Concordance between clinician-reported toxicities and patient-reported symptom severity in early-stage breast cancer: an analysis of the randomized phase III PANTHER trial

BACKGROUND: Although advances in systemic therapy have markedly improved survival in breast cancer, treatment-related toxicities remain a major determinant of health-related quality of life (HRQoL). Clinician reports may not fully capture the symptom burden experienced by patients. This study examined the concordance between clinician-reported toxicities and patient-reported symptom scores during adjuvant chemotherapy for early-stage, high-risk breast cancer in a randomized phase III clinical trial. PATIENTS AND METHODS: Clinician-reported toxicities were graded using the Common Terminology Criteria for Adverse Events (CTCAE) version 3.0, and patient-reported symptom scores were obtained with the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30) and EORTC QLQ-BR23 questionnaires at mid- and end-of-treatment. Analyses focused on six symptomatic domains with conceptual overlap between CTCAE and EORTC measures: diarrhea, nausea, vomiting, mucositis, pain, and fatigue. Concordance between clinician- and patient-reported scores was evaluated using weighted Cohen's kappa coefficients. RESULTS: A total of 1566 patients with available CTCAE and HRQoL data were included in the analysis. Agreement between clinician-reported toxicities and patient-reported symptom scores was low across all symptoms and time points, with kappa values indicating no to minimal concordance (kappa = 0.07-0.34). Clinician-assigned CTCAE grades were predominantly lower than patient-reported symptom severity. The highest discrepancy rates were observed for fatigue (36%-54%) and pain (23%-38%) across treatment arms and time points. Except for nausea and vomiting, the magnitude of the discrepancy increased from mid- to end-of-treatment in both study arms. Binary analyses comparing the presence of any versus no symptom confirmed persistently low agreement between clinicians and patients (kappa = 0.03-0.35). CONCLUSION: Clinician-reported toxicities showed poor concordance with patient-reported symptom severity, and the discrepancy increased over time. Incorporating patient-reported assessments into both clinical trials and routine care is essential to ensure more accurate evaluation of treatment tolerability.