A study on the temporal network characteristics of early post-chemotherapy core symptoms in colorectal cancer patients based on ecological momentary assessment

Background Patients with colorectal cancer (CRC) often experience multiple concurrent symptoms during the early post-chemotherapy period, and both symptom severity and inter-symptom associations may change over time. Previous low-frequency assessments may not adequately capture short-term fluctuations. Objective To characterize the dynamic trajectories and network features of early symptoms in patients receiving postoperative adjuvant chemotherapy for CRC and to identify candidate symptoms for enhanced monitoring. Methods A prospective longitudinal study was conducted. Ecological momentary assessment (EMA) was performed three times daily for seven consecutive days after infusion in the first and second chemotherapy cycles and covered 13 symptoms. Linear mixed-effects models characterized symptom trajectories, and a multilevel vector autoregressive (mlVAR) model estimated temporal and contemporaneous networks. Edge uncertainty, centrality stability, sensitivity analyses, and between-cycle differences were evaluated. Results A total of 110 patients contributed 4,134 valid assessments (completion rate, 89.5%). All 13 symptoms showed significant time effects. Fatigue, loss of appetite, gastrointestinal symptoms, and sleep disturbance fluctuated, whereas numbness/tingling and depression showed cumulative trends. Loss of appetite, dry mouth, and nausea had comparatively high outgoing temporal associations. The strongest individual temporal edge was modest (standardized coefficient approximately 0.12). No statistically significant between-cycle differences were found in overall network structure or global strength. Conclusion Early post-chemotherapy symptoms showed dynamic changes and a broadly comparable network structure across the first two cycles. Loss of appetite, dry mouth, and nausea may be candidates for enhanced monitoring and future intervention research; the observational findings do not establish causal effects or intervention efficacy.