A multi-population phenome-wide association study of genetically-predicted height in the Million Veteran Program

Novel Genes Associated with Atrial Fibrillation and the Predictive Models for AF Incorporating Polygenic Risk Scores and PheWAS-Derived Risk Factors

BACKGROUND

Atrial fibrillation (AF), the most common atrial arrhythmia, presents varied clinical manifestations. Despite the identification of genetic loci associated with AF, particularly in specific populations, research within Asian ethnicities remains limited. This study aimed to develop predictive models for AF using AF-associated single-nucleotide polymorphisms (SNPs) from a Genome-Wide Association Study (GWAS) on a substantial cohort of Taiwanese individuals, evaluating the models' predictive efficacy.

METHODS

Involving 75,121 subjects, including 5,694 AF patients and 69,427 normal controls with GWAS data, the study merged polygenic risk scores (PRS) from AF-associated SNPs with Phenome-wide association study (PheWAS)-derived risk factors. Advanced statistical and machine learning techniques were employed to develop and evaluate AF predictive models for discrimination and calibration.

RESULTS

The study identified the top 30 significant SNPs associated with AF, predominantly on chromosomes 10 and 16, implicating genes like NEURL1, SH3PXD2A, INA, NT5C2, STN1, and ZFHX3. Notably, INA, NT5C2, and STN1 were newly linked to AF. The GWAS predictive power using PRS-CS analysis for AF exhibited an area under the curve (AUC) of 0.600 (P < 0.001), improving to 0.855 (P < 0.001) after adjusting for age and gender. PheWAS analysis showed the top 10 diseases associated with these genes were circulatory system diseases.

CONCLUSIONS

Integrating genetic and phenotypic data enhanced the accuracy and clinical relevance of AF predictive models. The findings suggest promise for refining AF risk assessment, enabling personalized interventions, and reducing AF-related morbidity and mortality burdens.

Gene Expression Signatures Predict First-Year Response to Somapacitan Treatment in Children With Growth Hormone Deficiency

CONTEXT

The pretreatment blood transcriptome predicts growth response to daily growth hormone (GH) therapy with high accuracy.

OBJECTIVE

Investigate response prediction using pretreatment transcriptome in children with GH deficiency (GHD) treated with once-weekly somapacitan, a novel long-acting GH.

METHODS

REAL4 is a randomized, multinational, open-label, active-controlled parallel group phase 3 trial, comprising a 52-week main phase and an ongoing 3-year safety extension (NCT03811535). A total of 128/200 treatment-naïve prepubertal children with GHD consented to baseline blood transcriptome profiling. They were randomized 2:1 to subcutaneous somapacitan (0.16 mg/kg/week) or daily GH (0.034 mg/kg/day). Differential RNA-seq analysis and machine learning were used to predict therapy response.

RESULTS

121/128 samples passed quality control. Children treated with somapacitan (n = 76) or daily GH (n = 45) were categorized based on fastest and slowest growing quartiles at week 52. Prediction of height velocity (HV; cm/year) was excellent for both treatments (out of bag [OOB] area under curve [AUC]: 0.98-0.99; validation AUC: 0.83-0.84), as was prediction of secondary markers of growth response: HV standard deviation score (SDS) (0.99-1.0; 0.75-0.78), change from baseline height SDS (ΔHSDS) (0.98-1.0; 0.61-0.75), and change from baseline insulin-like growth factor-I SDS (ΔIGF-I SDS) (0.96-1.0; 0.85-0.88). Genes previously identified as predictive of GH therapy response were consistently better at predicting the fastest growers in both treatments in this study (OOB AUC: 0.93-0.97) than the slowest (0.67-0.85).

CONCLUSION

Pretreatment transcriptome predicts first-year growth response in somapacitan-treated children with GHD. A common set of genes can predict the treatment response to both once-weekly somapacitan and conventional daily GH. This approach could potentially be developed into a clinically applicable pretreatment test to improve clinical management.

A missense SNP in the tumor suppressor SETD2 reduces H3K36me3 and mitotic spindle integrity in Drosophila

Mutations in SETD2 are among the most prevalent drivers of renal cell carcinoma (RCC). We identified a novel single nucleotide polymorphism (SNP) in SETD2, E902Q, within a subset of RCC patients, which manifests as both an inherited or tumor-associated somatic mutation. To determine if the SNP is biologically functional, we used CRISPR-based genome editing to generate the orthologous mutation within the Drosophila melanogaster Set2 gene. In Drosophila, the homologous amino acid substitution, E741Q, reduces H3K36me3 levels comparable to Set2 knockdown, and this loss is rescued by reintroduction of a wild-type Set2 transgene. We similarly uncovered significant defects in spindle morphogenesis, consistent with the established role of SETD2 in methylating α-Tubulin during mitosis to regulate microtubule dynamics and maintain genome stability. These data indicate the Set2 E741Q SNP affects both histone methylation and spindle integrity. Moreover, this work further suggests the SETD2 E902Q SNP may hold clinical relevance.

Opportunistic Screening for Atrial Fibrillation on Routine Chest Computed Tomography

PURPOSE

Quantitative biomarkers from chest computed tomography (CT) can facilitate the incidental detection of important diseases. Atrial fibrillation (AFib) substantially increases the risk for comorbid conditions including stroke. This study investigated the relationship between AFib status and left atrial enlargement (LAE) on CT.

MATERIALS AND METHODS

A total of 500 consecutive patients who had undergone nongated chest CTs were included, and left atrium maximal axial cross-sectional area (LA-MACSA), left atrium anterior-posterior dimension (LA-AP), and vertebral body cross-sectional area (VB-Area) were measured. Height, weight, age, sex, and diagnosis of AFib were obtained from the medical record. Parametric statistical analyses and receiver operating characteristic curves were performed. Machine learning classifiers were run with clinical risk factors and LA measurements to predict patients with AFib.

RESULTS

Eighty-five patients with a diagnosis of AFib were identified. Mean LA-MACSA and LA-AP were significantly larger in patients with AFib than in patients without AFib (28.63 vs. 20.53 cm 2 , P <0.000001; 4.34 vs. 3.5 cm, P <0.000001, respectively), both with area under the curves (AUCs) of 0.73. Multivariable logistic regression analysis including age, sex, and VB-Area with LA-MACSA improved the AUC for predicting AFib (AUC=0.77). An LA-MACSA threshold of 30 cm 2 demonstrated high specificity for AFib diagnosis at 92% and sensitivity of 48%, and LA-AP threshold at 4.5 cm demonstrated 90% specificity and 42% sensitivity. A Bayesian machine learning model using age, sex, height, body surface area, and LA-MACSA predicted AFib with an AUC of 0.743.

CONCLUSIONS

LA-MACSA or LA-AP can be rapidly measured from routine chest CT, and when >30 cm 2 and >4.5 cm, respectively, are specific indicators to predict patients at increased risk for AFib.

Instrumental variable model average with applications in Mendelian randomization

Mendelian randomization is a technique used to examine the causal effect of a modifiable exposure on a trait using an observational study by utilizing genetic variants. The use of many instruments can help to improve the estimation precision but may suffer bias when the instruments are weakly associated with the exposure. To overcome the difficulty of high-dimensionality, we propose a model average estimator which involves using different subsets of instruments (single nucleotide polymorphisms, SNPs) to predict the exposure in the first stage, followed by weighting the submodels' predictions using penalization by common penalty functions such as least absolute shrinkage and selection operator (LASSO), smoothly clipped absolute deviation (SCAD) and minimax concave penalty (MCP). The model averaged predictions are then used as a genetically predicted exposure to obtain the estimation of the causal effect on the response in the second stage. The novelty of our model average estimator also lies in that it allows the number of submodels and the submodels' sizes to grow with the sample size. The practical performance of the estimator is examined in a series of numerical studies. We apply the proposed method on a real genetic dataset investigating the relationship between stature and blood pressure.

Do the short die young? Evidence from a large sample of deceased Polish adults

Body height is associated with various socioeconomic and health-related outcomes. Despite numerous studies, the relationship between stature and longevity remains uncertain. This study explores the association between self-reported height and lifespan. Data from 848,860 adults who died between 2004 and 2008 in Poland were collected. After excluding a small proportion of records due to missing data or errors, we examined records for 848,387 individuals (483,281 men, age range: 20–110 years; 365,106 women, age range: 20–112 years). Height was expressed as standardized residual variance derived from linear regression in order to eliminate the variance of year of birth on height. After the elimination of the cohort effect, five height classes were designated using centiles: very short, short, medium, tall and very tall. The differences between sexes and among classes were evaluated with two-way ANOVA and post hoc Tukey’s test. The effect size was assessed using partial eta squared (η2). Pearson’s r coefficients of correlation were calculated. The effect of sex on lifespan was nearly 17 times stronger than the effect of height. No correlation between height and lifespan was found. In conclusion, these findings do not support the hypothesis that taller people have a longevity advantage. We offer tentative explanations for the obtained results.

Insights into Causal Cardiovascular Risk Factors from Mendelian Randomization

PURPOSE OF THE REVIEW

This review summarizes major insights into causal risk factors for cardiovascular disease (CVD) by using Mendelian randomization (MR) to obtain unconfounded estimates, contextualized within its strengths and weaknesses.

RECENT FINDINGS

MR studies have confirmed the role of major CVD risk factors, including alcohol, smoking, adiposity, blood pressure, type 2 diabetes, lipids, and possibly inflammation, but added that the relation with alcohol is likely linear, confirmed the role of diastolic blood pressure, identified apolipoprotein B as the major target lipid, and foreshadowed results of some trials concerning anti-inflammatories. Identifying a healthy diet and the role of early life influences, such as birth weight, has proved more difficult. Use of MR has winnowed empirically driven hypotheses about CVD into a set of genetically validated targets of intervention. Greater inclusion of global diversity in genetic studies and the use of an overarching framework would enable even more informative MR studies.