No abstract

BackgroundIn esophageal cancer, circumferential resection margins (CRMs) are considered to be of relevant prognostic value, but a reliable definition of tumor-free CRM is still unclear. The aim of this study was to appraise the clinical prognostic value of microscopic CRM involvement and to determine the optimal limit of CRM.MethodsTo define the optimal tumor-free CRM we included 98 consecutive patients who underwent extended esophagectomy with microscopic tumor-free resection margins (R0) between 1997 and 2006. CRMs were measured in tenths of millimeters with inked lateral margins. Outcome of patients with CRM involvement was compared with a statistically comparable control group of 21 patients with microscopic positive resection margins (R1).ResultsA cutoff point of CRM at ≤1.0 mm and >1.0 mm appeared to be an adequate marker for survival and prognosis (both P < 0.001). The outcome in patients with CRMs ≤1.0 and >0 mm was equal to that in patients with CRM of 0 mm (P = 0.43). CRM involvement was an independent prognostic factor for both recurrent disease (P = 0.001) and survival (P < 0.001). Survival of patients with positive CRMs (≤1 mm) did not significantly differ from patients with an R1 resection (P = 0.12).ConclusionInvolvement of the circumferential resection margins is an independent prognostic factor for recurrent disease and survival in esophageal cancer. The optimal limit for a positive CRM is ≤1 mm and for a free CRM is >1.0 mm. Patients with unfavorable CRM should be approached as patients with R1 resection with corresponding outcome.

In a population-based study in the Netherlands, we found patients with stable NDBE to have a low risk of progression to HGD or EAC. These findings indicate that surveillance intervals might be lengthened or even discontinued in subgroups patients with persistent NDBE.

Loss of CD44 expression and loss of SOX2 expression are prognostic factors of poor survival in EAC patients. This suggests a role of these proteins in EAC that requires further investigation.