A network medicine approach to study comorbidities in heart failure with preserved ejection fraction

Abstract

Background:
Comorbidities are expected to impact pathophysiology of heart failure (HF) with preserved ejection fraction (HFpEF). However, comorbidity profiles are usually reduced to a few comorbid disorders. Systems medicine approaches can model multiple comorbidities to help to improve our understanding of HFpEF.

Methods:
We retrospectively explored 569 comorbidities documented in 29,074 HF patients, including 8,062 HFpEF and 6,585 heart failure with reduced ejection fraction (HFrEF) patients from a German university hospital. To analyze comorbidity profiles, first we assessed differences between comorbidity profiles by HF subtype. We fit machine learning classifiers to discriminate between HFpEF and HFrEF patients to interpret important comorbidities. Then we built a network to represent comorbid relationships in HF patients (HFnet) and identified main disease clusters. Lastly, we performed novel gene candidate prediction for HFpEF by linking the HFnet to a multi-layer gene network by integrating multiple databases (HFhetnet). To corroborate HFpEF candidate genes, we collected transcriptomic data in a murine HFpEF model. We compared predicted genes with the murine disease signature as well as with the literature.

Results:
We found that comorbidity profiles vary more in association with HFpEF/HFrEF status than with age or sex. The comorbidities present in HFpEF patients were more diverse than those in HFrEF, and included neoplastic, osteologic and rheumatoid disorders, among others. Disease communities in the HFnet captured important comorbidity concepts of HF patients which could be assigned to HF subtypes, age groups and sex. Within the HFhetnet, we predicted novel gene candidates, including genes involved in fibrosis (COL3A1, MMP1, LOX), oxidative stress (NOS1, GSST1, XDH) and endoplasmic reticulum stress (ATF6). Finally, predicted genes were significantly overrepresented in the murine transcriptomic disease signature.

Conclusions:
We applied systems medicine concepts to analyze comorbidity profiles in a HF patient cohort. We were able to identify disease clusters that helped to characterize HF patients. We derived a distinct comorbidity profile for HFpEF, which was leveraged to suggest novel candidate genes via network propagation. The identification of distinctive comorbidity profiles and candidate genes from routine clinical data will provide insights that may be leveraged to improve diagnosis and treatment for HFpEF patients.

Analysis Workflow

Utilitary functions

1. Multiple Correspondence Analysis
   MCA

2. Patient Classifier
   HFpEF_HFrEF_classifier
   feature_forward_selection time_point_and_time_to_HF

3. HFnet Construction and Analysis
   test_comorbidities_between_HFpEF/HFrEF
   construct_HFnet
   cluster_HFnet
   HFnet_graph_metrics Disease_network_comparison Patient_cohort_and_disease_cluster_comparison

4. HFhetnet Construction and Analysis
   Database_processings
   HFhetnet_construction
   HFhetnet_graph_structure
   HFhetnet_randomization
   LOO_CV_disease_gene_prediction HFhetnet_gene_prediction

5. RNAseq
   processing_and_gene_annotations
   Funcomics
   Differntial_expression_analysis

Citation

Jan David Lanzer, Alberto Valdeolivas, Mark Pepin et al. A Network medicine approach to study comorbidities in heart failure with preserved ejection fraction, 11 January 2023, PREPRINT (Version 1) available at Research Square [https://doi.org/10.21203/rs.3.rs-2429581/v1]

Please find our preprint

Funded by Informatics for Life

A workflowr project.