766. LASTING SYMPTOMS AFTER OESOPHAGEAL RESECTIONAL SURGERY (LASORS): MULTICENTRE VALIDATION COHORT STUDY

Background As outcomes following curative-intent oesophageal cancer treatment improve, more patients are living with the long-term morbidity of cancer therapies, the symptoms of which are poorly recognised and understood. Existing symptomatology and quality-of-life tools are cumbersome and moreover are not designed specifically for evaluation of post-treatment survivorship. The LASER study and subsequent consultation process identified six key symptoms thought to predict poor health-related quality-of-life (HRQoL) as measured by validated European Organisation for Research and Treatment of Cancer (EORTC) tools. The current study aimed to validate this six-symptom LAsting Symptoms after Oesophageal Resection (LASOR) clinical tool, and assess its clinical utility. Methods In this multi-centre cohort study, patients who underwent oesophagectomy between January 2015 and June 2019 across the United Kingdom, and were disease-free at least one-year post-treatment, were asked to complete LASOR, EORTC-QLQ-C30 and QLQ-OG25 questionnaires. LASOR symptoms (low mood, reduced energy, thoracotomy pain, heartburn, diarrhoea, and bloating after eating) were evaluated using composite scores based on frequency and quality of life impact, and correlated with EORTC HRQoL scores. The ability of the LASOR tool to predict patients with poor HRQoL was validated using receiver operating characteristic (ROC) curve analysis. Patient acceptability of the tool was assessed using a separate questionnaire. Results 263 participants were included. 192 (73%) were males. Surgical technique included open (61%), hybrid (18%), and minimally-invasive (21%). 203 patients (77%) received neoadjuvant chemotherapy or chemoradiotherapy. 148 patients (56%) experienced a post-operative complication. Four LASOR symptoms were associated with significantly lower HRQoL: reduced energy (OR=2.13; 95% CI 1.20-2.87), low mood (OR=1.86; 95% CI 1.45–3.12), diarrhoea (OR=1.48; 95% CI 1.06–2.06), and bloating (OR=1.35; 95%CI 1.03-1.77). In combination, the LASOR symptoms produced an area under the ROC curve of 0.85 (sensitivity = 0.82, specificity = 0.73). Conclusion The six-symptom LASOR tool generated a reliable model for identification of patients with a poor HRQoL, with an overall diagnostic accuracy of over 80%. This is the first clinical symptom tool to be validated in the post-curative-treatment setting for patients with oesophageal cancer. The LASOR tool is straightforward to administer and highly acceptable to patients, and can be used to identify those at risk of high morbidity and poor quality of life in surveillance programmes. In turn, evidence-based survivorship services can be integrated into patient follow-up to alleviate the burden of cancer treatment on oesophagectomy patients.