450 FACTORS AFFECTING CIRCUMFERENTIAL RESECTION MARGIN AFTER ESOPHAGECTOMY AND ITS PROGNOSTIC SIGNIFICANCE-LARGE SINGLE CENTRE EXPERIENCE

Abstract

Abstract The evidence regarding the importance of circumferential resection margin (CRM) as a prognostic factor after esophagectomy is inconclusive in the era of neoadjuvant therapy. We retrospectively analysed our prospectively maintained database for factors that affect CRM positivity, and whether a positive CRM affects event free and overall survival. 2843 patients underwent esophagectomy with curative intent from October 2004 to 2019 at our centre. CRM was analysed as negative, close but technically free (<1 mm) and involved. Methods Data on the following variables was retrospectively extracted from prospective database. CRM status was noted for clinic-radiological T and N stage, level of growth, histology, differentiation grade and neoadjuvant treatment. Intra-operative details such as surgical procedure, approach, surgeon grade, lymphadenectomy and resection status were analysed. On final histopathology; proximal and distal margins, lymph node positivity, lymphovascular invasion(LVI), tumour regression grade(TRG) were analysed. The effect of CRM on development of recurrence and overall survival was evaluated. CRM data was available for 2439 (85.78%) patients. 71.2% of the patients received neoadjuvant chemotherapy. Factors were analysed separately for both close and positive margins. Results 75.8% had negative, 15.6% close and 8.6% positive CRM. Univariately, T stage, adenocarcinoma, poor differentiation, transhiatal approach, R+ resection, positive margins, TRG > 3, LVI and upfront surgery predicted positive CRM. On multivariate, negative CRM was seen in T1/T2 stage [OR 0.325, 95% CI-0.144-0.732, p = 0.007], squamous carcinoma [OR 0.574, 95% CI-0.351-0.958, p = 0.027], R0 resection [OR 0.228, 95% CI-0.086-0.599, p = 0.003] while positive CRM was seen in upfront surgery [OR 2.32, 95% CI-1.55-3.46, p < 0.001], positive nodes [OR 1.748, 95% CI-1.19-2.56, p = 0.004] and LVI [OR 2.73, 95% CI-1.87-3.98, p < 0.001]. Median event-free survival in CRM negative was 64 months compared to 14 months in CRM positive (p < 0.001). Conclusion Positive CRM involvement is a prognostic indicator in patients undergoing esophagectomy and associated with worse event-free and overall survival. CRM-positive disease in esophageal cancer may represent residual tumor, advanced disease, aggressive biology, or poor response to neoadjuvant treatment. All attempts should be made to achieve a clear circumferential resection margin. More evidence is needed to evaluate if adjuvant therapy is justified in these patients and the type of therapy also needs to be determined.