Mining cancer-specific disease comorbidities from a large observational health database

Multimorbidity and mortality: A data science perspective

Background

With multimorbidity becoming the norm rather than the exception, the management of multiple chronic diseases is a major challenge facing healthcare systems worldwide.

Methods

Using a large, nationally representative database of electronic medical records from the United Kingdom spanning the years 2005–2016 and consisting over 4.5 million patients, we apply statistical methods and network analysis to identify comorbid pairs and triads of diseases and identify clusters of chronic conditions across different demographic groups. Unlike many previous studies, which generally adopt cross-sectional designs based on single snapshots of closed cohorts, we adopt a longitudinal approach to examine temporal changes in the patterns of multimorbidity. In addition, we perform survival analysis to examine the impact of multimorbidity on mortality.

Results

The proportion of the population with multimorbidity has increased by approximately 2.5 percentage points over the last decade, with more than 17% having at least two chronic morbidities. We find that the prevalence and the severity of multimorbidity, as quantified by the number of co-occurring chronic conditions, increase progressively with age. Stratifying by socioeconomic status, we find that people living in more deprived areas are more likely to be multimorbid compared to those living in more affluent areas at all ages. The same trend holds consistently for all years in our data. In general, hypertension, diabetes, and respiratory-related diseases demonstrate high in-degree centrality and eigencentrality, while cardiac disorders show high out-degree centrality.

Conclusions

We use data-driven methods to characterize multimorbidity patterns in different demographic groups and their evolution over the past decade. In addition to a number of strongly associated comorbid pairs (e.g., cardiac-vascular and cardiac-metabolic disorders), we identify three principal clusters: a respiratory cluster, a cardiovascular cluster, and a mixed cardiovascular-renal-metabolic cluster. These are supported by established pathophysiological mechanisms and shared risk factors, and largely confirm and expand on the results of existing studies in the medical literature. Our findings contribute to a more quantitative understanding of the epidemiology of multimorbidity, an important pre-requisite for developing more effective medical care and policy for multimorbid patients.

Mental Health Comorbidity Analysis in Pediatric Patients with Autism Spectrum Disorder Using Rhode Island Medical Claims Data

Identification of comorbidity subgroups linked with Autism Spectrum Disorder (ASD) could provide promising insight into learning more about this disorder. This study sought to use the Rhode Island All-Payer Claims Database to examine mental health conditions linked to ASD. Medical claims data for ASD patients and one or more mental health conditions were analyzed using descriptive statistics, association rule mining (ARM), and sequential pattern mining (SPM). The results indicated that patients with ASD have a higher proportion of mental health diagnoses than the general pediatric population. ARM and SPM methods identified patterns of comorbidities commonly seen among ASD patients. Based on the observed patterns and temporal sequences, suicidal ideation, mood disorders, anxiety, and conduct disorders may need focused attention prospectively. Understanding more about groupings of ASD patients and their comorbidity burden can help bridge gaps in knowledge and make strides toward improved outcomes for patients with ASD.

Comorbidity in Older Patients Hospitalized with Cancer in Northeast China based on Hospital Discharge Data

Patients with cancer often carry the dual burden of the cancer itself and other co-existing medical conditions. The problems associated with comorbidities among elderly cancer patients are more prominent compared with younger patients. This study aimed to identify common cancer-related comorbidities in elderly patients through routinely collected hospital discharge data and to use association rules to analyze the prevalence and patterns of these comorbidities in elderly cancer patients at different cancer sites. We collected the discharge data of 80,574 patients who were diagnosed with cancers of the esophagus, stomach, colorectum, liver, lung, female breast, cervix, and thyroid between 2016 and 2018. The same number of non-cancer patients were randomly selected as the control group and matched with the case group by age and gender. The results showed that cardiovascular diseases, metabolic diseases, digestive diseases, and anemia were the most common comorbidities in elderly patients with cancer. The comorbidity patterns differed based on the cancer site. Elderly patients with liver cancer had the highest risk of comorbidities, followed by lung cancer, gastrointestinal cancer, thyroid cancer, and reproductive cancer. For example, elderly patients with liver cancer had the higher risk of the comorbid infectious and digestive diseases, whereas patients with lung cancer had the higher risk of the comorbid respiratory system diseases. The findings can assist clinicians in diagnosing comorbidities and contribute to the allocation of medical resources.

Sensitivity of comorbidity network analysis

OBJECTIVES

Comorbidity network analysis (CNA) is a graph-theoretic approach to systems medicine based on associations revealed from disease co-occurrence data. Researchers have used CNA to explore epidemiological patterns, differentiate populations, characterize disorders, and more; but these techniques have not been comprehensively evaluated. Our objectives were to assess the stability of common CNA techniques.

MATERIALS AND METHODS

We obtained seven co-occurrence data sets, most from previous CNAs, coded using several ontologies. We constructed comorbidity networks under various modeling procedures and calculated summary statistics and centrality rankings. We used regression, ordination, and rank correlation to assess these properties' sensitivity to the source of data and construction parameters.

RESULTS

Most summary statistics were robust to variation in link determination but somewhere sensitive to the association measure. Some more effectively than others discriminated among networks constructed from different data sets. Centrality rankings, especially among hubs, were somewhat sensitive to link determination and highly sensitive to ontology. As multivariate models incorporated additional effects, comorbid associations among low-prevalence disorders weakened while those between high-prevalence disorders shifted negative.

DISCUSSION

Pairwise CNA techniques are generally robust, but some analyses are highly sensitive to certain parameters. Multivariate approaches expose additional conceptual and technical limitations to the usual pairwise approach.

CONCLUSION

We conclude with a set of recommendations we believe will help CNA researchers improve the robustness of results and the potential of follow-up research.

Large-scale mining disease comorbidity relationships from post-market drug adverse events surveillance data

Background

Systems approaches in studying disease relationship have wide applications in biomedical discovery, such as disease mechanism understanding and drug discovery. The FDA Adverse Event Reporting System (FAERS) contains rich information about patient diseases, medications, drug adverse events and demographics of 17 million case reports. Here, we explored this data resource to mine disease comorbidity relationships using association rule mining algorithm and constructed a disease comorbidity network.

Results

We constructed a disease comorbidity network with 1059 disease nodes and 12,608 edges using association rule mining of FAERS (14,157 rules). We evaluated the performance of comorbidity mining from FAERS using known disease comorbidities of multiple sclerosis (MS), psoriasis and obesity that represent rare, moderate and common disease respectively. Comorbidities of MS, obesity and psoriasis obtained from our network achieved precisions of 58.6%, 73.7%, 56.2% and recalls 87.5%, 69.2% and 72.7% separately. We performed comparative analysis of the disease comorbidity network with disease semantic network, disease genetic network and disease treatment network. We showed that (1) disease comorbidity clusters exhibit significantly higher semantic similarity than random network (0.18 vs 0.10); (2) disease comorbidity clusters share significantly more genes (0.46 vs 0.06); and (3) disease comorbidity clusters share significantly more drugs (0.64 vs 0.17). Finally, we demonstrated that the disease comorbidity network has potential in uncovering novel disease relationships using asthma as a case study.

Conclusions

Our study presented the first comprehensive attempt to build a disease comorbidity network from FDA Adverse Event Reporting System. This network shows well correlated with disease semantic similarity, disease genetics and disease treatment, which has great potential in disease genetics prediction and drug discovery.

Applications of network analysis to routinely collected health care data: a systematic review

Objective

To survey network analyses of datasets collected in the course of routine operations in health care settings and identify driving questions, methods, needs, and potential for future research.

Materials and Methods

A search strategy was designed to find studies that applied network analysis to routinely collected health care datasets and was adapted to 3 bibliographic databases. The results were grouped according to a thematic analysis of their settings, objectives, data, and methods. Each group received a methodological synthesis.

Results

The search found 189 distinct studies reported before August 2016. We manually partitioned the sample into 4 groups, which investigated institutional exchange, physician collaboration, clinical co-occurrence, and workplace interaction networks. Several robust and ongoing research programs were discerned within (and sometimes across) the groups. Little interaction was observed between these programs, despite conceptual and methodological similarities.

Discussion

We use the literature sample to inform a discussion of good practice at this methodological interface, including the concordance of motivations, study design, data, and tools and the validation and standardization of techniques. We then highlight instances of positive feedback between methodological development and knowledge domains and assess the overall cohesion of the sample.

Interrogating Patient-level Genomics and Mouse Phenomics towards Understanding Cytokines in Colorectal Cancer Metastasis

Background: Colorectal cancer is the second leading cancer-related death worldwide and a majority of patients die from metastasis. Chronic intestinal inflammation plays an important role in tumor progression of colorectal cancer. However, few study works on systematically predicting colorectal cancer metastasis using inflammatory cytokine genes. Results: We developed a supervised machine learning approach to predict colorectal cancer tumor progression using patient level genomic features. To better understand the role of cytokines, we integrated the metastatic-related genes from mouse phenotypic data. In addition, pathway analysis and network visualization were also applied to top significant genes ranked by feature weights of the final prediction model. The combined model of cytokines and mouse phenotypes achieved a predictive accuracy of 75.54%, higher than the model based on mouse phenotypes independently (70.42%, p-value<0.05). In additional, the combined model outperformed the model based on the existing metastatic-related epithelial-to-mesenchymal transition (EMT) genes (75.54% vs. 71.61%, p-value<0.05). We also observed that the most important cytokine gene features of the our model interact with the cancer driver genes and are highly associated with the colorectal cancer metastasis signaling pathway. Conclusion: We developed a combined model using both cytokine and mouse phenotype information to predict colorectal cancer metastasis. The results suggested that the inflammatory cytokines increase the power of predicting metastasis. We also systematically demonstrated the critical role of cytokines in progression of colorectal tumor.

Context-sensitive network-based disease genetics prediction and its implications in drug discovery

Motivation

Disease phenotype networks play an important role in computational approaches to identifying new disease-gene associations. Current disease phenotype networks often model disease relationships based on pairwise similarities, therefore ignore the specific context on how two diseases are connected. In this study, we propose a new strategy to model disease associations using context-sensitive networks (CSNs). We developed a CSN-based phenome-driven approach for disease genetics prediction, and investigated the translational potential of the predicted genes in drug discovery.

Results

We constructed CSNs by directly connecting diseases with associated phenotypes. Here, we constructed two CSNs using different data sources; the two networks contain 26 790 and 13 822 nodes respectively. We integrated the CSNs with a genetic functional relationship network and predicted disease genes using a network-based ranking algorithm. For comparison, we built Similarity-Based disease Networks (SBN) using the same disease phenotype data. In a de novo cross validation for 3324 diseases, the CSN-based approach significantly increased the average rank from top 12.6 to top 8.8% for all tested genes comparing with the SBN-based approach ( p<e-22 ). The area under the receiver operating characteristic curve for the CSN approach was also significantly higher than the SBN approach (0.91 versus 0.87, p<e-3 ). In addition, we predicted genes for Parkinson's disease using CSNs, and demonstrated that the top-ranked genes are highly relevant to PD pathologenesis. We pin-pointed a top-ranked drug target gene for PD, and found its association with neurodegeneration supported by literature. In summary, CSNs lead to significantly improve the disease genetics prediction comparing with SBNs and provide leads for potential drug targets.

Availability and Implementation

nlp.case.edu/public/data/.

Contact

rxx@case.edu.

Phenome-based gene discovery provides information about Parkinson's disease drug targets

BACKGROUND

Parkinson disease (PD) is a severe neurodegenerative disease without curative drugs. The highly complex and heterogeneous disease mechanisms are still unclear. Detecting novel PD associated genes not only contributes in revealing the disease pathogenesis, but also facilitates discovering new targets for drugs.

METHODS

We propose a phenome-based gene prediction strategy to identify disease-associated genes for PD. We integrated multiple disease phenotype networks, a gene functional relationship network, and known PD genes to predict novel candidate genes. Then we investigated the translational potential of the predicted genes in drug discovery.

RESULTS

In a cross validation analysis, the average rank for 15 known PD genes is within top 0.8 %. We also tested the algorithm with an independent validation set of 669 PD-associated genes detected by genome-wide association studies. The top ranked genes predicted by our approach are enriched for these validation genes. In addition, our approach prioritized the target genes for FDA-approved PD drugs and the drugs that have been tested for PD in clinical trials. Pathway analysis shows that the prioritized drug target genes are closely associated with PD pathogenesis. The result provides empirical evidence that our computational gene prediction approach identifies novel candidate genes for PD, and has the potential to lead to rapid drug discovery.

Insights into Population Health Management Through Disease Diagnoses Networks

The increasing availability of electronic health care records has provided remarkable progress in the field of population health. In particular the identification of disease risk factors has flourished under the surge of available data. Researchers can now access patient data across a broad range of demographics and geographic locations. Utilizing this Big healthcare data researchers have been able to empirically identify specific high-risk conditions found within differing populations. However to date the majority of studies approached the issue from the top down, focusing on the prevalence of specific diseases within a population. Through our work we demonstrate the power of addressing this issue bottom-up by identifying specifically which diseases are higher-risk for a specific population. In this work we demonstrate that network-based analysis can present a foundation to identify pairs of diagnoses that differentiate across population segments. We provide a case study highlighting differences between high and low income individuals in the United States. This work is particularly valuable when addressing population health management within resource-constrained environments such as community health programs where it can be used to provide insight and resource planning into targeted care for the population served.

Phenome-driven disease genetics prediction toward drug discovery

MOTIVATION

Discerning genetic contributions to diseases not only enhances our understanding of disease mechanisms, but also leads to translational opportunities for drug discovery. Recent computational approaches incorporate disease phenotypic similarities to improve the prediction power of disease gene discovery. However, most current studies used only one data source of human disease phenotype. We present an innovative and generic strategy for combining multiple different data sources of human disease phenotype and predicting disease-associated genes from integrated phenotypic and genomic data.

RESULTS

To demonstrate our approach, we explored a new phenotype database from biomedical ontologies and constructed Disease Manifestation Network (DMN). We combined DMN with mimMiner, which was a widely used phenotype database in disease gene prediction studies. Our approach achieved significantly improved performance over a baseline method, which used only one phenotype data source. In the leave-one-out cross-validation and de novo gene prediction analysis, our approach achieved the area under the curves of 90.7% and 90.3%, which are significantly higher than 84.2% (P < e(-4)) and 81.3% (P < e(-12)) for the baseline approach. We further demonstrated that our predicted genes have the translational potential in drug discovery. We used Crohn's disease as an example and ranked the candidate drugs based on the rank of drug targets. Our gene prediction approach prioritized druggable genes that are likely to be associated with Crohn's disease pathogenesis, and our rank of candidate drugs successfully prioritized the Food and Drug Administration-approved drugs for Crohn's disease. We also found literature evidence to support a number of drugs among the top 200 candidates. In summary, we demonstrated that a novel strategy combining unique disease phenotype data with system approaches can lead to rapid drug discovery.

AVAILABILITY AND IMPLEMENTATION

nlp.

CASE

edu/public/data/DMN

Disease Comorbidity Network Guides the Detection of Molecular Evidence for the Link Between Colorectal Cancer and Obesity

Epidemiological studies suggested that obesity increases the risk of colorectal cancer (CRC). The genetic connection between CRC and obesity is multifactorial and inconclusive. In this study, we hypothesize that the study of shared comorbid diseases between CRC and obesity can offer unique insights into common genetic basis of these two diseases. We constructed a comorbidity network based on mining health data for millions of patients. We developed a novel approach and extracted the diseases that play critical roles in connecting obesity and CRC in the comorbidity network. Our approach was able to prioritize metabolic syndrome and diabetes, which are known to be associated with obesity and CRC through insulin resistance pathways. Interestingly, we found that osteoporosis was highly associated with the connection between obesity and CRC. Through gene expression meta-analysis, we identified novel genes shared among CRC, obesity and osteoporosis. Literature evidences support that these genes may contribute in explaining the genetic overlaps between obesity and CRC.