BACKGROUND: The outcome of treatment in non-small cell lung cancer (NSCLC) remains poor. One of the reasons is that in many patients its biological behavior does not follow a definite pattern, and can not be accurately predicted prior to treatment. In the present study we have examined the significant prognostic predictors. METHODS: One hundred and fifty-eight patients with NSCLC entered this study. They received surgery alone (95 cases) or combined therapy with postoperative irradiation (63 cases). Three types of data have been collected: (1) clinical characteristics: age, sex, Karnofsky performance status, weight loss, T stage, and N stage; (2) histopathology studies: histological types, tumor differentiation, status of vascular and lymphatic vessel invasions; (3) laboratory measurements by immunohistochemistry assay: oncoprotein overexpression, including pan-ras, c-myc, neu, epidermal growth factor receptor (EGFR) and p53, and tumor cell proliferation by proliferating cell nuclear antigen (PCNA). RESULTS: For the entire group, 5-year actuarial survival, local control and distant metastasis rates were 44, 63 and 40%, respectively. In the univariate analyses, T stage, N stage and lymphatic vessel invasion correlated to survival; T stage and N stage to local control; N stage, lymphatic vessel invasion and pan-ras protein positive stain to distant metastasis. When the index of oncoprotein positive stains was used, the higher index was associated with a higher distant metastasis rate. In the multivariate analyses, T stage, N stage and lymphatic vessel invasion could be independent predictors for survival; T stage for local control; N stage, lymphatic vessel invasion and index of positive oncoprotein stains for distant metastasis. CONCLUSIONS: Late T and N stages, lymphatic vessel invasion and multi-oncoprotein positive stains would predict poor prognoses for NSCLC.
BACKGROUND: The outcome of treatment in non-small cell lung cancer (NSCLC) remains poor. One of the reasons is that in many patients its biological behavior does not follow a definite pattern, and can not be accurately predicted prior to treatment. In the present study we have examined the significant prognostic predictors. METHODS: One hundred and fifty-eight patients with NSCLC entered this study. They received surgery alone (95 cases) or combined therapy with postoperative irradiation (63 cases). Three types of data have been collected: (1) clinical characteristics: age, sex, Karnofsky performance status, weight loss, T stage, and N stage; (2) histopathology studies: histological types, tumor differentiation, status of vascular and lymphatic vessel invasions; (3) laboratory measurements by immunohistochemistry assay: oncoprotein overexpression, including pan-ras, c-myc, neu, epidermal growth factor receptor (EGFR) and p53, and tumor cell proliferation by proliferating cell nuclear antigen (PCNA). RESULTS: For the entire group, 5-year actuarial survival, local control and distant metastasis rates were 44, 63 and 40%, respectively. In the univariate analyses, T stage, N stage and lymphatic vessel invasion correlated to survival; T stage and N stage to local control; N stage, lymphatic vessel invasion and pan-ras protein positive stain to distant metastasis. When the index of oncoprotein positive stains was used, the higher index was associated with a higher distant metastasis rate. In the multivariate analyses, T stage, N stage and lymphatic vessel invasion could be independent predictors for survival; T stage for local control; N stage, lymphatic vessel invasion and index of positive oncoprotein stains for distant metastasis. CONCLUSIONS: Late T and N stages, lymphatic vessel invasion and multi-oncoprotein positive stains would predict poor prognoses for NSCLC.