Sa1127 Propensity Score Matched Analysis of Outcome of Esophageal Cancer Treatment With Surgery or Definitive Chemoradiotherapy

Abstract

Introduction: Restoration of GI-continuity without compromise of oncological outcome remains the holy grail of rectal cancer surgery. From an oncological perspective, the importance of the circumferential resection margin (CRM) is well recognised. However, it is the distal clearance that determines the potential for restoration of GI-continuity, leading to a progressive reduction in the ‘acceptable' distal resection margin (DRM) to 1cm. In the pursuit of permanent stoma avoidance, GI restoration may, in some cases, be possible if this 1 cm DRM were to be compromised. However, the oncological safety of such a strategy remains uncertain. Therefore, this study aimed to investigate the oncological outcomes of a subcentimetre DRM following rectal cancer surgery. Method: A prospective observational cohort study of consecutive patients undergoing rectal cancer surgery between 1995 to 2011 at a tertiary centre in Sydney, Australia, was performed. The primary outcome measure was mortality, and secondary outcome measures were local and distant disease relapse. Cox regression modelling was performed to assess the impact of important clinicopathological features (including age, tumour stage, tumour height and position, DRM, CRM, neoadjuvant therapy) on outcomes. CRM was regarded as ‘positive' if ≤1 mm. Kaplan-Meier survival analyses were performed to assess the impact of disease relapse on overall survival. Results: Of 935 patients undergoing surgery, 32% and 16% were staged as N-positive and M-positive, respectively. The overall 5-year survival rate was 59%. Local and distant relapse occurred in 12% and 16% of patients, respectively, both of which were associated with worsened survival on Kaplan-Meier analyses. A sub-centimetre DRM was identified in 3% of patients and a positive CRM in 13%. A sub-centimetre DRM did not increase local relapse (P=0.62) or worsen survival (P=0.91), but patients with positive CRM demonstrated increased rates of local and distant relapse and higher mortality (P,0.001). Associations between outcome measures and other clinico-pathological features studied are presented as forest plots (Figs 1-2). Conclusion: In the relatively small number of cases in whom a sub-centimetre DRM was achieved, there was no increase in risk of local relapse or worsened overall survival. This suggests that sub-centimetre clearance may be accepted in selected cases, if required, to permit restoration of GI-continuity. In addition, this study adds support that a positive CRM and increased tumour stage are strongly predictive of disease relapse and increased mortality.