Grading xerostomia by physicians or by patients after intensity-modulated radiotherapy of head-and-neck cancer

Purpose: To assess observer-based vs. patient self-reported scoring of xerostomia after intensity-modulated radiotherapy (IMRT) of head-and-neck (HN) cancer. Methods: A total of 38 patients who had received IMRT for HN cancer underwent xerostomia evaluations 6 to 24 months after completion of therapy using three methods each time: (1) Grading by 3 observers according to the Radiotherapy Oncology Group/European Organization for Research and Therapy of Cancer (RTOG/EORTC) system; (2) patient self-reported validated xerostomia questionnaire (XQ); and (3) major salivary gland flow measurements. Results: The interobserver agreement regarding the RTOG/EORTC grades was moderate: κ-coefficient 0.54 (95% CI = 0.31-0.76). The correlations between the average RTOG/EORTC grades and the salivary flow rates were not statistically significant. A trend for significant correlation was observed between these grades and the percent (relative to the pretherapy) nonstimulated salivary flow rates (p = 0.07), but not with the percent stimulated flow rates. Better correlations were found between grading made more than the median time (15 min) after the last liquid sipping and the nonstimulated (but not the stimulated) flows compared with grading made shortly after sipping. In contrast, significant correlations were found between the XQ scores and the nonstimulated (p < 0.005) and the stimulated (p < 0.005) salivary flow rates, as well as with the percentages of the corresponding pretherapy values (p = 0.002 and 0.038, respectively). No significant correlation was found between the RTOG/EORTC grades and the XQ scores. The observer-based grades underestimated the severity of xerostomia compared with the patient self-reported scores. Conclusions: Patient self-reported, rather than physician-assessed scores, should be the main end points in evaluating xerostomia.