BACKGROUND: Adenocarcinoma of the gastro-oesophageal junction is considered a distinct clinical entity, although the current pathological tumour node metastasis (pTNM) classification does not consider this tumour specifically. A prospective study was undertaken to determine the prognostic importance of lymph node involvement in adenocarcinoma of the gastro-oesophageal junction, analysing both a number- and site-based classification, in order to develop a clinically useful nodal staging system. METHODS: Two classification systems were analysed in 116 patients who underwent resection for adenocarcinoma of the gastro-oesophageal junction from January 1988 to August 2001. The Cox regression model was used to evaluate the prognostic significance of the site and number of positive nodes. RESULTS: The number- and site-based staging systems coincided only in 42 (56 per cent) of 75 patients; in particular, the old pN1 classification was upstaged in 13 of 41 patients and the old pN2 was downstaged in 13 of 34 patients. Lymph node involvement was the most important prognostic factor in both classifications (P < 0.001). The risk of death was significantly influenced by the site of nodal metastasis among patients with a similar number of involved nodes (relative risk with respect to pN0: 2.18 for pN1 with one to six nodes; 6.53 for pN2 with one to six nodes; 7.53 for pN1 with more than six nodes; 39.13 for pN2 with more than six nodes). CONCLUSION: Adenocarcinoma of the gastro-oesophageal junction requires a specific lymph node classification which should take into account both the number and site of nodal metastases. Copyright 2004 British Journal of Surgery Society Ltd.
BACKGROUND:Adenocarcinoma of the gastro-oesophageal junction is considered a distinct clinical entity, although the current pathological tumour node metastasis (pTNM) classification does not consider this tumour specifically. A prospective study was undertaken to determine the prognostic importance of lymph node involvement in adenocarcinoma of the gastro-oesophageal junction, analysing both a number- and site-based classification, in order to develop a clinically useful nodal staging system. METHODS: Two classification systems were analysed in 116 patients who underwent resection for adenocarcinoma of the gastro-oesophageal junction from January 1988 to August 2001. The Cox regression model was used to evaluate the prognostic significance of the site and number of positive nodes. RESULTS: The number- and site-based staging systems coincided only in 42 (56 per cent) of 75 patients; in particular, the old pN1 classification was upstaged in 13 of 41 patients and the old pN2 was downstaged in 13 of 34 patients. Lymph node involvement was the most important prognostic factor in both classifications (P < 0.001). The risk of death was significantly influenced by the site of nodal metastasis among patients with a similar number of involved nodes (relative risk with respect to pN0: 2.18 for pN1 with one to six nodes; 6.53 for pN2 with one to six nodes; 7.53 for pN1 with more than six nodes; 39.13 for pN2 with more than six nodes). CONCLUSION:Adenocarcinoma of the gastro-oesophageal junction requires a specific lymph node classification which should take into account both the number and site of nodal metastases. Copyright 2004 British Journal of Surgery Society Ltd.