OBJECTIVE: To investigate the prognostic factors and influence of the number of lymph node metastases on survival and UICC-TNM classification in patients with thoracic esophageal cancer after curative resection. METHODS: From 1985 to 1990, 1224 patients were surgically treated for thoracic esophageal cancer. The patients who died within 30 days after operation were not included in this study. Fifteen factors possibly influencing survival of these patients were selected and analyzed. A multivariate analysis of these individual variables was performed by Cox proportional hazard model. According to the number of lymph node metastases (0, 1 and > or = 2), a new modification of the TNM classification was suggested: stage IIa (T2N0M0 and T3N0M0), stage IIb [T1N1M0 and T2N1(1)M0], stage IIIa [T2N1 (2)M0 and T3N1 (1) M0] and stage IIIb [T3N1 (2) M0 and T4N any M0]. RESULTS: According to multivariate analysis, lymph node metastases, depth of invasion, location of tumor, histological classification and length of the tumor were of prognostic significance (P < 0.01). There was obvious correlation between the rate of lymph node metastasis and the depth of invasion, length of tumor and grade of differentiation. The 5-year survival rate of the patients with 0, 1 and > or = 2 positive metastatic lymph nodes was 59.1%, 32.0% and 8. 9%, respectively. The 5-year survival rate of the patients with stage T2N1 M0 and stage T3N1 M0 was significantly higher in those with only one lymph node involved than in those with two or more lymph nodes involved (43.1% vs. 18.0% and 28.0% vs. 9.6%, P < 0.01). The 5-year survival rate of the modified stage IIa, IIb, IIIa and IIIb was 56.5%, 43.9%, 25.6% and 11.1%, respectively, with a statistically significant difference among different stages (P < 0.01). CONCLUSION: The lymph node metastasis is the most important prognostic factor for thoracic esophageal cancer after resection. The major influencing factors of lymph node metastasis are the depth of invasion, length of tumor and grade of differentiation. Therefore, the lymphadenectomy along with esophagectomy and subsequently combined modality therapy against lymph node metastasis is necessary to improve the 5-year survival rate. Our proposed new classification based on number of lymph node metastases (0, 1, > or = 2 positive nodes) is more applicable because it can well reflect the correlation between lymph node metastasis and the survival, and provides evidence for the modification of the currently used UICC TNM staging system for surgically treated thoracic esophageal cancer.
OBJECTIVE: To investigate the prognostic factors and influence of the number of lymph node metastases on survival and UICC-TNM classification in patients with thoracic esophageal cancer after curative resection. METHODS: From 1985 to 1990, 1224 patients were surgically treated for thoracic esophageal cancer. The patients who died within 30 days after operation were not included in this study. Fifteen factors possibly influencing survival of these patients were selected and analyzed. A multivariate analysis of these individual variables was performed by Cox proportional hazard model. According to the number of lymph node metastases (0, 1 and > or = 2), a new modification of the TNM classification was suggested: stage IIa (T2N0M0 and T3N0M0), stage IIb [T1N1M0 and T2N1(1)M0], stage IIIa [T2N1 (2)M0 and T3N1 (1) M0] and stage IIIb [T3N1 (2) M0 and T4N any M0]. RESULTS: According to multivariate analysis, lymph node metastases, depth of invasion, location of tumor, histological classification and length of the tumor were of prognostic significance (P < 0.01). There was obvious correlation between the rate of lymph node metastasis and the depth of invasion, length of tumor and grade of differentiation. The 5-year survival rate of the patients with 0, 1 and > or = 2 positive metastatic lymph nodes was 59.1%, 32.0% and 8. 9%, respectively. The 5-year survival rate of the patients with stage T2N1 M0 and stage T3N1 M0 was significantly higher in those with only one lymph node involved than in those with two or more lymph nodes involved (43.1% vs. 18.0% and 28.0% vs. 9.6%, P < 0.01). The 5-year survival rate of the modified stage IIa, IIb, IIIa and IIIb was 56.5%, 43.9%, 25.6% and 11.1%, respectively, with a statistically significant difference among different stages (P < 0.01). CONCLUSION: The lymph node metastasis is the most important prognostic factor for thoracic esophageal cancer after resection. The major influencing factors of lymph node metastasis are the depth of invasion, length of tumor and grade of differentiation. Therefore, the lymphadenectomy along with esophagectomy and subsequently combined modality therapy against lymph node metastasis is necessary to improve the 5-year survival rate. Our proposed new classification based on number of lymph node metastases (0, 1, > or = 2 positive nodes) is more applicable because it can well reflect the correlation between lymph node metastasis and the survival, and provides evidence for the modification of the currently used UICC TNM staging system for surgically treated thoracic esophageal cancer.