The Alzheimer's comorbidity phenome: mining from a large patient database and phenome-driven genetics prediction

Data-driven comorbidity analysis of 100 common disorders reveals patient subgroups with differing mortality risks and laboratory correlates

The populational heterogeneity of a disease, in part due to comorbidity, poses several complexities. Individual comorbidity profiles, on the other hand, contain useful information to refine phenotyping, prognostication, and risk assessment, and they provide clues to underlying biology. Nevertheless, the spectrum and the implications of the diagnosis profiles remain largely uncharted. Here we mapped comorbidity patterns in 100 common diseases using 4-year retrospective data from 526,779 patients and developed an online tool to visualize the results. Our analysis exposed disease-specific patient subgroups with distinctive diagnosis patterns, survival functions, and laboratory correlates. Computational modeling and real-world data shed light on the structure, variation, and relevance of populational comorbidity patterns, paving the way for improved diagnostics, risk assessment, and individualization of care. Variation in outcomes and biological correlates of a disease emphasizes the importance of evaluating the generalizability of current treatment strategies, as well as considering the limitations that selective inclusion criteria pose on clinical trials.

Mining comorbidities of opioid use disorder from FDA adverse event reporting system and patient electronic health records

BACKGROUND

Opioid use disorder (OUD) has become an urgent health problem. People with OUD often experience comorbid medical conditions. Systematical approaches to identifying co-occurring conditions of OUD can facilitate a deeper understanding of OUD mechanisms and drug discovery. This study presents an integrated approach combining data mining, network construction and ranking, and hypothesis-driven case-control studies using patient electronic health records (EHRs).

METHODS

First, we mined comorbidities from the US Food and Drug Administration Adverse Event Reporting System (FAERS) of 12 million unique case reports using frequent pattern-growth algorithm. The performance of OUD comorbidity mining was measured by precision and recall using manually curated known OUD comorbidities. We then constructed a disease comorbidity network using mined association rules and further prioritized OUD comorbidities. Last, novel OUD comorbidities were independently tested using EHRs of 75 million unique patients.

RESULTS

The OUD comorbidities from association rules mining achieves a precision of 38.7% and a recall of 78.2 Based on the mined rules, the global DCN was constructed with 1916 nodes and 32,175 edges. The network-based OUD ranking result shows that 43 of 55 known OUD comorbidities were in the first decile with a precision of 78.2%. Hypothyroidism and type 2 diabetes were two top-ranked novel OUD comorbidities identified by data mining and network ranking algorithms. Based on EHR-based case-control studies, we showed that patients with OUD had significantly increased risk for hyperthyroidism (AOR = 1.46, 95% CI 1.43-1.49, p value < 0.001), hypothyroidism (AOR = 1.45, 95% CI 1.42-1.48, p value < 0.001), type 2-diabetes (AOR = 1.28, 95% CI 1.26-1.29, p value < 0.001), compared with individuals without OUD.

CONCLUSION

Our study developed an integrated approach for identifying and validating novel OUD comorbidities from health records of 87 million unique patients (12 million for discovery and 75 million for validation), which can offer new opportunities for OUD mechanism understanding, drug discovery, and multi-component service delivery for co-occurring medical conditions among patients with OUD.

Multiple comorbid conditions and healthcare resource utilization among adult patients with hyperkalemia: A retrospective observational cohort study using association rule mining

Objectives

To estimate the prevalence of specific comorbid conditions (CCs) and multiple comorbid conditions (MCCs) among adult patients with hyperkalemia and examine the associations between MCCs and healthcare resource utilization (HRU) and costs.

Methods

This retrospective observational cohort study was conducted using a large administrative claims database. We identified patients with hyperkalemia (ICD-10-CM: E87.5; or serum potassium >5.0 mEq/L; or NDC codes for either patiromer or sodium polystyrene sulfonate) during the study period (1/1/2016–6/30/2019). The earliest service/claim date with evidence of hyperkalemia was identified as index date. Qualified patients had ≥12 months of enrolment before and after index date, ≥18 years of age. Comorbid conditions were assessed using all data within 12 months prior to the index date. Healthcare resource utilization and costs were estimated using all data within 12 months after the index date. Association rule mining was applied to identify MCCs. Generalized linear models were used to examine the associations between MCCs and HRU and costs.

Results

Of 22,154 patients with hyperkalemia, 94% had ≥3 CCs. The most common individual CCs were chronic kidney disease (CKD, 85%), hypertension (HTN, 83%), hyperlipidemia (HLD, 81%), and diabetes mellitus (DM, 47%). The most common dyad combination of CCs was CKD+HTN (71%). The most common triad combination was CKD+HTN+HLD (62%). The most common quartet combination was CKD+HTN+HLD+DM (36%). The increased number of CCs were significantly associated with increased ED visits, length of hospital stays, and total healthcare costs (all p-value < 0.0001).

Conclusions

MCCs are very prevalent among patients with hyperkalemia and are strongly associated with HRU and costs.

The Role of Medication Data to Enhance the Prediction of Alzheimer's Progression Using Machine Learning

Early detection of Alzheimer's disease (AD) progression is crucial for proper disease management. Most studies concentrate on neuroimaging data analysis of baseline visits only. They ignore the fact that AD is a chronic disease and patient's data are naturally longitudinal. In addition, there are no studies that examine the effect of dementia medicines on the behavior of the disease. In this paper, we propose a machine learning-based architecture for early progression detection of AD based on multimodal data of AD drugs and cognitive scores data. We compare the performance of five popular machine learning techniques including support vector machine, random forest, logistic regression, decision tree, and K-nearest neighbor to predict AD progression after 2.5 years. Extensive experiments are performed using an ADNI dataset of 1036 subjects. The cross-validation performance of most algorithms has been improved by fusing the drugs and cognitive scores data. The results indicate the important role of patient's taken drugs on the progression of AD disease.

Prognosis prediction model for conversion from mild cognitive impairment to Alzheimer's disease created by integrative analysis of multi-omics data

BACKGROUND

Mild cognitive impairment (MCI) is a precursor to Alzheimer's disease (AD), but not all MCI patients develop AD. Biomarkers for early detection of individuals at high risk for MCI-to-AD conversion are urgently required.

METHODS

We used blood-based microRNA expression profiles and genomic data of 197 Japanese MCI patients to construct a prognosis prediction model based on a Cox proportional hazard model. We examined the biological significance of our findings with single nucleotide polymorphism-microRNA pairs (miR-eQTLs) by focusing on the target genes of the miRNAs. We investigated functional modules from the target genes with the occurrence of hub genes though a large-scale protein-protein interaction network analysis. We further examined the expression of the genes in 610 blood samples (271 ADs, 248 MCIs, and 91 cognitively normal elderly subjects [CNs]).

RESULTS

The final prediction model, composed of 24 miR-eQTLs and three clinical factors (age, sex, and APOE4 alleles), successfully classified MCI patients into low and high risk of MCI-to-AD conversion (log-rank test P = 3.44 × 10^-4 and achieved a concordance index of 0.702 on an independent test set. Four important hub genes associated with AD pathogenesis (SHC1, FOXO1, GSK3B, and PTEN) were identified in a network-based meta-analysis of miR-eQTL target genes. RNA-seq data from 610 blood samples showed statistically significant differences in PTEN expression between MCI and AD and in SHC1 expression between CN and AD (PTEN, P = 0.023; SHC1, P = 0.049).

CONCLUSIONS

Our proposed model was demonstrated to be effective in MCI-to-AD conversion prediction. A network-based meta-analysis of miR-eQTL target genes identified important hub genes associated with AD pathogenesis. Accurate prediction of MCI-to-AD conversion would enable earlier intervention for MCI patients at high risk, potentially reducing conversion to AD.

Analysis of disease organ as a novel phenotype towards disease genetics understanding

Discerning the modular nature of human diseases through computational approaches calls for diverse data. The finding sites of diseases, like other disease phenotypes, possess rich information in understanding disease genetics. Yet, analysis of the rich knowledge of disease finding sites has not been comprehensively investigated. In this study, we built a large-scale disease organ network (DON) based on 76,561 disease-organ associations (for 37,615 diseases and 3492 organs) extracted from the United Medical Language System (UMLS) Metathesaurus. We investigated how phenotypic organ similarity among diseases in DON reflects disease gene sharing. We constructed a disease genetic network (DGN) using curated disease-gene associations and demonstrated that disease pairs with higher organ similarities not only are more likely to share genes, but also tend to share more genes. Based on community detection algorithm, we showed that phenotypic disease clusters on DON significantly correlated with genetic disease clusters on DGN. We compared DON with a state-of-art disease phenotype network, disease manifestation network (DMN), that we have recently constructed, and demonstrated that DON contains complementary knowledge for disease genetics understanding.