Constructing cancer-specific patient similarity network with clinical significance

Clinical molecular genetic testing and molecular imaging dramatically increase the quantity of clinical data. Combined with the extensive application of electronic health records, medical data ecosystem is forming, which summons big-data-based medicine model. We tried to use big data analytics to search for similar patients in a cancer cohort and to promote personalized patient management. In order to overcome the weaknesses of most data processing algorithms that rely on expert labelling and annotation, we uniformly adopted one-hot encoding for all types of clinical data, calculating Euclidean distance to measure patient similarity, and subgrouping via unsupervised learning model. Overall survival was investigated to assess the clinical validity and clinical relevance of the model. Thereafter, we built a high-dimensional network cPSN (clinical patient similarity network). When performing overall survival analysis, we found Cluster_2 had the longest survival rates while Cluster_5 had the worst prognosis among all subgroups. Because patients in the same subgroup share some clinical characteristics, clinical feature analysis found that Cluster_2 harbored more lower distal GCs than upper proximal GCs, shedding light on the debates. Overall, we constructed a cancer-specific cPSN with excellent interpretability and clinical significance, which would recapitulate patient similarity in the real-world. The constructed cPSN model is scalable, generalizable, and performs well for various data types. The constructed cPSN could be used to accurately "locate" interested patients, classify the patient into a disease subtype, support medical decision making, and predict clinical outcomes.