Phenome-Wide Association Study of Polygenic Risk Score for Alzheimer's Disease in Electronic Health Records

Research Advance of Causal Inference in Clinical Medicine: A Bibliometrics Analysis via Citespace

Objective

Causal inference in clinical medicine provides scientific evidence for precision medicine and individualized treatment by revealing the true associations between interventions and health outcomes. This study aims to conduct a comprehensive bibliometric analysis to identify current research trends, primary themes, and future directions for the application of causal inference in clinical medicine.

Methods

We conducted a literature search in the Web of Science database using causal inference and medical terminology as subject keywords, covering the period from January 1986 to December 2024. After screening, we obtained 4,316 documents for analysis. Utilizing CiteSpace to generate network diagrams, we analyzed data related to the number of publications, citation analysis, collaboration relationships, keyword co-occurrence, and highlighted terms to illustrate the knowledge map and collaboration network in this field.

Results

Publications on medical causal inference shows a fluctuating growth trend over time. The United States was the top contributors to this field. Harvard University is the leading research institution. George David Smith is the most prolific author, Robbins JM is the most cited scholar. The major research hotspots concentrated in fields such as epidemiology, coronary heart disease and health. Notably, marginal structural models, counterfactual forecasting, and Mendelian randomization have consistently been key methodologies in research. The burstness of keywords reveals that big data, DNA methylation, and robust estimation are emerging research directions.

Conclusion

In clinical research, counterfactual forecasting provides prospective guidance for optimizing clinical strategies; Mendelian randomization helps uncover potential therapeutic targets; and marginal structural models enhance the accuracy of causal effect estimation in clinical studies. The future integration of various data sources to improve causal inference methods is anticipated to enhance the sensitivity and specificity of trials, ultimately elucidating the complex mechanisms of diseases and drug effects. The literature retrieve strategy and the metrics of the tools adopted may have a certain impact on the results of this study.

Evaluating the Bidirectional Causal Effects of Alzheimer's Disease Across Multiple Conditions: A Systematic Review and Meta-Analysis of Mendelian Randomization Studies

This study systematically evaluates and meta-analyzes Mendelian randomization studies on the bidirectional causal relationship between Alzheimer's disease (AD) and systemic diseases. We searched five databases, assessed study quality, and extracted data. Diseases were classified using ICD-11, and the meta-analysis was performed with RevMan 5.4. A total of 56 studies identified genetic links between AD susceptibility and systemic diseases. Notably, genetic proxies for hip osteoarthritis (OR = 0.80; p = 0.007) and rheumatoid arthritis (OR = 0.97; p = 0.004) were inversely associated with AD risk, while gout (OR = 1.02; p = 0.049) showed a positive association. Genetic liability to depression (OR = 1.03; p = 0.001) elevated AD risk, and AD genetic risk increased susceptibility to delirium (OR = 1.32; p = 0.0005). Cardiovascular traits, including coronary artery disease (OR = 1.07; p = 0.021) and hypertension (OR = 4.30; p = 0.044), were causally linked to a higher AD risk. Other conditions, such as insomnia, chronic periodontitis, migraine, and certain cancers, exhibited significant genetic correlations. Intriguingly, herpes zoster (OR = 0.87; p = 0.005) and cataracts (OR = 0.96; p = 0.012) demonstrated inverse genetic associations with AD. These findings suggest potential therapeutic targets and preventive strategies, emphasizing the need to address comorbid systemic diseases to reduce AD risk and progression.

Exploring beyond diagnoses in electronic health records to improve discovery: a review of the phenome-wide association study

Objective

The phenome-wide association study (PheWAS) systematically examines the phenotypic spectrum extracted from electronic health records (EHRs) to uncover correlations between phenotypes and exposures. This review explores methodologies, highlights challenges, and outlines future directions for EHR-driven PheWAS.

Materials and Methods

We searched the PubMed database for articles spanning from 2010 to 2023, and we collected data regarding exposures, phenotypes, cohorts, terminologies, replication, and ancestry.

Results

Our search yielded 690 articles. Following exclusion criteria, we identified 291 articles published between January 1, 2010, and December 31, 2023. A total number of 162 (55.6%) articles defined phenomes using phecodes, indicating that research is reliant on the organization of billing codes. Moreover, 72.8% of articles utilized exposures consisting of genetic data, and the majority (69.4%) of PheWAS lacked replication analyses.

Discussion

Existing literature underscores the need for deeper phenotyping, variability in PheWAS exposure variables, and absence of replication in PheWAS. Current applications of PheWAS mainly focus on cardiovascular, metabolic, and endocrine phenotypes; thus, applications of PheWAS in uncommon diseases, which may lack structured data, remain largely understudied.

Conclusions

With modern EHRs, future PheWAS should extend beyond diagnosis codes and consider additional data like clinical notes or medications to create comprehensive phenotype profiles that consider severity, temporality, risk, and ancestry. Furthermore, data interoperability initiatives may help mitigate the paucity of PheWAS replication analyses. With the growing availability of data in EHR, PheWAS will remain a powerful tool in precision medicine.

A phenome-wide association study of polygenic scores for selected childhood cancer: Results from the UK Biobank

The aim of this study was to scan phenotypes in adulthood associated with polygenic risk scores (PRS) for childhood cancers with well-articulated genetic architectures-acute lymphoblastic leukemia (ALL), Ewing sarcoma, and neuroblastoma-to examine genetic pleiotropy. Furthermore, we aimed to determine which SNPs could drive associations. Per-SNP summary statistics were extracted for PRS calculation. Participants with white British ancestry were exclusively included for analyses. SNPs were queried from the UK Biobank genotype imputation data. Records from the cancer registry, death registry, and inpatient diagnoses were abstracted for phenome-wide scans. Firth logistic regression was used to estimate odds ratios (ORs) and 95% confidence intervals (CIs) alongside corresponding p values, adjusting for age at recruitment and sex. A total of 244,332 unrelated white British participants were included. We observed a significant association between ALL-PRS and ALL (OR: 1.20e+24, 95% CI: 9.08e+14-1.60e+33). In addition, we observed a significant association between high-risk neuroblastoma PRS and nonrheumatic aortic valve disorders (OR: 43.9, 95% CI: 7.42-260). There were no significant phenotype associations with Ewing sarcoma and neuroblastoma PRS. Regarding individual SNPs, rs17607816 increased the risk of ALL (OR: 6.40, 95% CI: 3.26-12.57). For high-risk neuroblastoma, rs80059929 elevated the risk of atrioventricular block (OR: 3.04, 95% CI: 1.85-4.99). Our findings suggest that individuals with genetic susceptibility to ALL may face a lifelong risk for developing ALL, along with a genetic pleiotropic association between high-risk neuroblastoma and circulatory diseases.

Balancing the efforts of chart review and gains in PRS prediction accuracy: An empirical study

OBJECTIVE

Phenotypic misclassification in genetic association analyses can impact the accuracy of PRS-based prediction models. The bias reduction method proposed by Tong et al. (2019) has demonstrated its efficacy in reducing the effects of bias on the estimation of association parameters between genotype and phenotype while minimizing variance by employing chart reviews on a subset of the data for validating phenotypes, however its improvement of subsequent PRS prediction accuracy remains unclear. Our study aims to fill this gap by assessing the performance of simulated PRS models and estimating the optimal number of chart reviews needed for validation.

METHODS

To comprehensively assess the efficacy of the bias reduction method proposed by Tong et al. in enhancing the accuracy of PRS-based prediction models, we simulated each phenotype under different correlation structures (an independent model, a weakly correlated model, a strongly correlated model) and introduced error-prone phenotypes using two distinct error mechanisms (differential and non-differential phenotyping errors). To facilitate this, we used genotype and phenotype data from 12 case-control datasets in the Alzheimer's Disease Genetics Consortium (ADGC) to produce simulated phenotypes. The evaluation included analyses across various misclassification rates of original phenotypes as well as quantities of validation set. Additionally, we determined the median threshold, identifying the minimal validation size required for a meaningful improvement in the accuracy of PRS-based predictions across a broad spectrum.

RESULTS

This simulation study demonstrated that incorporating chart review does not universally guarantee enhanced performance of PRS-based prediction models. Specifically, in scenarios with minimal misclassification rates and limited validation sizes, PRS models utilizing debiased regression coefficients demonstrated inferior predictive capabilities compared to models using error-prone phenotypes. Put differently, the effectiveness of the bias reduction method is contingent upon the misclassification rates of phenotypes and the size of the validation set employed during chart reviews. Notably, when dealing with datasets featuring higher misclassification rates, the advantages of utilizing this bias reduction method become more evident, requiring a smaller validation set to achieve better performance.

CONCLUSION

This study highlights the importance of choosing an appropriate validation set size to balance between the efforts of chart review and the gain in PRS prediction accuracy. Consequently, our study establishes a valuable guidance for validation planning, across a diverse array of sensitivity and specificity combinations.

Improving genetic risk modeling of dementia from real-world data in underrepresented populations

Genetic risk modeling for dementia offers significant benefits, but studies based on real-world data, particularly for underrepresented populations, are limited. We employ an Elastic Net model for dementia risk prediction using single-nucleotide polymorphisms prioritized by functional genomic data from multiple neurodegenerative disease genome-wide association studies. We compare this model with APOE and polygenic risk score models across genetic ancestry groups (Hispanic Latino American sample: 610 patients with 126 cases; African American sample: 440 patients with 84 cases; East Asian American sample: 673 patients with 75 cases), using electronic health records from UCLA Health for discovery and the All of Us cohort for validation. Our model significantly outperforms other models across multiple ancestries, improving the area-under-precision-recall curve by 31-84% (Wilcoxon signed-rank test p-value <0.05) and the area-under-the-receiver-operating characteristic by 11-17% (DeLong test p-value <0.05) compared to the APOE and the polygenic risk score models. We identify shared and ancestry-specific risk genes and biological pathways, reinforcing and adding to existing knowledge. Our study highlights the benefits of integrating functional mapping, multiple neurodegenerative diseases, and machine learning for genetic risk models in diverse populations. Our findings hold potential for refining precision medicine strategies in dementia diagnosis.

Improving genetic risk modeling of dementia from real-world data in underrepresented populations

BACKGROUND

Genetic risk modeling for dementia offers significant benefits, but studies based on real-world data, particularly for underrepresented populations, are limited.

METHODS

We employed an Elastic Net model for dementia risk prediction using single-nucleotide polymorphisms prioritized by functional genomic data from multiple neurodegenerative disease genome-wide association studies. We compared this model with APOE and polygenic risk score models across genetic ancestry groups, using electronic health records from UCLA Health for discovery and All of Us cohort for validation.

RESULTS

Our model significantly outperforms other models across multiple ancestries, improving the area-under-precision-recall curve by 21-61% and the area-under-the-receiver-operating characteristic by 10-21% compared to the APOEand the polygenic risk score models. We identified shared and ancestry-specific risk genes and biological pathways, reinforcing and adding to existing knowledge.

CONCLUSIONS

Our study highlights benefits of integrating functional mapping, multiple neurodegenerative diseases, and machine learning for genetic risk models in diverse populations. Our findings hold potential for refining precision medicine strategies in dementia diagnosis.

Improving genetic risk modeling of dementia from real-world data in underrepresented populations

BACKGROUND

Genetic risk modeling for dementia offers significant benefits, but studies based on real-world data, particularly for underrepresented populations, are limited.

METHODS

We employed an Elastic Net model for dementia risk prediction using single-nucleotide polymorphisms prioritized by functional genomic data from multiple neurodegenerative disease genome-wide association studies. We compared this model with APOE and polygenic risk score models across genetic ancestry groups, using electronic health records from UCLA Health for discovery and All of Us cohort for validation.

RESULTS

Our model significantly outperforms other models across multiple ancestries, improving the area-under-precision-recall curve by 21-61% and the area-under-the-receiver-operating characteristic by 10-21% compared to the APOE and the polygenic risk score models. We identified shared and ancestry-specific risk genes and biological pathways, reinforcing and adding to existing knowledge.

CONCLUSIONS

Our study highlights benefits of integrating functional mapping, multiple neurodegenerative diseases, and machine learning for genetic risk models in diverse populations. Our findings hold potential for refining precision medicine strategies in dementia diagnosis.

Defining the distance between diseases using SNOMED CT embeddings

Characterizing disease relationships is essential to biomedical research to understand disease etiology and improve clinical decision-making. Measurements of distance between disease pairs enable valuable research tasks, such as subgrouping patients and identifying common time courses of disease onset. Distance metrics developed in prior work focused on smaller, targeted disease sets. Distance metrics covering all diseases have not yet been defined, which limits the applications to a broader disease spectrum. Our current study defines disease distances for all disease pairs within the International Classification of Diseases, version 10 (ICD-10), the diagnostic classification system universally used in electronic health records. Our proposed distance is computed based on a biomedical ontology, SNOMED CT (Systemized Nomenclature of Medicine, Clinical Terms), which can also be viewed as a structured knowledge graph. We compared the knowledge graph-based metric to three other distance metrics based on the hierarchical structure of ICD, clinical comorbidity, and genetic correlation, to evaluate how each may capture similar or unique aspects of disease relationships. We show that our knowledge graph-based distance metric captures known phenotypic, clinical, and molecular characteristics at a finer granularity than the other three. With the continued growth of using electronic health records data for research, we believe that our distance metric will play an important role in subgrouping patients for precision health, and enabling individualized disease prevention and treatments.

Polygenic Liability to Alzheimer's Disease Is Associated with a Wide Range of Chronic Diseases: A Cohort Study of 312,305 Participants

BACKGROUND

Alzheimer's disease (AD) patients rank among the highest levels of comorbidities compared to persons with other diseases. However, it is unclear whether the conditions are caused by shared pathophysiology due to the genetic pleiotropy for AD risk genes.

OBJECTIVE

To figure out the genetic pleiotropy for AD risk genes in a wide range of diseases.

METHODS

We estimated the polygenic risk score (PRS) for AD and tested the association between PRS and 16 ICD10 main chapters, 136 ICD10 level-1 chapters, and 377 diseases with cases more than 1,000 in 312,305 individuals without AD diagnosis from the UK Biobank.

RESULTS

After correction for multiple testing, AD PRS was associated with two main ICD10 chapters: Chapter IV (endocrine, nutritional and metabolic diseases) and Chapter VII (eye and adnexa disorders). When narrowing the definition of the phenotypes, positive associations were observed between AD PRS and other types of dementia (OR = 1.39, 95% CI [1.34, 1.45], p = 1.96E-59) and other degenerative diseases of the nervous system (OR = 1.18, 95% CI [1.13, 1.24], p = 7.74E-10). In contrast, we detected negative associations between AD PRS and diabetes mellitus, obesity, chronic bronchitis, other retinal disorders, pancreas diseases, and cholecystitis without cholelithiasis (ORs range from 0.94 to 0.97, FDR < 0.05).

CONCLUSION

Our study confirms several associations reported previously and finds some novel results, which extends the knowledge of genetic pleiotropy for AD in a range of diseases. Further mechanistic studies are necessary to illustrate the molecular mechanisms behind these associations.

Discovery of Drug-Responsive Phenomic Alteration-Related Driver Genes in the Treatment of Coronary Heart Disease

Background

The Xueyu Zheng (XYZ) phenome is central to coronary heart disease (CHD), but efforts to detect genetic associations in the XYZ phenome have been disappointing.

Methods

The phenomic alteration-related genes (PARGs) for the XYZ phenome were screened using |ρ| > 0.4 and p < 0.05 after treatment with Danhong injection at day 14 and day 30. Then, the driver genes for the Protein-Protein Interaction (PPI) networks of the PARGs established using STRING 11.0 were detected using a personalized network control algorithm (PNC). Finally, the molecular correlations of the driver genes with the XYZ phenome were analyzed with the Gene Ontology (GO) biological processes and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways from a holistic viewpoint.

Results

A total of 525 and 309 PARGs in the XYZ phenome at day 14 and day 30 were identified. These genes were separately enriched in 48 and 35 pathways. Furthermore, five driver genes were detected. These genes were mainly correlated with endoplasmic reticulum stress-mediated apoptosis and autophagy regulation, which could suppress atherosclerosis progression.

Conclusion

Our study detected the drug-responsive PARGs of the XYZ phenome in CHD and provides an exemplary strategy to investigate the genetic associations among this common phenome and its component symptoms in patients with CHD.

Trial Registration

ClinicalTrials.gov, NCT01681316; registered on September 7, 2012.