Developing a CTCAE patient questionnaire for late toxicity after head and neck radiotherapy

Purpose:
Patient-based reporting of symptoms is increasingly important in providing treatment toxicity information. However, observer-based scoring systems such as the CTCAEs which incorporate the LENT-SOMA scales are not adapted for patient-based reporting. We aim to (1) report the late toxicity in patients following head and neck radiotherapy using a LENT-SOMA patient-based questionnaire, (2) describe how the responses help to improve the questionnaire and (3) adapt the questionnaire for patient reporting using CTCAEs.

Methods:
A 31-item LENT-SOMA patient questionnaire was administered prospectively to 220 patients pre-treatment and at eight time periods post-radical head and neck radiotherapy over 3 years. Exploratory factor analysis was carried out and questionnaire reliability was evaluated using Cronbach's a coefficient.

Results:
At 3-years follow-up, grade 3/4 toxicity was recorded for xerostomia (44%), hoarseness (14.3%), altered taste (6.1%) and oropharyngeal pain (1.9%). Factor analysis indicated that questionnaire division according to anatomical sub-site was reasonable. Cronbach's alpha was 0.851 (95% CI: 0.820-0.883) indicating high reliability. Good compliance was obtained with all questions except for the 'weight loss' item. A satisfaction survey showed that the questionnaire was clear and concise. Teeth and mandible sections have been removed. Dietary change due to xerostomia has been incorporated in line with CTCAEs. LENT-SOMA scoring of analgesic needs and dysphagia not described in CTCAEs were found useful and have been retained.

Conclusions:
The questionnaire has enabled reporting of late toxicity and the responses have enabled refinement of the questionnaire. It is reliable, feasible and has been validated for patient-based collection of CTCAEs late toxicity data.