Abstract:Cancer patients' symptom experiences are complex and exhibit a large amount of interindividual variability. This clinical reality poses major challenges in the prioritization of symptom management interventions. This study is the first to use computational network analysis and unsupervised clustering approaches with a sample of 3088 cancer patients to model symptom interconnectedness, group patients based on shared symptom patterns, and identify core symptoms to target in each group. Background Interindividual variability in oncology patients' symptom experiences poses significant challenges in prioritizing symptoms for targeted intervention(s). In this study, computational approaches were used to unbiasedly characterize the heterogeneity of the symptom experience of oncology patients to elucidate symptom patterns and drivers of symptom burden. Methods Severity ratings for 32 symptoms on the Memorial Symptom Assessment Scale from 3088 oncology patients were analyzed. Gaussian Graphical Model symptom networks were constructed for the entire cohort and patient subgroups identified through unsupervised clustering of symptom co‐severity patterns. Network characteristics were analyzed and compared using permutation‐based statistical tests. Differences in demographic and clinical characteristics between subgroups were assessed using multinomial logistic regression. Results Network analysis of the entire cohort revealed three symptom clusters: constitutional, gastrointestinal‐epithelial, and psychological. Lack of energy was identified as central to the network which suggests that it plays a pivotal role in patients' overall symptom experience. Unsupervised clustering of patients based on shared symptom co‐severity patterns identified six patient subgroups with distinct symptom patterns and demographic and clinical characteristics. The centrality of individual symptoms across the subgroup networks differed which suggests that different symptoms need to be prioritized for treatment within each subgroup. Age, treatment status, and performance status were the strongest determinants of subgroup membership. Conclusions Computational approaches that combine unbiased stratification of patients and in‐depth modeling of symptom relationships can capture the heterogeneity in patients' symptom experiences. When validated, the core symptoms for each of the subgroups and the associated clinical determinants may inform precision‐based symptom management.

Defining the Symptom Cluster: How Far Have We Come?

The Elusive Concept of the Symptom Cluster

Future Directions in Symptom Cluster Research

The Concept of Symptom Cluster

Symptom Cluster Research: Conceptual, Design, Measurement, and Analysis Issues

Symptom Clusters: The New Frontier in Symptom Management Research

Advancing Symptom Science Through Symptom Cluster Research: Expert Panel Proceedings and Recommendations

Cancer Symptom Clusters: Old Concept But New Data

Symptom Cluster Patterns During the First Year After Diagnosis with Cancer

Characteristics of Cancer Symptom Clusters Reported through a Patient-Centered Symptom Cluster Assessment

Cancer Symptom Cluster Management

Symptom Network Analysis and Unsupervised Clustering of Oncology Patients Identifies Drivers of Symptom Burden and Patient Subgroups With Distinct Symptom Patterns

Symptom clusters experienced by breast cancer patients at various treatment stages: A systematic review

Evidence-Based and Person-Centered Symptom Cluster Management in Cancer Care: The Value of Symptom Theory

Symptom clustering in advanced cancer

Research progress on symptom clusters

Common Biological Pathways Underlying the Psychoneurological Symptom Cluster in Cancer Patients

Prospective Longitudinal Evaluation of a Symptom Cluster in Breast Cancer

Cancer symptom cluster research in pediatric oncology: a work in progress