Abstract

Purpose Information about symptomatic toxicities of anticancer treatments is not based on direct report by patients, but rather on reports by clinicians in trials. Given the potential for under-reporting, our aim was to compare reporting by patients and physicians of six toxicities (anorexia, nausea, vomiting, constipation, diarrhea, and hair loss) within three randomized trials. Patients and Methods In one trial, elderly patients with breast cancer received adjuvant chemotherapy; in two trials, patients with advanced non–small-cell lung cancer received first-line treatment. Toxicity was prospectively collected by investigators (graded by National Cancer Institute Common Toxicity Criteria [version 2.0] or Common Terminology Criteria for Adverse Events [version 3]). At the end of each cycle, patients completed the European Organisation for Research and Treatment of Cancer quality-of-life questionnaires, including toxicity-related symptom items. Possible answers were “not at all,” “a little,” “quite a bit,” and “very much.” Analysis was limited to the first three cycles. For each toxicity, agreement between patients and physicians and under-reporting by physicians (ie, toxicity reported by patients but not reported by physicians) were calculated. Results Overall, 1,090 patients (2,482 cycles) were included. Agreement between patients and physicians was low for all toxicities. Toxicity rates reported by physicians were always lower than those reported by patients. For patients who reported toxicity (any severity), under-reporting by physicians ranged from 40.7% to 74.4%. Examining only patients who reported “very much” toxicity, under-reporting by physicians ranged from 13.0% to 50.0%. Conclusion Subjective toxicities are at high risk of under-reporting by physicians, even when prospectively collected within randomized trials. This strongly supports the incorporation of patient-reported outcomes into toxicity reporting in clinical trials.