included 573 patients with advanced gastric cancer, 265 treated with chemotherapy and 308 with ICIs plus chemotherapy. In the overall cohort and HER-2-negative patients, the combination therapy significantly improved progression-free survival and overall survival, without marked increases in severe adverse events. Notably, patients with PD-L1 CPS 1–4 showed significant overall survival prolongation and a trend towards improved progression-free survival with combination therapy. Patients with unknown PD-L1 status also benefitted from ICIs. SMARCA4 and BRCA2 mutations were more frequent in patients with responses, while CCNE1 and ZFHX3 alternation, alongside high “ABC transporters” signatures, were more common in non-responsive patients. A novel risk model, PGFIC, outperformed traditional biomarkers in predicting treatment outcomes.ConclusionsAdding ICIs to first-line treatment significantly prolongs survival in overall patients and in those with PD-L1 CPS 1–4 or unknown. This study also provides valuable insights into prognostic markers and resistance mechanisms, potentially guiding immunotherapy strategies.

Identification of a pyroptosis-related prognostic model for colorectal cancer and validation of the core gene SPTBN5

BackgroundPyroptosis, an emerging type of programmed cell death. The mechanisms of pyroptosis mainly include inflammasome-activated pyroptosis and non-inflammasome-activated pyroptosis. Multiple prognostic scoring systems that utilize pyroptosis-related gene expression have been validated as effective predictors of patient outcomes. But the relationship between pyroptosis and colorectal cancer remains unclear. This study has established a gene signature associated with pyroptosis to forecast the prognosis of CRC patients. MethodsAn analysis of 52 pyroptosis genes was conducted in both CRC and normal colorectal tissues, leading to the discovery of differentially expressed genes (DEGs). Core pyroptosis-related genes were identified using least absolute shrinkage and selection operator (LASSO) Cox regression to establish a prognostic risk score (PRS) for predicting CRC patient outcomes. The TCGA cohort was split into high-risk and low-risk groups based on the PRS, followed by Gene Ontology (GO) and KEGG pathway analyses. Additionally, differences in the enrichment scores of 16 immune cell types and the activity of 13 immune-related pathways were compared. The role of SPTBN5, a core pyroptosis-related gene, was validated through functional experiments on human colorectal adenocarcinoma cells (SW480).Results40 differentially expressed genes were identified from 52 pyroptosis genes. A risk model was subsequently developed using 25 core pyroptosis-related genes identified through LASSO Cox regression analysis, and this model was validated in GEO cohorts. GO and KEGG pathway analyses showed that the DEGs are predominantly associated with mineral absorption, thyroid hormone synthesis, and pancreatic secretion. Functional experiments demonstrated that down-regulation of SPTBN5 expression through transfection led to significant decreases in the proliferation, migration, and clonogenicity of SW480 cells.ConclusionThe PRS can identify high-risk CRC patient groups and predict patient prognosis. SPTBN5 may present a potential therapeutic target for CRC.

Prognostic feature based on androgen-responsive genes in bladder cancer and screening for potential targeted drugs

ObjectivesBladder cancer (BLCA) is a tumor that affects men more than women. The biological function and prognostic value of androgen-responsive genes (ARGs) in BLCA are currently unknown. To address this, we established an androgen signature to determine the prognosis of BLCA.MethodsSequencing data for BLCA from the TCGA and GEO datasets were used for research. The tumor microenvironment (TME) was measured using Cibersort and ssGSEA. Prognosis-related genes were identified and a risk score model was constructed using univariate Cox regression, LASSO regression, and multivariate Cox regression. Drug sensitivity analysis was performed using Genomics of drug sensitivity in cancer (GDSC). Real-time quantitative PCR was performed to assess the expression of representative genes in clinical samples.ResultsARGs (especially the CDK6, FADS1, PGM3, SCD, PTK2B, and TPD52) might regulate the progression of BLCA. The different expression patterns of ARGs may lead to different immune cell infiltration. The risk model indicates that patients with higher risk scores have a poorer prognosis, more stromal infiltration, and an enrichment of biological functions. Single-cell RNA analysis, bulk RNA data, and PCR analysis support the reliability of this risk model, and a nomogram was also established for clinical use. Drug prediction analysis showed that high-risk patients had a better response to fludarabine, AZD8186, and carmustine.ConclusionARGs played an important role in the progression, immune infiltration, and prognosis of BLCA. The ARGs model has high accuracy in predicting the prognosis of BLCA patients and provides more effective medication guidelines.

Construction and validation of a readmission risk prediction model for elderly patients with coronary heart disease

BackgroundTo investigate the risk factors for readmission of elderly patients with coronary artery disease, and to construct and validate a predictive model for readmission risk of elderly patients with coronary artery disease within 3 years by applying machine learning method.MethodsWe selected 575 elderly patients with CHD admitted to the Affiliated Lu’an Hospital of Anhui Medical University from January 2020 to January 2023. Based on whether patients were readmitted within 3 years, they were divided into two groups: those readmitted within 3 years (215 patients) and those not readmitted within 3 years (360 patients). Lasso regression and multivariate logistic regression were used to compare the predictive value of these models. XGBoost, LR, RF, KNN and DT algorithms were used to build prediction models for readmission risk. ROC curves and calibration plots were used to evaluate the prediction performance of the model. For external validation, 143 patients who were admitted between February and June 2023 from a different associated hospital in Lu'an City were also used.ResultsThe XGBoost model demonstrated the most accurate prediction performance out of the five machine learning techniques. Diabetes, Red blood cell distribution width (RDW), and Triglyceride glucose-body mass index (TyG-BMI), as determined by Lasso regression and multivariate logistic regression. Calibration plot analysis demonstrated that the XGBoost model maintained strong calibration performance across both training and testing datasets, with calibration curves closely aligning with the ideal curve. This alignment signifies a high level of concordance between predicted probabilities and observed event rates. Additionally, decision curve analysis highlighted that both decision trees and XGBoost models achieved higher net benefits within the majority of threshold ranges, emphasizing their significant potential in clinical decision-making processes. The XGBoost model's area under the ROC curve (AUC) reached 0.903, while the external validation dataset yielded an AUC of 0.891, further validating the model's predictive accuracy and its ability to generalize across different datasets.ConclusionTyG-BMI, RDW, and diabetes mellitus at the time of admission are the factors affecting readmission of elderly patients with coronary artery disease, and the model constructed based on the XGBoost algorithm for readmission risk prediction has good predictive efficacy, which can provide guidance for identifying high-risk patients and timely intervention strategies.

Predicting maternal risk level using machine learning models

BackgroundMaternal morbidity and mortality remain critical health concerns globally. As a result, reducing the maternal mortality ratio (MMR) is part of goal 3 in the global sustainable development goals (SDGs), and previously, it was an important indicator in the Millennium Development Goals (MDGs). Therefore, identifying high-risk groups during pregnancy is crucial for decision-makers and medical practitioners to mitigate mortality and morbidity. However, the availability of accurate predictive models for maternal mortality and maternal health risks is challenging. Compared with traditional predictive models, machine learning algorithms have emerged as promising predictive modelling methods providing accurate predictive models.MethodsThis work aims to explore the potential of machine learning (ML) algorithms in maternal risk level prediction using a nationwide maternal mortality dataset from Oman for the first time. A total of 402 maternal deaths from 1991 to 2023 in Oman were included in this study. We utilised principal component analysis (PCA) in the ML algorithms and compared them to the results of model performance without PCA. We employed and compared ten ML algorithms, including decision tree (DT), random forest (RF), K—Nearest Neighbors (KNN), Naïve Bayes (NB), Extreme Gradient Boosting (xgboost), Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QDA), Logistic Regression (LR), Support Vector Machine (SVM) and Artificial Neural Network (ANN). Different metrics, including, accuracy, sensitivity, precision, and the F1- score, were utilised to assess Model performance.ResultsThe results indicated that the RF model outperformed the other methods in predicting the risk level (low or high) with an accuracy of 75.2%, precision of 85.7% and F1- score of 73% after PCA was applied.ConclusionsWe applied several machine learning models to predict maternal risk levels for the first time using real data from Oman. RF outperformed the other algorithms in this classification problem. A reliable estimate of maternal risk level would facilitate intervention plans for medical practitioners to reduce maternal death.

Clinical characteristics and prediction model of re-positive nucleic acid tests among Omicron infections by machine learning: a real-world study of 35,488 cases

BackgroundDuring the Omicron BA.2 variant outbreak in Shanghai, China, from April to May 2022, PCR nucleic acid test re-positivity (TR) occurred frequently, yet the risk factors and predictive models for TR remain unclear. This study aims to identify the factors influencing Omicron TR and to develop machine learning models to predict TR risk. Accurately predicting re-positive patients is crucial for identifying high-risk individuals, optimizing resource allocation, and developing personalized treatment and management plans, thereby effectively controlling the spread of the epidemic, reducing community burden, and ensuring public health.MethodsA retrospective study was conducted among individuals infected with Omicron BA.2 variant from April 12 to May 25, 2022, in the largest Shanghai Fangcang shelter hospital. Five machine learning models were compared, including k-nearest-neighbors (KNN), logistic regression (logistic), bootstrap aggregation (bagging), error back-propagation (BP) neural network, and support vector machines (SVM), to select the best prediction model for the TR risk factors.ResultsA total of 35,488 cases were included in this real-world study. The TR and control groups comprised of 6,171 and 29,317 cases respectively, with a re-positive rate of 17.39%. Higher occurrence of TR was observed in young age, males, those with obvious symptoms, underlying diseases, and a low Ct value. The KNN model proved to be the best in predicting the prognosis in the overall evaluation (accuracy = 0.8198, recall = 0.8026, and AUC = 0.8110 in the test set).InterpretationHigher TR risk was found in infected cases who were underage or with underlying diseases; vaccine brand and inoculation status were not significantly associated with TR. KNN was the most effective machine learning model to predict TR occurrence in isolation.

Adjustment of the ACEF score by the triglyceride glucose index improves the prediction of clinical outcomes in patients undergoing percutaneous coronary intervention

BackgroundThe ACEF score (age, creatinine, and left ventricular ejection fraction) and the triglyceride-glucose (TyG) index have been identified as robust risk prediction models for adverse outcomes post-percutaneous coronary intervention (PCI) in coronary atherosclerotic heart disease (CHD) patients. This study aimed to assess whether incorporating the TyG index enhances the predictive ability of the ACEF score and enhances risk stratification for CHD patients undergoing PCI.MethodsThis observational cohort study enrolled 1248 patients diagnosed with CHD who underwent PCI at the Third People’s Hospital of Chengdu, China, between May 2018 and December 2022. Receiver operating characteristic (ROC) curves were employed to establish the optimal cutoff values for the TyG index and ACEF score. The primary endpoint event was defined as major adverse cardiac and cerebrovascular events (MACCEs). Cumulative survival curves were plotted using the Kaplan–Meier method. Univariable and multivariable Cox regression analyses were conducted to identify predictors of MACCEs. The predictive ability of the ACEF score and the TyG index was assessed using the area under the ROC curve, net reclassification improvement (NRI), and integrated discrimination improvement (IDI).ResultsOver a median follow-up period of 30.60 months, 214 patients experienced MACCEs. Kaplan-Meier survival analysis demonstrated a significant association between elevated ACEF score (> 1.27) and TyG index levels (> 8.82) with a heightened incidence of MACCEs (HR = 2.018, 95%CI 1.593–2.789; HR = 2.057, 95%CI 1.572–2.691; Log-Rank test, both P < 0.001). Multivariable Cox regression analysis revealed that even after adjusting for multiple confounders, both the ACEF score and TyG index remained as predictors of MACCEs (HR = 1.702, 95%CI 1.210–2.394, P = 0.002; HR = 1.575, 95%CI 1.296–1.916, P < 0.001). Additionally, the addition of the ACEF score and TyG index into the conventional risk model significantly improved the ability to predict MACCEs, as evidenced by the increase in the C-statistic value from 0.664 to 0.703, along with notable improvements in continuous NRI (0.391), categorical NRI (0.107), and IDI (0.033) (all P < 0.001).ConclusionThe combination of the TyG index and the ACEF score enhances the predictive ability for long-term MACCEs in patients with CHD undergoing PCI.

Identification of pyroptosis-associated miRNAs in the immunoscape and prognosis of hepatocellular carcinoma

BackgroundHepatocellular carcinoma is one of the most prevalent types of liver malignancy and poses a severe threat to global health. Despite recent improvements in therapeutic approaches, treatment options for patients with advanced or recurrent HCC are still limited.Materials and methodsOur study analyzed miRNA differential expression using data from hepatocellular carcinoma patients in the Cancer Genome Atlas. Pyroptosis-related genes were identified from gene cards. Differential expression of miRNAs was analyzed using DESeq2 and visualized using ggplot2 and pheatmap. A prognostic risk model for pyroptosis-associated miRNAs was constructed using LASSO regression and validated by principal component analysis, Kaplan-Meier survival and ROC curve analysis. We also performed gene and pathway enrichment analysis. Immune cell infiltration and function in HCC were assessed using single-sample genomic enrichment analysis, and correlations with immune cells and function were explored. Also, CCK-8 assay as well as migration and invasion assays were performed after knockdown of miR-6844.ResultsWe have established and validated a prognostic risk model based on ten DEmiRNAs, which is important for the survival of HCC patients. Significant changes in immune cell infiltration and immune function were also found in high-risk patients. It also demonstrated that knockdown of miR-6844 inhibited HCC cell proliferation, migration and invasion, highlighting its role in HCC progression.ConclusionOur study reveals the implications of pyroptosis-associated differential miRNAs on the prognosis of patients with hepatocellular carcinoma and provides a foundation for novel HCC therapies, especially immunotherapy.

Dual-specificity phosphatase 23 functions as a promising prognostic biomarker in non-small cell lung cancer.

The mechanical remodeling of tumor microenvironment is critical for non-small cell lung cancer (NSCLC) progression. Dual-specificity phosphatase 23 (DUSP23) has been previously identified as a mechano-responsive gene, but its role in NSCLC progression remains unknown. We aim to elucidate the clinical significance of DUSP23 in NSCLC progression. We analyzed the expression of DUSP23 in cancer using polyacrylamide hydrogels designed to mimic the stiffness of normal (soft; ~0.5kPa) and cancerous (stiff; ~40kPa) tissues. The prognostic significance of DUSP23 expression in patients was examined using public databases. Additionally, we conducted various cell-based assays and transcriptomic analyses in DUSP23-silenced NSCLC cell lines. A risk score prognosis model was constructed using univariate Cox regression and Kaplan-Meier analysis. Our findings show that DUSP23 is upregulated in stiff matrices and is highly associated with poor prognosis in patients with solid cancers such as NSCLC and breast cancer. Silencing of DUSP23 resulted in decreased cell proliferation and invasion. Transcriptomic profiling revealed that 182 genes were downregulated, and 230 genes upregulated following DUSP23-depletion. Notably, 182 downregulated genes were enriched in cancer-related pathways, including cell cycle progression and cytoskeleton organization. Through KEGG pathway analysis, we identified 11 cancer-related genes and developed a prognostic risk model. In this model, the high-risk group of NSCLC patients exhibited significantly shorter overall survival compared to low-risk group, based on public datasets. Our study demonstrates the clinical significance of DUSP23 as a prognostic marker in NSCLC and highlights its potential role of DUSP23 in promoting NSCLC progression.

Neurobehavioral disorders among children born to mothers exposed to illicit substances during pregnancy

BackgroundExposure to illicit substances during pregnancy may have long-term impacts on children’s neurodevelopment. This study explores subsequent risks for intellectual disability, autistic disorders, and attention deficit and hyperactivity disorders in children born to mothers exposed to illicit substances before or during pregnancy.MethodsWe identified women with illicit drug use by linking the police records from the “Substance Abuse Control Databases” and Taiwan Birth Registration and Birth Notification records from 2004 to 2014. Children whose mothers that had exposed to illicit substances during pregnancy identified from the police records were the “substance-exposed cohort.” A 1:1 ratio exact-matched comparison cohort based on child’s gender, child’s birth year, mother’s birth year, and child’s first use of the health insurance card, as well as a “propensity score (PS)-matched” comparison cohort of children born by substance-unexposed mothers, was established. Multivariate Cox regression analyses with competing risk models were performed.ResultsHigher incidences of intellectual disability (adjusted hazard ratio (aHR) = 2.41, 95% confidence interval (CI): 1.15–5.03) and attention deficit and hyperactivity disorder (ADHD) (aHR = 2.35, 95% CI: 1.63–3.28) were found in children prenatally exposed to illicit substances during pregnancy compared to exact-matched non-exposed cohorts. Adjusted risks of ADHD were significantly higher in mothers exposed to substances during pregnancy (aHR = 1.77 (1.42–2.21)) and before pregnancy (aHR = 1.43 (1.14–1.80)) compared to PS-matched unexposed cohorts after adjusting for covariates.ConclusionsThis is one of the first studies using large population-based data linked to criminal records to reveal increased risks of intellectual disability and ADHD in children with prenatal exposure to illicit substances compared to matched unexposed controls. Our results also highlight the importance of preventive measures and interventions for the well-being of both the mother and the child.

Association and predictive ability between significant perioperative cardiovascular adverse events and stress glucose rise in patients undergoing non-cardiac surgery

BackgroundThe predictive importance of the stress hyperglycemia ratio (SHR), which is composed of admission blood glucose (ABG) and glycated hemoglobin (HbA1c), has not been fully established in noncardiac surgery. This study aims to evaluate the association and predictive capability the SHR for major perioperative adverse cardiovascular events (MACEs) in noncardiac surgery patients.MethodsIndividuals who underwent noncardiac surgical procedures between 2011 and 2020, including both diabetic and non-diabetic patients, were identified in the perioperative medicine database (INSPIRE 1.1) and classified into tertiles based on their SHR. The connection between the SHR and the risk of MACEs was studied using Cox proportional hazards regression analysis, then restricted cubic spline (RCS) was employed to assess the association’s form. Additionally, the SHR’s incremental predictive utility for MACEs was assessed by the C-statistic, continuous net reclassification improvement (NRI), and integrated discrimination improvement (IDI), thereby quantifying the enhancement in predictive accuracy brought by incorporating the SHR into existing risk models. Feature importance and predictive models were generated utilizing the Boruta algorithm and machine learning approaches.ResultsA total of 5609 patients were enrolled. With an upwards shift in SHR vertices, the rate of perioperative MACEs and cardiac death event steadily rose. The RCS analysis for perioperative MACEs and cardiac death event both indicated J-shaped associations. Inflection points occurred at SHR = 0.81 for MACEs and SHR = 0.97 for cardiac death. The model’s fit improved significantly, with a continuous NRI of 0.067 (95% CI: 0.025–0.137, P < 0.001) and an IDI of 0.305 (95% CI: 0.155–0.430, P < 0.001). When SHR was added as a categorical variable (> 0.81), the C-statistic increased to 0.785 (95% CI: 0.756–0.814) with a ΔC-statistic of 0.035 (P = 0.009), a continuous NRI of 0.007 (95% CI: 0.000-0.021, P = 0.016), and an IDI of 0.076 (95% CI -0.024-0.142, P = 0.092). In the Boruta algorithm, variables identified as important features in the green area were incorporated into the machine learning models development.ConclusionsThe SHR was related with an increased risk of perioperative MACEs in patients following noncardiac surgery, highlighting its potential as a useful and reliable predictive tool for assessing the risk of perioperative MACEs.Graphical

Validation and modification of existing bleeding complications prediction models for percutaneous renal biopsy: a prospective study

Background Bleeding complications following percutaneous renal biopsy (PRB) are a significant clinical concern. This study aimed to validate and refine existing prediction models for post-biopsy bleeding to support more accurate clinical decision-making. Methods Clinical data from 471 PRB patients were examined in this prospective analysis. Ultrasounds were performed immediately and 6 h post-biopsy to identify perinephric hematomas. Patients exhibiting severe pain, a hemoglobin drop of &gt;10 g/L, symptomatic hypotension, hematuria within 7 days post-procedure underwent repeat ultrasound to assess for bleeding complications. Univariate and multivariable logistic regression analyses were conducted to identify factors associated with bleeding risk. The predictive performance of three kidney biopsy risk calculators (KBRC) was evaluated using the area under the receiver operating characteristic (AUROC) curve, net reclassification improvement (NRI), integrated discrimination improvement (IDI), and decision curve analysis (DCA) to determine clinical utility. Nomograms were developed for each model to facilitate clinical application. Results Univariate analysis identified body mass index (BMI), hemoglobin, and ultrasound findings as significant predictors of bleeding complications. In multivariable analysis, BMI, immediate ultrasound, and 6-h ultrasound data remained significant (p &lt; 0.05). The three models compared included: KBRC-5 (age, body mass index (BMI), platelet count, hemoglobin, kidney size), KBRC-5 with immediate ultrasound data (IKBRC), and KBRC-5 with 6-h hematoma size (SKBRC). The AUROC values for these models were 0.683, 0.786, and 0.867, respectively (p &lt; 0.001). NRI and IDI analyses demonstrated that adding immediate or 6-h ultrasound data significantly improved the risk reclassification ability of the KBRC-5 model (p &lt; 0.05). DCA indicated that IKBRC provided the highest net benefit for risk thresholds between 25% and 77%, while SKBRC was superior for thresholds between 10% and 95%. Nomograms were constructed for each model, allowing clinicians to estimate the probability of bleeding complications by summing scores for each predictor. Calibration curves showed good agreement between predicted and observed probabilities. Conclusion Incorporating real-time ultrasound data post-PRB significantly enhances the predictive accuracy and risk reclassification capability of bleeding risk models. These findings provide critical insights for guiding clinical management decisions in patients undergoing renal biopsy.

Novel models by machine learning to predict the risk of cardiac disease-specific death in young patients with breast cancer

BackgroundWith the tremendous leap of various adjuvant therapies, breast cancer (BC)-related deaths have decreased significantly. Increasing attention was focused on the effect of cardiac disease on BC survivors, while limited existing population-based studies lay emphasis on the young age population.MethodData of BC patients aged less than 50 years was collected from the SEER database. A competing risk model was introduced to analyze the effects of clinicopathology variables on the cardiac disease-specific death (CDSD) risks of these patients. Further, an XGBoost prediction model was constructed to predict the risk of CDSD. Prediction performance was assessed using the receiver operating characteristic (ROC) analysis, area under the POC curve (AUC) values, calibration curves, decision curves, and confusion matrix, and SHapley Additive exPlanations (SHAP) were used to interpret the models.ResultsOur competing risk analysis proved that young BC patients with older age, low household income, non-metropolitan residential environment, black race, unmarried status, HR + subtype, higher T stage (T2-4), receiving chemotherapy, and non-surgery are under higher risk of CDSD. Further, five machine learning models were constructed to predict the CDSD risks of young BC patients, among which the XGBoost models showed the highest AUC value (train set: AUC = 0.846; test set: AUC = 0.836). The confusion matrix of the XGBoost model demonstrated that the sensitivity, specificity, and correction were 0.81, 0.94, and 0.94 for the train set, and 0.82, 0.95, and 0.96 for the test set, respectively. The SHAP graph indicated that median household income, marital status, race, and age at diagnosis were the top four strongest predictors.ConclusionIndependent CDSD risk factors for young BC patients were identified, and machine-learning prognostic models were constructed to predict their CDSD risks. Our validation results indicated that the predicted probability of our XGBoost model agrees well with the actual CDSD risks, and it can help recognize high-risk populations and therefore develop effective cardioprotection strategies. Hopefully, our findings can support the growth of the new field of cardio-oncology.

Development and validation of a nomogram for predicting the risk of obstructive coronary artery disease in rheumatoid arthritis patients based on LDL-C, Th17 cells, and IL-17

ObjectiveThis study aims to develop and validate a nomogram model for predicting the risk of obstructive coronary artery disease (CAD) in patients with rheumatoid arthritis (RA), incorporating low-density lipoprotein cholesterol (LDL-C), Th17 cells, and interleukin (IL)-17 levels. The proposed model seeks to enable personalized cardiovascular risk assessment for RA patients, thereby optimizing clinical management strategies.MethodsA total of 120 patients with rheumatoid arthritis (RA) who were treated at the Second Hospital of Shanxi Medical University between January 2019 and September 2023 were enrolled in this study. Based on coronary angiography results, patients were categorized into the RA-obstructive CAD group and the RA-non-obstructive CAD group. Additionally, 53 healthy controls (HC group) were included. Clinical characteristics, laboratory parameters, peripheral blood lymphocyte subsets, and cytokine levels were collected for analysis. Univariate logistic regression was used to identify risk factors associated with RA-obstructive CAD. These variables were further refined using a random forest model for optimal selection. Finally, multivariate logistic regression analysis was performed with the selected variables to develop a nomogram model, which was subsequently validated to assess its performance.ResultsCompared with the RA-non-obstructive CAD group, the RA-obstructive CAD group demonstrated significantly elevated levels of immune cell subsets, such as Th17 cells, and cytokines, including IL-17, IL-2, and IL-4, along with a reduction in Treg cells. (2) In the training cohort, univariate and multivariate logistic regression analyses identified LDL-C (OR = 0.04, P &amp;lt; 0.001), Th17 cells (OR = 0.76, P = 0.005), and IL-17 (OR = 0.75, P = 0.001) as independent risk factors for obstructive CAD in RA patients. Subsequently, a predictive nomogram model for RA-obstructive CAD risk was developed based on these indicators, incorporating LDL-C, Th17 cells, and IL-17.ConclusionThis study developed a predictive nomogram for RA-obstructive CAD by combining traditional risk factors, such as LDL-C, with immune biomarkers Th17 and IL-17. The model demonstrated robust predictive accuracy, enabling more precise risk assessment of CAD in RA patients. It offers clinicians a valuable tool for advancing cardiovascular risk management in RA, underscoring its significant potential for clinical application.

Risk assessment and predictive modeling of suicide in multiple myeloma patients.

Despite advancements in treatment that have extended survival, multiple myeloma (MM) remains a distressing diagnosis with significant health impacts, including an elevated risk of suicide. This study aims to investigate suicide risk among MM patients and develop a predictive model to identify high-risk individuals. We analyzed 83,333 MM cases from the latest Surveillance, Epidemiology, and End Results (SEER) database (2001-2020) to identify suicide risk predictors and develop prediction nomograms. The cohort was randomly allocated into training and validation groups. Validation included assessing the consistency index (C-index), receiver operating characteristic (ROC) curve, and calibration curve. Among the cohort, 89 MM patients died by suicide, reflecting a significantly higher rate compared to the general US population (SMR = 2.186). Key risk factors included household income ≤ $50,000 (SMR = 3.82), male sex (SMR = 3.68), and age ≥ 80years at diagnosis (SMR = 3.05). Additional predictors were unmarried status, Black race, and diagnosis post-2007. The nomogram incorporating these factors demonstrated strong predictive accuracy in both training and validation groups. This study identified critical suicide risk factors in MM patients and developed a predictive nomogram that aids physicians in the early identification of at-risk individuals, facilitating more effective preventive measures. Utilizing the factors and predictive model for suicide risk among MM survivors allows for earlier identification and intervention, significantly enhancing their quality of life and psychological relief in the context of improved MM survival rates.

Economic Tremors From a Perfect Storm: The Ukrainian Crisis and Its Impact on Regional Stock Market Volatility

This study investigates the impact of the Russia-Ukraine war on stock market volatility in neighboring countries, analyzing five stock market indices (BET, BUX, WIG, SAX, and MOEX) over a three-year period encompassing one year before and two years after the conflict’s outbreak. Employing four volatility estimators (Close-to-Close, Parkinson, Garman-Klass, and Rogers-Satchell), this research examines the evolution of market volatility and inter-market correlations. Findings reveal a general increase in volatility across most indices following the war’s commencement, with the MOEX index experiencing the highest turbulence. The concept of a “proximity penalty” is partially supported, as geographical closeness to the conflict zone does not uniformly correspond to increased volatility. Also, findings show an initial strengthening of correlations between markets in the first year of the war, suggesting a “contagion effect.” However, this is followed by a weakening of correlations in the second year, indicating a potential “decoupling effect” as markets begin to respond more to local economic conditions. These results have significant implications for investors, policymakers, and risk managers, highlighting the need for dynamic portfolio management strategies, tailored policy responses, and flexible risk models that can adapt to changing market conditions during prolonged geopolitical crises. This study contributes to the existing literature by extending the temporal scope of analysis beyond the immediate aftermath of the war’s outbreak and providing insights into the “proximity effect” in the context of a major European conflict. The observed patterns of initial volatility spikes followed by varying degrees of persistence and changing correlation structures offer a nuanced picture of how geopolitical events impact financial markets over time, emphasizing the complex interplay between political events and financial market dynamics. JEL classification: G15, F51 Article History: Received: October 14, 2024; Reviewed: November 27, 2024; Accepted: December 2, 2024; Available online: December 17, 2024.

A new gene signature associated with pyroptosis for identifying high-risk myeloma patients

Multiple myeloma (MM) is a complicated hematologic malignancy of plasma cells. However, the existing stratification systems cannot accurately predict the prognosis of MM. This study aims to evaluate the role of pyroptosis in identifying high-risk MM patients. RNA expression profiles were obtained from the Cancer Genome Atlas-MMRF CoMMpass and GTEx databases, which were treated as MM cases and controls, respectively. By applying univariate Cox regression analysis and consensus clustering, 20 pyroptosis-related genes (PRGs) were initially identified to effectively stratify MM patients into two distinct subgroups. To identify prognostic gene signature, a stepwise LASSO regression analysis was conducted following by univariate and multivariate Cox regression analyses. We identified a set of signature genes &amp;ndash; CASP3, CHMP2A, CHMP3, CHMP6, GZMB, CASP8, NOD2, PLCG1, and FOXO3 &amp;ndash; that could significantly distinguish MM patients based on overall survival. We further identified that the risk score can serve as an independent prognostic indicator. The same prognostic model was also successfully constructed in the internal test cohort. Thus, a prognostic risk model for clinically predicting the survival rate of MM patients was established. Single-sample gene set enrichment analysis was employed to analyze immune infiltrating cells and immune-related pathways between two risk groups. Moreover, the mRNA expression levels of the prognostic risk signature genes were confirmed by quantitative reverse-transcription polymerase chain reaction in MM cells treated with the pyroptosis-inducing agent etoposide. In conclusion, we identified a 9-PRG signature that enables effective stratification of MM patients and serves as an independent prognostic marker. These findings underscore the need for further exploration of pyroptosis in MM therapy.

Metformin lowers risk of hearing loss and mortality in type 2 diabetes.

To assess the association between metformin use and the risk of sudden sensorineural hearing loss (SSNHL) in patients with Type 2 diabetes (T2D), a population at elevated risk for SSNHL. This cohort study utilized data from Taiwan's National Health Insurance Research Database, following T2D patients from 2008 to 202 database's baseline. Metformin use was defined as achieving ≥80% of the medication possession ratio (MPR) and ≥28 cumulative defined daily doses (cDDD) within three months. The control group included patients with ≥80% MPR from other antidiabetic agents, ensuring active treatment comparability. Propensity score matching was applied to balance covariates, while competing risk models accounted for mortality. Hazard ratios (HRs), incidence rates (IRs), and incidence rate ratios (IRRs) were calculated. Metformin users demonstrated a lower SSNHL incidence (IR: 11.48 per 10,000 person-years) compared to non-users (IR: 15.66 per 10,000 person-years), with an IRR of 0.73 (95% CI: 0.66-0.82; p < 0.0001). Adjusted HRs indicated a 27% reduction in SSNHL risk (HR: 0.73; 95% CI: 0.66-0.82). Higher cumulative doses (Q4: HR 0.36; 95% CI: 0.29-0.46) and daily doses ≥1 DDD (HR: 0.78; 95% CI: 0.69-0.87) were linked to further risk reductions. Metformin use was also associated with lower overall mortality. Metformin use is associated with a dose-dependent reduction in SSNHL risk and lower mortality in T2D patients. The rigorous definitions of metformin exposure and an actively treated comparator group emphasize these findings, suggesting metformin's potential role in SSNHL prevention and improved survival.