68-4

44/136

364pathological degree of differentiation is unknown. One study showed more micrometastases in well differentiated esophageal carcinoma5), but another found opposite results8), and currently, no conclu-sion has been reached. Therefore, we examined this relationship and that of micrometastasis to lymph nodes with prognosis in patients with esoph-ageal SCC, including cases of advanced cancer.This study was approved by the clinical research committee of Juntendo University Hospital (E21- 0056-H01). Informed consent was not required because of the retrospective study design. Twen-ty-five patients with esophageal cancer underwent esophagectomy in our department at Juntendo University Medical School from January to December 2000. The inclusion criteria were [1] SCC, [2] no preoperative chemotherapy, [3] three-field lymphadenectomy, [4] completion of follow-up, and [5] complete R0 resection. The exclusion criteria were >10 metastases to lymph nodes detected by conventional HE stain. SurgeryAll patients underwent esophagectomy with three-field lymphadenectomy by right thora-cotomy and laparotomy, as reported previously9). All patients also underwent lymphadenectomy along the bilateral recurrent laryngeal nerves and around the supraclavicular area. The gastric tube was pulled up through the retrosternal route and a hand-sewn esophagogastrostomy was created in the neck.Postoperative adjuvant therapyPatients who were pathologically confirmed to have ≥3 lymph node metastases received two courses of postoperative adjuvant therapy with docetaxel, cisplatin and 5-fluorouracil10).ImmunostainingFor histopathological examination of resected lymph nodes, samples were subjected to formalin fixation and paraffin embedding. Five neighboring tissue sections (3 μm) were prepared from each slice and 3 central sections were used for staining. After deparaffinizing each section, antigen retrieval was carried out by autoclave treatment (1.2 atm, 10 min). Immunostaining was performed using anti-cytokeratin13 Ks 13.1 mouse monoclonal anti-body (CK13, American Research Products) diluted 20 times with antibody diluent (1% BSA/PBS) using automated staining equipment (Ventana NX System, Ventana). After immunostaining, back-ground staining was performed with Mayer’s hematoxylin for microscopic examination. Of the cells within the lymph node capsules, nucleated cells with evenly cytokeratin-stained cytoplasm were defined as metastasis-positive.Clinicopathologic parametersData for clinicopathologic parameters, including tumor stage by TNM staging (ver. 8)11), were obtained retrospectively from a hospital database. TNM staging reflects the result of HE stain. The degree of differentiation was evaluated using most parts of the tumor.Statistical analysis Comparisons of two groups were performed by chi-square test. Survival curves were estimated with the Kaplan-Meier method and significant differences in survival rate were analyzed by log-rank test. Multivariate analysis was performed using logistic regression analysis for categorical data and Cox regression analysis for survival. Significance was defined as P<0.05 in all analyses. Due to the small number of cases, a result with P<0.2 was considered not to be significant, but to show a tendency. All calculations were performed using IBM SPSS Statistics ver. 23.0.The characteristics of the 25 patients in the study are shown in Table 1. The most common disease was middle thoracic esophageal cancer, pT3, pN1, pStage III. There were 14 (56%) well-dif-ferentiated, 11 (44%) moderately differentiated, and no poorly differentiated cases. Intramural metastasis was found in resected specimens in 4 patients.Conventional lymph node metastasesA total of 2,915 lymph nodes were collected from the 25 patients. The median number of lymph nodes per patient was 116. In histopathologic exam-ination using HE stain, metastases were detected MethodsResults