Combined Risk Score Research Articles

Abstract 4136501: Machine Learning Uncovers the Contribution of Rare Genetic Variants and Enhances Risk Prediction for Coronary Artery Disease in the Japanese Population

Background: Genome-wide association studies (GWASs) have enhanced our understanding of the genetic basis of coronary artery disease (CAD), and polygenic risk scores (PRSs) have facilitated the assessment of genetic risk. However, these methods predominantly focus on common variants due to statistical power, potentially leaving rare variants insufficiently analyzed and thus limiting the predictive performance of PRS. Methods: We conducted whole genome sequencing (WGS) of 1,752 Japanese early-onset myocardial infarction (MI) patients and 3,019 controls from Biobank Japan (BBJ). We performed case-control association studies including GWAS and gene-based tests, as well as a novel machine learning-based framework. In this framework, we developed a penalized regression model to predict the CAD status from genome-wide rare nonsynonymous variants. The model identified the minimal set of most distinguishing features (genes) and generated a rare variant-based risk score (RVS). The RVS was evaluated on an independent validation WGS cohort of 200 cases and 824 controls. We also derived a PRS based on CAD-GWAS (25,668 CAD cases vs 141,667 controls from BBJ) to compare the properties and performance between the RVS and PRS. Results: In the case-control studies, only two common variants in chromosome 12 were identified in GWAS, with no genes in gene-based analysis (SKAT-O), suggesting the challenges in rare variant analysis. On the other hand, our machine-learning framework identified 59 CAD-related genes, including LDLR , a causal gene of familial hypercholesteremia. Functional analyses revealed that various biological pathways, including lipid metabolism, immune system, and vessel development, are involved in CAD. For the genetic risk prediction, RVS significantly predicted CAD (area under the curve [AUC], 0.58; p=0.001, pseudo-R 2 , 0.051; p=7.92*10 -9 ). RVS was significantly associated with LDL cholesterol levels and coagulation function (Pearson’s r, 0.21; p=4.5*10 -5 and 0.10; p=0.03, respectively) and MI patients with high RVS (top 5%) showed higher cardiovascular mortality rate (p=0.03, log-rank test), highlighting the clinical importance of RVS. Finally, the combined risk score (CRS) of RVS and PRS significantly improved CAD prediction compared to PRS (AUC, 0.66 (CRS) vs 0.61 (PRS); p=0.007, pseudo-R 2 , 0.093 (CRS) vs 0.040 (PRS); p=0.0018, Figure ). Conclusions: Our machine learning framework successfully characterized rare variants and enhanced genetic risk prediction in CAD.

Prediction of Treatment Response Based on Nutritional Status and Tumor Immunity in Oropharyngeal Cancer Patients Treated With Chemoradiotherapy.

Radiotherapy (RT) for advanced oropharyngeal cancer (OPC) is effective, especially when combined with chemotherapy (CRT). However, its success can vary depending on factors, such as tumor stage, HPV infection (p16 status), and the patient's nutritional and immune status. This study examined the controlling nutritional status (CONUT) score and tumor immunity as predictive factors for treatment outcomes in OPC, aiming to develop a personalized risk score. A retrospective analysis was conducted on 84 patients with OPC treated with definitive RT or CRT, and survival outcomes were compared based on various factors, including BMI, CONUT score, CD8 expression, and HLA class II expression. We observed better overall survival (OS) rates in CD8-positive patients and those with higher HLA class II expression. The univariate analysis identified stage, p16 status, BMI, CONUT score, and CD8 expression as significantly associated with OS. In multivariate analysis, stage, BMI, and CONUT score remained significant predictors of OS. A risk scoring system was developed based on stage, p16 status, BMI, CONUT score, and CD8 expression. Patients were categorized into low-risk and high-risk groups, with significantly better survival in the low-risk group. A combined risk score incorporating clinical, nutritional, and immune factors can improve the prediction of treatment outcomes for OPC patients. This risk stratification may enable personalized treatment plans and improve ΟS rates.

Plasma proteomic and polygenic profiling improve risk stratification and personalized screening for colorectal cancer

This study aims to identify colorectal cancer (CRC)-related proteomic profiles and develop a prediction model for CRC onset by integrating proteomic profiles with genetic and non-genetic factors (QCancer-15) to improve the risk stratification and estimate of personalized initial screening age. Here, using a two-stage strategy, we prioritize 15 protein biomarkers as predictors to construct a protein risk score (ProS). The risk prediction model integrating proteomic profiles with polygenic risk score (PRS) and QCancer-15 risk score (QCancer-S) shows improved performance (C-statistic: 0.79 vs. 0.71, P = 4.94E–03 in training cohort; 0.75 vs 0.69, P = 5.49E–04 in validation cohort) and net benefit than QCancer-S alone. The combined model markedly stratifies the risk of CRC onset. Participants with high ProS, PRS, or combined risk score are proposed to start screening at age 46, 41, or before 40 years old. In this work, the integration of blood proteomics with PRS and QCancer-15 demonstrates improved performance for risk stratification and clinical implication for the derivation of risk-adapted starting ages of CRC screening, which may contribute to the decision-making process for CRC screening.

Improving prediction models of amyotrophic lateral sclerosis (ALS) using polygenic, pre-existing conditions, and survey-based risk scores in the UK Biobank.

Amyotrophic lateral sclerosis (ALS) causes profound impairments in neurological function, and a cure for this devastating disease remains elusive. This study aimed to identify pre-disposing genetic, phenotypic, and exposure-related factors for amyotrophic lateral sclerosis using multi-modal data and assess their joint predictive potential. Utilizing data from the UK (United Kingdom) Biobank, we analyzed an unrelated set of 292 ALS cases and 408,831 controls of European descent. Two polygenic risk scores (PRS) are constructed: "GWAS Hits PRS" and "PRS-CS," reflecting oligogenic and polygenic ALS risk profiles, respectively. Time-restricted phenome-wide association studies (PheWAS) were performed to identify pre-existing conditions increasing ALS risk, integrated into phenotypic risk scores (PheRS). A poly-exposure score ("PXS") captures the influence of environmental exposures measured through survey questionnaires. We evaluate the performance of these scores for predicting ALS incidence and stratifying risk, adjusting for baseline demographic covariates. Both PRSs modestly predicted ALS diagnosis but with increased predictive power when combined (covariate-adjusted receiver operating characteristic [AAUC] = 0.584 [0.525, 0.639]). PheRS incorporated diagnoses 1year before ALS onset (PheRS1) modestly discriminated cases from controls (AAUC = 0.515 [0.472, 0.564]). The "PXS" did not significantly predict ALS. However, a model incorporating PRSs and PheRS1 improved the prediction of ALS (AAUC = 0.604 [0.547, 0.667]), outperforming a model combining all risk scores. This combined risk score identified the top 10% of risk score distribution with a fourfold higher ALS risk (95% CI [2.04, 7.73]) versus those in the 40%-60% range. By leveraging UK Biobank data, our study uncovers pre-disposing ALS factors, highlighting the improved effectiveness of multi-factorial prediction models to identify individuals at highest risk for ALS.

The risk stratification and predictive performance of a new combined polygenic risk score for hepatocellular carcinoma.

Recent genome-wide association studies (GWASs) in liver diseases have generated some polygenic risk scores (PRSs), but their predictive effectiveness on hepatocellular carcinoma (HCC) risk assessment remains unclear. Here, we constructed a novel combined polygenic risk score and evaluated its increment to the well-established risk model. We used 15 HCC-associated genetic loci from two PRSs and FinnGen GWAS data to calculate a PRS-combined score and to fit the related PRS model in the UK Biobank cohort (N = 436,162). The PRS-combined score was further assessed for risk stratification for HCC integrating with the recommended clinical risk scores. The PRS-combined model achieved a better AUC (0.657) than that of PRS-HFC (0.637) and PRS-cirrhosis (0.645). The top 20% of the PRS-combined distribution had a 3.25 increased risk of HCC vs. the middle decile (45-55%). At the population level, the addition of PRS-combined to the CLivD score significantly increased the C-statistic (from 0.716 to 0.746) and provided a remarkable improvement in reclassification (NRI = 0.088) at the 10-year risk threshold of 0.2%. In clinic, additional assessment of PRS-combined would reclassify 34,647 intermediate-risk participants as high genetic risk, corresponding to an increase of 63.92% (62/97) of the HCC events classified at high risk using the Fibrosis-4 alone. The PRS may enhance HCC risk prediction effectiveness in the general population and refine risk stratification of the conventional clinical indicator.

Validation of a clinical breast cancer risk assessment tool combining a polygenic score for all ancestries with traditional risk factors

PurposeWe previously described a combined risk score (CRS) that integrates a multiple-ancestry polygenic risk score (MA-PRS) with the Tyrer-Cuzick (TC) model to assess breast cancer (BC) risk. Here, we present a longitudinal validation of CRS in a real-world cohort. MethodsThis study included 130,058 patients referred for hereditary cancer genetic testing and negative for germline pathogenic variants in BC-associated genes. Data were obtained by linking genetic test results to medical claims (median follow-up 12.1 months). CRS calibration was evaluated by the ratio of observed to expected BCs. ResultsThree hundred forty BCs were observed over 148,349 patient-years. CRS was well-calibrated and demonstrated superior calibration compared with TC in high-risk deciles. MA-PRS alone had greater discriminatory accuracy than TC, and CRS had approximately 2-fold greater discriminatory accuracy than MA-PRS or TC. Among those classified as high risk by TC, 32.6% were low risk by CRS, and of those classified as low risk by TC, 4.3% were high risk by CRS. In cases where CRS and TC classifications disagreed, CRS was more accurate in predicting incident BC. ConclusionCRS was well-calibrated and significantly improved BC risk stratification. Short-term follow-up suggests that clinical implementation of CRS should improve outcomes for patients of all ancestries through personalized risk-based screening and prevention.

Association of polygenic-based breast cancer risk prediction with patient management.

10527 Background: Guidelines recommend individuals with ≥20% lifetime risk of breast cancer (BC) undergo enhanced management, including annual screening mammography (SM) as early as age 30, annual breast MRI, and genetic counseling (GC). Lifetime BC risk can be estimated using a validated risk model, such as Tyrer-Cuzick (TC). A risk predictor that combines TC with a polygenic risk score (“combined risk score” or CRS) significantly improves risk prediction over TC alone. Little is known about how clinicians manage patients based on CRS. Here, we describe management following receipt of CRS results. Methods: De-identified administrative claims data from the Optum Labs Data Warehouse were linked with de-identified TC and CRS results originally provided to ordering clinicians. Patients were divided into four subgroups based on their lifetime risk predicted by CRS and by TC alone (high risk: ≥20%, average risk: <20%): A (CRS-high/TC-high), B (CRS-high/TC-avg), C (CRS-avg/TC-high), and D (CRS-avg/TC-avg). Patient management claims were compared 360 days before and after the CRS test date and differences evaluated with McNemar’s test. Differences in post-test management were analyzed using multivariable logistic regression with adjustment for clinical factors. Results: Before receiving risk results, 11.9%, 7.7%, 8.8%, and 5.0% of patients in A, B, C, and D received a SM before age 40, respectively; 3.4%, 2.6%, 2.3%, and 0.9% received an MRI, respectively; and 4.3%, 4.7%, 6.0%, and 3.6% received GC, respectively. After receiving risk results, SM in those under age 40 significantly increased 1.6-2.2-fold in A, B, and C; MRI significantly increased 4.7-5.6-fold in A, B, and C; and GC significantly increased 1.9-2.3-fold in A and B. SM, MRI, and GC did not increase in D. Prophylactic mastectomy occurred in only 0.17% of all patients after receiving results. After adjustment for confounders, those in A, B, and C were 3.8-5.2 times more likely to undergo SM; 11.6-23.1 times more likely to undergo MRI; and 2.0-2.9 times more likely to undergo GC, compared to those in Group D (all p<0.001). Conclusions: Patients with a ≥20% lifetime risk for BC were more likely to undergo enhanced management compared to those with a lifetime risk <20%, regardless of whether their risk was based on the CRS or on TC. These results suggest that clinicians recommended management aligned with guidelines for those with ≥20% lifetime risk, even when such risk was predicted by the CRS. [Table: see text]

Cancer-associated fibroblast-derived gene signature discriminates distinct prognoses by integrated single-cell and bulk RNA-seq analyses in breast cancer.

Cancer-associated fibroblasts (CAFs) are one of the most predominant cellular subpopulations in the tumor stroma and play an integral role in cancer occurrence and progression. However, the prognostic role of CAFs in breast cancer remains poorly understood. We identified a number of CAF-related biomarkers in breast cancer by combining single-cell and bulk RNA-seq analyses. Based on univariate Cox regression as well as Least Absolute Shrinkage and Selection Operator (LASSO) regression analysis, a novel CAF-associated prognostic model was developed. Breast cancer patients were grouped according to the median risk score and further analyzed for outcome, clinical characteristic, pathway activity, genomic feature, immune landscape, and drug sensitivity. A total of 341 CAF-related biomarkers were identified from single-cell and bulk RNA-seq analyses. We eventually screened eight candidate prognostic genes, including CERCAM, EMP1, SDC1, PRKG1, XG, TNN, WLS, and PDLIM4, and constructed the novel CAF-related prognostic model. Grouped by the median risk score, high-risk patients showed a significantly worse prognosis and exhibited distinct pathway activities such as uncontrolled cell cycle progression, angiogenesis, and activation of glycolysis. In addition, the combined risk score and tumor mutation burden significantly improved the ability to predict patient prognosis. Importantly, patients in the high-risk group had a higher infiltration of M2 macrophages and a lower infiltration of CD8+ T cells and activated NK cells. Finally, we calculated the IC50 for a range of anticancer drugs and personalized the treatment regimen for each patient. Integrating single-cell and bulk RNA-seq analyses, we identified a list of compositive CAF-associated biomarkers and developed a novel CAF-related prognostic model for breast cancer. This robust CAF-derived gene signature acts as an excellent predictor of patient outcomes and treatment responses in breast cancer.

Gefitinib metabolism-related lncRNAs for the prediction of prognosis, tumor microenvironment and drug sensitivity in lung adenocarcinoma

The complete compound of gefitinib is effective in the treatment of lung adenocarcinoma. However, the effect on lung adenocarcinoma (LUAD) during its catabolism has not yet been elucidated. We carried out this study to examine the predictive value of gefitinib metabolism-related long noncoding RNAs (GMLncs) in LUAD patients. To filter GMLncs and create a prognostic model, we employed Pearson correlation, Lasso, univariate Cox, and multivariate Cox analysis. We combined risk scores and clinical features to create nomograms for better application in clinical settings. According to the constructed prognostic model, we performed GO/KEGG and GSEA enrichment analysis, tumor immune microenvironment analysis, immune evasion and immunotherapy analysis, somatic cell mutation analysis, drug sensitivity analysis, IMvigor210 immunotherapy validation, stem cell index analysis and real-time quantitative PCR (RT-qPCR) analysis. We built a predictive model with 9 GMLncs, which showed good predictive performance in validation and training sets. The calibration curve demonstrated excellent agreement between the expected and observed survival rates, for which the predictive performance was better than that of the nomogram without a risk score. The metabolism of gefitinib is related to the cytochrome P450 pathway and lipid metabolism pathway, and may be one of the causes of gefitinib resistance, according to analyses from the Gene Set Enrichment Analysis (GSEA), Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG). Immunological evasion and immunotherapy analysis revealed that the likelihood of immune evasion increased with risk score. Tumor microenvironment analysis found most immune cells at higher concentrations in the low-risk group. Drug sensitivity analysis found 23 sensitive drugs. Twenty-one of these drugs exhibited heightened sensitivity in the high-risk group. RT-qPCR analysis validated the characteristics of 9 GMlncs. The predictive model and nomogram that we constructed have good application value in evaluating the prognosis of patients and guiding clinical treatment.

Use of polygenic and clinical risk scores in wide spectrum congenital heart malformations: a diagnostic and therapeutic perspective

Worldwide, congenital heart defects (CHDs) are a leading cause of mortality and morbidity. Globally, the annual prevalence rate of CHDs is about 1.5 million. It is one of the most common life-threatening congenital malformations, with a visible difference between high-income countries (HICs) and developing countries. This difference may be attributed to the diagnostic facilities and interventions.1 Moreover, the use of genetic and clinical risk scores is less studied in this field compared to other diagnostic and therapeutic approaches. Thus, we comprehensively review the available literature on adult and pediatric syndromic or non-syndromic CHDs and propose limitations and future directions in this field. Risk factors for this fatal disease include genetic and environmental factors.2,3 Risk stratification based on prenatal diagnosis with genetic counseling has shown better clinical outcomes. The modified World Health Organization (mWHO) guidelines are among the first proposed for risk stratification based on clinical scores.4 The combination of polygenic risk scores (PRS) and clinical risk evaluation has been adopted in various populations as better diagnostic and therapeutic approaches for various diseases, including coronary artery disease,5 diabetes,6 obesity,7 and cancers.8,9 However, unlike these diseases, polygenic risk scores are rarely used for the genetic diagnosis of CHD patients. The clinical utility of polygenic risk scores for CHDs is still in the early stages and requires validation studies in different cohorts. Incorporating PRS into the clinical setting may improve CHD diagnosis and therapeutic strategies.10 Recently, an initial attempt was made to determine the clinical utility of polygenic risk scores in a large dataset of the European cohort. Data from this study suggested a high potential for PRS in distinguishing between clinical phenotypes (mild, moderate, and severe) of congenital heart malformations.11 Furthermore, this hypothesis was tested in patients with transposition of the great arteries (TGA) to assess the utility of risk stratification after arterial switch operation. The cumulative risk score was computed for single nucleotide polymorphisms (SNPs) that reached the threshold statistical significance of p<1×10−5 in the initial analysis. Combined clinical and polygenic risk scores showed improved risk stratification in mild, moderate, and severe TGA phenotypes.12 Similarly, in another European cohort, the risk allele frequency was higher in the diseased group compared to the control group. However, statistical analysis suggested a non-significant association with p>0.05. Interestingly, combined risk scores constructed for selected SNPs showed a significant association with the disease with a p-value of 0.002. This evidence further supports the hypothesis that the cumulative effect of polymorphisms has better diagnostic and therapeutic efficacy than individual SNPs.13 There are certain limitations to its clinical use. Firstly, social stigma and lack of awareness regarding this disease are major hurdles. Secondly, there is a need to recruit a large number of samples from different populations to validate the clinical utility of PRS. Additionally, a lack of resources in developing countries also contributes to fatal outcomes. We recommend conducting large cohort studies in different ethnic groups to predict better risk associations, develop better therapeutic strategies, and formulate management guidelines.

Combined Serologic and Genetic Risk Score and Prognostication of Phospholipase A2 receptor-Associated Membranous Nephropathy.

The aim of this study was to test whether a combined risk score on the basis of genetic risk and serology can improve the prediction of kidney failure in phospholipase A2 receptor (PLA2R)-associated primary membranous nephropathy. We performed a retrospective analysis of 519 biopsy-proven PLA2R-associated primary membranous nephropathy patients with baseline eGFR ≥25 ml/min per 1.73 m 2 . The combined risk score was calculated by combining the genetic risk score with PLA2R ELISA antibody titers. The primary end point was kidney disease progression defined as a 50% reduction in eGFR or kidney failure. Cox proportional hazard regression analysis and C-statistics were applied to compare the performance of PLA2R antibody, genetic risk score, and combined risk score, as compared with clinical factors alone, in predicting primary outcomes. The median age was 56 years (range, 15-82 years); the male-to-female ratio was 1:0.6, the median eGFR at biopsy was 99 ml/min per 1.73 m 2 (range: 26-167 ml/min per 1.73 m 2 ), and the median proteinuria was 5.3 g/24 hours (range: 1.5-25.8 g/24 hours). During a median follow-up of 67 (5-200) months, 66 (13%) had kidney disease progression. In Cox proportional hazard regression models, PLA2R antibody titers, genetic risk score, and combined risk score were all individually associated with kidney disease progression with and without adjustments for age, sex, proteinuria, eGFR, and tubulointerstitial lesions. The best-performing clinical model to predict kidney disease progression included age, eGFR, proteinuria, serum albumin, diabetes, and tubulointerstitial lesions (C-statistic 0.76 [0.69-0.82], adjusted R 2 0.51). Although the addition of PLA2R antibody titer improved the performance of this model (C-statistic: 0.78 [0.72-0.84], adjusted R 2 0.61), replacing PLA2R antibody with the combined risk score improved the model further (C-statistic: 0.82 [0.77-0.87], adjusted R 2 0.69, difference of C-statistics with clinical model=0.06 [0.03-0.10], P < 0.001; difference of C-statistics with clinical-serologic model=0.04 [0.01-0.06], P < 0.001). In patients with PLA2R-associated membranous nephropathy, the combined risk score incorporating inherited risk alleles and PLA2R antibody enhanced the prediction of kidney disease progression compared with PLA2R serology and clinical factors alone.

A simple clinical risk score (ABCDMP) for predicting mortality in patients with AECOPD and cardiovascular diseases

BackgroundThe morbidity and mortality among hospital inpatients with AECOPD and CVDs remains unacceptably high. Currently, no risk score for predicting mortality has been specifically developed in patients with AECOPD and CVDs. We therefore aimed to derive and validate a simple clinical risk score to assess individuals’ risk of poor prognosis.Study design and methodsWe evaluated inpatients with AECOPD and CVDs in a prospective, noninterventional, multicenter cohort study. We used multivariable logistic regression analysis to identify the independent prognostic risk factors and created a risk score model according to patients’ data from a derivation cohort. Discrimination was evaluated by the area under the receiver-operating characteristic curve (AUC), and calibration was assessed by the Hosmer–Lemeshow goodness-of-fit test. The model was validated and compared with the BAP-65, CURB-65, DECAF and NIVO models in a validation cohort.ResultsWe derived a combined risk score, the ABCDMP score, that included the following variables: age > 75 years, BUN > 7 mmol/L, consolidation, diastolic blood pressure ≤ 60 mmHg, mental status altered, and pulse > 109 beats/min. Discrimination (AUC 0.847, 95% CI, 0.805–0.890) and calibration (Hosmer‒Lemeshow statistic, P = 0.142) were good in the derivation cohort and similar in the validation cohort (AUC 0.811, 95% CI, 0.755–0.868). The ABCDMP score had significantly better predictivity for in-hospital mortality than the BAP-65, CURB-65, DECAF, and NIVO scores (all P < 0.001). Additionally, the new score also had moderate predictive performance for 3-year mortality and can be used to stratify patients into different management groups.ConclusionsThe ABCDMP risk score could help predict mortality in AECOPD and CVDs patients and guide further clinical research on risk-based treatment.Clinical trial registrationChinese Clinical Trail Registry NO.:ChiCTR2100044625; URL: http://www.chictr.org.cn/showproj.aspx?proj=121626.

Development and Internal Validation of a Clinical and Genetic Risk Score for Rheumatoid Arthritis-Associated Interstitial Lung Disease.

Although clinical and genetic risk factors have been identified for rheumatoid arthritis-associated interstitial lung disease (RA-ILD), there are no current tools allowing for risk stratification. We sought to develop and validate an ILD risk model in a large, multicentre, prospective RA cohort. Participants in the Veterans Affairs RA (VARA) registry were genotyped for 12 single nucleotide polymorphisms (SNPs) associated with idiopathic pulmonary fibrosis. ILD was validated through systematic record review. A genetic risk score (GRS) was computed from minor alleles weighted by effect size with ILD, using backward selection. The GRS was combined with clinical risk factors within a logistic regression model. Internal validation was completed using bootstrapping, and model performance was assessed by the area under the receiver operating curve (AUC). Of 2,386 participants (89% male, mean age 69.5 years), 9.4% had ILD. Following backward selection, five SNPs contributed to the GRS. The GRS and clinical factors outperformed clinical factors alone in discriminating ILD (AUC 0.675 vs 0.635, p< 0.001). The shrinkage-corrected performance for combined and clinical-only models was 0.667 (95% CI 0.628, 0.712) and 0.623 (95% CI 0.584, 0.651), respectively. Twenty percent of the cohort had a combined risk score below a cut-point with >90% sensitivity. A clinical and genetic risk model discriminated ILD in a large, multicentre RA cohort better than a clinical-only model, excluding 20% of the cohort from low-yield testing. These results demonstrate the potential utility of a GRS in RA-ILD and support further investigation into individualized risk stratification and screening.

Associations of combined polygenic risk score and glycemic status with atrial fibrillation, coronary artery disease and ischemic stroke

BackgroundIt is unknown whether high hemoglobin A1c (HbA1c) is associated with increases in the risk of cardiovascular disease among individuals with elevated genetic susceptibility. We aimed to investigate the association between HbA1c and atrial fibrillation (AF), coronary artery disease (CAD), and ischemic stroke according to the polygenic risk score (PRS).MethodsThe UK Biobank cohort included 502,442 participants aged 40–70 years who were recruited from 22 assessment centers across the United Kingdom from 2006 to 2010. This study included 305,605 unrelated individuals with available PRS and assessed new-onset AF, CAD, and ischemic stroke. The participants were divided into tertiles based on the validated PRS for each outcome. Within each PRS tertiles, the risks of incident events associated with HbA1c levels were investigated and compared with HbA1c < 5.7% and low PRS. Data were analyzed from November 2022 to May 2023.ResultsOf 305,605 individuals, 161,605 (52.9%) were female, and the mean (SD) age was 56.6 (8.1) years. During a median follow-up of 11.9 (interquartile range 11.1–12.6) years, the incidences of AF, CAD, and ischemic stroke were 4.6, 2.9 and 1.1 per 100 person-years, respectively. Compared to individuals with HbA1c < 5.7% and low PRS, individuals with HbA1c ≥ 6.5% and high PRS had a 2.67-times higher risk for AF (hazard ratio [HR], 2.67; 95% confidence interval (CI), 2.43–2.94), 5.71-times higher risk for CAD (HR, 5.71; 95% CI, 5.14–6.33) and 2.94-times higher risk for ischemic stroke (HR, 2.94; 95% CI, 2.47–3.50). In the restricted cubic spline models, while a U-shaped trend was observed between HbA1c and the risk of AF, dose-dependent increases were observed between HbA1c and the risk of CAD and ischemic stroke regardless PRS tertile.ConclusionsOur results suggest that the nature of the dose-dependent relationship between HbA1c levels and cardiovascular disease in individuals with different PRS is outcome-specific. This adds to the evidence that PRS may play a role together with glycemic status in the development of cardiovascular disease.

Using Multi-Modal Electronic Health Record Data for the Development and Validation of Risk Prediction Models for Long COVID Using the Super Learner Algorithm

Background: Post-Acute Sequelae of COVID-19 (PASC) have emerged as a global public health and healthcare challenge. This study aimed to uncover predictive factors for PASC from multi-modal data to develop a predictive model for PASC diagnoses. Methods: We analyzed electronic health records from 92,301 COVID-19 patients, covering medical phenotypes, medications, and lab results. We used a Super Learner-based prediction approach to identify predictive factors. We integrated the model outputs into individual and composite risk scores and evaluated their predictive performance. Results: Our analysis identified several factors predictive of diagnoses of PASC, including being overweight/obese and the use of HMG CoA reductase inhibitors prior to COVID-19 infection, and respiratory system symptoms during COVID-19 infection. We developed a composite risk score with a moderate discriminatory ability for PASC (covariate-adjusted AUC (95% confidence interval): 0.66 (0.63, 0.69)) by combining the risk scores based on phenotype and medication records. The combined risk score could identify 10% of individuals with a 2.2-fold increased risk for PASC. Conclusions: We identified several factors predictive of diagnoses of PASC and integrated the information into a composite risk score for PASC prediction, which could contribute to the identification of individuals at higher risk for PASC and inform preventive efforts.

Genomic Landscape of Ccus Compared to MDS Indicates a Potential Applicability of the IPSS-M

Background: The 5 th edition of the WHO classification newly included myeloid precursor lesions introducing CCUS as an entity. CCUS is defined as clonal hematopoiesis (CH) in the presence of unexplained, persistent cytopenia requiring detection of either somatic mutation in certain CH associated genes or clonal chromosomal abnormalities. Aim: Characterize a large CCUS cohort with respect to cytogenetics and molecular genetics and compare the findings to an MDS cohort to evaluate differences in the genomic landscape. Methods: The CCUS cohort comprised 222 cases (median age: 76 y; female: 33%), the MDS cohort 698 cases (median age: 73 y; female: 43%). The diagnoses were established following WHO 2022. All samples were analyzed by cytomorphology, cytogenetics, targeted NGS panel (median coverage 1500x) and whole genome (median coverage 100x) sequencing. Mutation (MUT) status of 59 genes associated with myeloid malignancies were analyzed in detail (54 CH associated; 5 relevant for IPSS-M). Results: Significant differences in age, gender, BM blasts, blood parameters and number and types of cytopenias were detected. CCUS patients (pts) were older, had more WBC, less PLT, harbored a higher HB, less BM blasts and were even more predominantly male compared to MDS pts (all p&amp;lt;0.05). Within the CCUS cohort 29 pts (13%) showed cytogenetic abnormalities (CA) only (thereof 69% loss of chromosome Y (Y-loss)), while the remaining 193 pts (87%) showed at least one MUT (VAF ≥2%) in any of the 54 CH associated genes (thereof 84% MUT only; 16% MUT+CA). Regarding cytogenetics 162 CCUS pts (73%) harbored normal karyotypes while CA were detected in 27% (MDS: 43%; p&amp;lt;0.001) with Y-loss being most frequent (39/60; 65%). Y-loss was more frequently found in CCUS than in MDS (17% vs. 5%; p&amp;lt;0.001), while complex karyotypes and del(5q) were more frequent in MDS (11% and 16% vs. 1% and 0%; both p&amp;lt;0.001). Based on WGS data additional pts with Y-loss (n=7; confirmed by FISH) and recurrent CN-LOH (4q: n=4; 7q: n=5) were observed. With respect to somatic MUTs, 356 MUTs were found within the entire CCUS cohort (in 28 CH genes, 3 non-CH genes). Overall, CCUS pts showed less MUTs (median: 1 [0-6]) compared to MDS (median: 2 [0-12]; p&amp;lt;0.001). CCUS pts most frequently harbored MUTs in DNMT3A (32%), TET2 (28%) and ASXL1 (14%) followed by MUTs in splicing genes and TP53 (Fig. 1A). MUTs in DNMT3A and PPM1D were more frequent in CCUS than MDS while MUTs in ASXL1, TP53, SF3B1, STAG2, RUNX1, NRAS, CUX1 were less frequent in CCUS (each p&amp;lt;0.05). Notably, biallelic TP53 MUTs were less frequent in CCUS compared to MDS ( TP53bi of mutated: 14% vs. 55% p=0.008). Within the 15 most frequently mutated genes in CCUS, MUTs in 6 genes ( DNMT3A, SF3B1, TP53, CBL, STAG2, PPM1D) showed a median VAF &amp;lt;10%. Nine of the top 15 genes showed significantly lower median VAFs than in MDS ( DNMT3A, TET2, ASXL1, SRSF2, U2AF1, SF3B1, TP53, PPM1D, IDH2; each p&amp;lt;0.05). Thus, the frequency and VAF of ASXL1, TP53, SF3B1 MUTs were significantly lower in CCUS than in MDS, while DNMT3A and PPM1D MUTs were significantly more often detected in CCUS but at a lower VAF than in MDS. Combining cytogenetics and mutational analysis we calculated the IPSS-M, a patient-specific risk score for MDS resulting in six risk categories, and detected a clear skewing towards low risk categories in CCUS (Fig. 1B). The frequency of categories very low (VL) and low (L) were significantly higher in CCUS than in MDS (30% vs. 15%; 50% vs. 40%; both p&amp;lt;0.013), while high and very high categories were more rare (1% vs. 13%; 1% vs. 14%; both p&amp;lt;0.001). Although CCUS follow-up data were short (median: 1.5 y), no differences in overall survival were observed between CCUS VL+L and MDS VL+L, while within CCUS survival differed between VL+L and the remaining cases ( p=0.05). Conclusions: Our data confirm the comparable mutational spectrum between CCUS and MDS but clearly show major differences in the frequency and VAF of distinct gene mutations. These biological differences hint towards different subgroups within CCUS with cases closer to MDS than others. A combined risk score for MDS and CCUS would reflect this continuous spectrum and has the potential to derive an objective risk assessment irrespective of observer-dependent grading of dysplasia. Our study indicates plausible short term results for CCUS patients stratified according to the IPSS-M. However, the definite applicability of the IPSS-M needs to be confirmed in larger CCUS studies with longer follow-up.

Integration of Biomarker Polygenic Risk Score Improves Prediction of Coronary Heart Disease

There are several established biomarkers for coronary heart disease (CHD), including blood pressure, cholesterol, and lipoproteins. It is of high interest to determine how a combined polygenic risk score (PRS) of CHD-associated biomarkers (BioPRS) can further improve genetic prediction of CHD. We developed CHDBioPRS, combining BioPRS with PRS of CHD in the UK Biobank and tested it on FinnGen. We found that BioPRS was clearly predictive of CHD and that CHDBioPRS improved the standard CHD PRS. The largest effect was observed with early onset cases in FinnGen, with HRs above 2 per standard deviation of CHDBioPRS.

External Validation of a Urinary Biomarker Risk Score for the Prediction of Steroid Responsiveness in Adults With Nephrotic Syndrome

In idiopathic nephrotic syndrome, response to corticosteroids remains the best indicator of prognosis. Noninvasive markers to predict a patient's response to steroids would allow improved prognostication and a more personalized approach to management. We have previously derived a urinary biomarker risk score which can differentiate steroid sensitive nephrotic syndrome (SSNS) from steroid resistant nephrotic syndrome (SRNS) in children. The goal of this study was to validate this previously derived biomarker risk score in a cohort of steroid-naïve adult patients, to determine whether the panel could be used to predict steroid responsiveness at the time of initial diagnosis. In this external validation study, clinical data, and urinary specimens (obtained before initiation of steroid treatment) from adult patients were used in the Nephrotic Syndrome Study Network (NEPTUNE) cohort. A panel of 5 previously identified and validated urinary biomarkers, including neutrophil gelatinase-associated lipocalin (NGAL), vitamin D binding protein (VDBP), Fetuin-A (FetA), Transthyretin (TTR), and alpha-1 acid glycoprotein 2 (AGP2) was measured. A summary risk score for steroid resistance was calculated based on biomarker concentrations. Receiver operating characteristic curves were created for each log-transformed biomarker concentration and for the individual and combined biomarker risk score. The urine biomarker risk score predicted development of steroid resistance, with optimal sensitivity and specificity of 0.74, and area under the receiver operating characteristic curve (AUC) of 0.79 using both absolute and creatinine-corrected concentrations. This study validates the previously derived urinary biomarker risk score to predict steroid resistance in adult patients with nephrotic syndrome at initial diagnosis.

Development of a novel, clinically relevant anoikis-related gene signature to forecast prognosis in patients with prostate cancer.

Introduction: Anoikis is a specific form of programmed cell death and is related to prostate cancer (PC) metastasis. This study aimed to develop a reliable anoikis-related gene signature to accurately forecast PC prognosis. Methods: Based on anoikis-related genes and The Cancer Genome Atlas (TCGA) data, anoikis-related molecular subtypes were identified, and their differences in disease-free survival (DFS), stemness, clinical features, and immune infiltration patterns were compared. Differential expression analysis of the two subtypes and weighted gene co-expression network analysis (WGCNA) were employed to identify clinically relevant anoikis-related differentially expressed genes (DEGs) between subtypes, which were then selected to construct a prognostic signature. The clinical utility of the signature was verified using the validation datasets GSE116918 and GSE46602. A nomogram was established to predict patient survival. Finally, differentially enriched hallmark gene sets were revealed between the different risk groups. Results: Two anoikis-related molecular subtypes were identified, and cluster 1 had poor prognosis, higher stemness, advanced clinical features, and differential immune cell infiltration. Next, 13 clinically relevant anoikis-related DEGs were identified, and five of them (CKS2, CDC20, FMOD, CD38, and MSMB) were selected to build a prognostic signature. This gene signature had a high prognostic value. A nomogram that combined Gleason score, T stage, and risk score could accurately predict patient survival. Furthermore, gene sets closely related with DNA repair were differentially expressed in the different risk groups. Conclusion: A novel, clinically relevant five-anoikis-related gene signature was a powerful prognostic biomarker for PC.

Cell-free DNA methylome analysis for early preeclampsia prediction.

Preeclampsia (PE) is a leading cause for peripartal morbidity, especially if developing early in gestation. To enable prophylaxis in the prevention of PE, pregnancies at risk of PE must be identified early-in the first trimester. To identify at-risk pregnancies we profiled methylomes of plasma-derived, cell-free DNA from 498 pregnant women, of whom about one-third developed early-onset PE. We detected DNA methylation differences between control and PE pregnancies that enabled risk stratification at PE diagnosis but also presymptomatically, at around 12 weeks of gestation (range 9-14 weeks). The first-trimester risk prediction model was validated in an external cohort collected from two centers (area under the curve (AUC) = 0.75) and integrated with routinely available maternal risk factors (AUC = 0.85). The combined risk score correctly predicted 72% of patients with early-onset PE at 80% specificity. These preliminary results suggest that cell-free DNA methylation profiling is a promising tool for presymptomatic PE risk assessment, and has the potential to improve treatment and follow-up in the obstetric clinic.