1192 - Longitudinal Trajectories, Quality Of Life Impact, and Intervention Thresholds of Radiation-Induced Adverse Events in Head And Neck Cancer: An ePRO-Based Cohort Study
Presenter(s)
Y. Jiang1, X. Zhang1, R. Huang2, and P. Zhang3; 1Department of Radiation Oncology, Sichuan Clinical Research Center for Cancer, Sichuan Cancer Hospital & Institute, Sichuan Cancer Center, Affiliated Cancer Hospital of University of Electronic Science and Technology of China, Chengdu, Sichuan, China, 2Sichuan Clinical ResearchCenter for Cancer, Sichuan Cancer Hospital & Institute, SichuanCancerCenter, University ofElectronic Science and Technology of China, Chengdu, China, 3Department of Radiation Oncology, Radiation Oncology Key Laboratory of Sichuan Province, Sichuan Cancer Hospital & Institute, Sichuan Cancer Center, Affiliated Cancer Hospital of University of Electronic Science and Technology of China, Chengdu, China
Purpose/Objective(s): Radiation-induced adverse events (AEs) significantly impair the quality of life (QoL) of patients with head and neck cancer (HNC). Using an electronic patient-reported outcomes (ePRO) system, this study prospectively characterizes the dynamic evolution of symptoms during radiotherapy, aiming to construct predictive models and identify optimal quantitative thresholds for clinical interventions.
Materials/Methods: Patients with HNC receiving radical radiotherapy (prescription dose: 66-70 Gy) were prospectively enrolled. An ePRO system was used to capture symptom burden and QoL using the MDASI-HN and the EQ-5D during and after radiotherapy. Linear mixed-effects models (LMM) were constructed to evaluate the relationship between symptom dynamics and primary tumor sites. Multivariable analysis quantified the impact of symptom clusters on QoL during the peak AE phase. Joint predictive models based on symptom clusters determined optimal thresholds for nutritional and analgesic interventions.
Results: A total of 181 patients with HNC (including nasopharyngeal carcinoma) receiving radical radiotherapy were enrolled, yielding 1,823 complete ePRO assessments. LMM analysis demonstrated that, after adjusting for baseline confounders including age, TNM stage, and treatment modality, the longitudinal evolution of core symptoms, such as dysphagia and oral mucositis, maintained significant heterogeneity across primary tumor sites (P < 0.001). Concurrently, multivariable analysis indicated that during the peak AE phase (weeks 4-7), each 1-point increase in the mucosal pain cluster independently resulted in a QoL (EQ-VAS) decrement of 2.08 points (P < 0.001). Fatigue, dry mouth, and older age emerged as additional independent risk factors for QoL deterioration (all P < 0.05). Furthermore, regarding the prediction of strong opioid initiation, the joint predictive performance of the mucosal pain cluster was superior to that of the single pain score (AUC: 0.938 vs. 0.914); similarly, the eating symptom cluster outperformed the single dysphagia score in predicting non-oral nutritional interventions (AUC: 0.894 vs. 0.872). The study established that a single pain/dysphagia score = 5, or a model-derived joint risk probability of 0.87 (analgesic threshold) and 0.71 (nutritional threshold), served as optimal cutoffs for triggering clinical interventions.
Conclusion: Longitudinal ePRO monitoring highlighted the site-specific heterogeneity of acute symptom evolution during HNC radiotherapy. The multidimensional symptom cluster predictive model not only identified the core risk factors driving QoL deterioration but also established intervention thresholds for nutrition and pain management. These quantitative criteria provide an evidence-based foundation for the proactive management of early radiation AEs.