Identification of Disease-Disease Network Communities in Subpopulations of Patients with Prostate Cancer

Prostate cancer (PCa) is ranked as one of the most common cancer diagnoses among men worldwide. Different research studies show that the current diagnosis and treatment strategies have improved the health condition of patients with PCa. Nonetheless, the number of new cases and the morbidity associated with PCa remain high. In this study, using the Medicare claims of patients identified from the SEER-Medicare database, we analyzed the disease-disease interactions in patients at different stages of PCa. Also, we implemented community detection to identify common co-occurring diseases in different subpopulations, identified by PCa stages. The similarity analysis of complication subgroups was performed to identify the most distinct complication subgroups among subpopulations. The results of the study show that the level of experiencing cooccurring diseases is different among subpopulations. Identifying distinct disease-disease interactions can inform the prediction of hospitalization rate and frequency and mortality rate among different subpopulations and enhance our understanding of the pathological correlations among diseases.