Validity and responsiveness of patient-reported outcomes common terminology criteria for adverse events (PRO-CTCAE) among Italian lung cancer patients

BACKGROUND: Patient-reported symptom monitoring may improve the outcomes of cancer treatment. Knowledge of psychometric properties may drive the selection of optimal tools. Here we report the analysis of PRO-CTCAE in Italian lung cancer patients. METHODS: A PRO-CTCAE list (74 items, 44 symptoms) was administered to on-treatment patients at registration (visit 1) and after 2-6 weeks (visit 2). Convergent validity with EORTC LC-13 and Hospital Anxiety and Depression Scale (HADS) was assessed by Pearson correlation. Known-group validity and responsiveness were tested with ECOG performance status (PS 0 vs 1) and Patients' Global Impression of Change (PGIC) scale as anchors, respectively. RESULTS: From September 2019 to November 2022, 186 patients filled PRO-CTCAE at visit 1 and 164 at visit 2. Median age (IQR) was 66.4 (60.9 - 73.1) years; 58.6% were male; 68.0% had PS 0. Convergent validity showed high correlation (r ≥ 0.50) of PRO-CTCAE hair loss (r = 0.78), dyspnoea (0.63), limb numbness (0.63), cough (0.58) and pain (0.51) with selected LC-13 anchors. PRO-CTCAE anxiety, depression and sadness highly correlated with HADS. For known-group validity, small to medium correlations were found for oral, cutaneous and respiratory toxicity, fatigue, appetite loss, insomnia and depression, that were worse for PS 1; limb swelling and pain were worse in PS 0. Responsiveness analysis at visit 2 found some PRO-CTCAE score changes consistently and directly correlated with impression of change measured with PGIC. CONCLUSIONS: These results support the use of selected PRO-CTCAE items for lung cancer patients as a reliable tool for detection of treatment related toxicities.