Prognostic value of nodal micrometastases in patients with cancer of the gastro-oesophageal junction

Aims Aim of this study was to examine the presence and the prognostic impact of immunohistochemically identified nodal micrometastases in patients with gastro-oesophageal junction (GEJ) carcinomas. Methods Between January 1988 and December 2000, 148 patients underwent a radical (R0) resection with a two-field lymphadenectomy for a GEJ carcinoma. Specimens of 60 patients in whom conventional haematoxylin and eosin (H & E) examination did not demonstrate lymph-node metastases (pN0) were available for immunohistochemical (IHC) analysis using antibodies AE1/AE3 directed against cytokeratins. Paraffin embedded material of all retrieved lymph nodes in these patients were serially sectioned and analysed by one pathologist after H & E examination for the presence of micrometastases by IHC. Results In 60 resection specimens initially staged as pN0 a total of 524 lymph nodes were available for IHC analyses. Micrometastases were detected in 126 out of 524 lymph nodes (24%), corresponding with 18 of the 60 patients (30%) who were upstaged by this technique. Compared with the pN0 group, the disease free survival (DFS) was significantly lower in patients with nodal involvement at IHC (p<0.001). Survival of patients with IHC identified micrometastatic disease was comparable to those with H & E positive lymph nodes. Conclusions Micrometastases in regional nodes were detected by cytokeratin-specific IHC in 30% of radical resected GEJ tumours without overt nodal involvement. Their presence conveys a worse prognosis with a significant reduced DFS, suggesting that the finding of micrometastases should be included in the staging system.