Abstract
There is an increasing need to measure treatment-related side effects in normal tissues following cancer therapy. The ALERT-B (Assessment of Late Effects of RadioTherapy - Bowel) questionnaire is a screening tool that is composed of four items related specifically to bowel symptoms. Those patients that respond with a "yes" to any of these items are referred on to gastroenterologist in order to improve the long-term consequences of these side effects of radiological treatment. Here we wish to test the ability of this questionnaire to identify these subsequent gastroenterological complications by tracking prostate cancer patients that were positive with respect to ALERT-B. We also carry out receiver-operator curve (ROC) analysis for baseline data for an overall ALERT-B questionnaire score with respect to subscale data for the Gastrointestinal Symptom Rating Scale (GSRS) and the Expanded Prostate Cancer Index Composite (EPIC-26) questionnaire. 84.4% and 95.7% of patients identified by the ALERT-B questionnaire demonstrated complications diagnosed at 6 and 12 months post-treatment, respectively. ROC curve analysis of baseline data showed that ALERT-B detected clinically relevant levels of side effects established at baseline by the GSRS diarrhoea subscale (AUC = 0.867, 95% CI = 0.795 to 0.926) and at the minimally important level of side effects for the EPIC bowel subscale (AUC = 0.765, 95% CI = 0.617 to 0.913). These results show that ALERT-B provides a simple and effective screening tool for identifying gastroenterological complications after treatment for prostate cancer.