OBJECTIVES: The expression of cyclooxygenase-2 (COX-2) and cell-cycle proteins (p21 and p27) proves useful in predicting prognosis and orientating therapy in many malignant tumours. Malignant pleural mesothelioma is an uncommon and lethal cancer for which there are limited treatment options. In this study, we evaluated the impact on prognosis and the influence on therapeutic strategy of immunohistochemical expression of COX-2, p21 and p27 in specimens from patients treated for malignant pleural mesothelioma. METHODS: We retrospectively reviewed immunohistochemical expression of COX-2, p21 and p27 dichotomised into high and low expression from specimens of 77 consecutive patients undergoing biopsy-plus-pleurodesis (n=6), pleurectomy-decortication (n=44) or extrapleural pneumonectomy (n=27) operations for malignant pleural mesothelioma between 1987 and 2007. Histology was of epithelioid (n=50), biphasic (n=17) and sarcomatoid (n=10) subtypes. Tumour node metastasis (TNM)-stage was I (n=21), II (n=36) and III (n=20). Therapies used were sole adjuvant radiotherapy (n=17), adjuvant radio-chemotherapy (n=56) and neo-adjuvant chemotherapy plus adjuvant radiotherapy (n=4). From 2005 on, preoperative maximal standard uptake value (SUV(MAX)) was also measured by fluorodeoxyglucose positron-emission-tomography. Significance was investigated by Kaplan-Meier survival and Cox regression analysis. RESULTS: The median survival was 10 months. Survival was negatively influenced by histology (epithelioid vs biphasic vs sarcomatoid) (p=0.009), positive macroscopic resection margins (p=0.016), metastatic mediastinal lymph nodes (p=0.016), high COX-2 (p=0.0001) expression, low p21 (p=0.0001) expression and low p27 (p=0.001) expression. Conversely, neither the type of surgery (biopsy-plus-pleurodesis vs pleurectomy-decortication vs extrapleural pneumonectomy), nor preoperative SUV(MAX) (> or = 6.0 vs <6.0), or combined therapies (sole radiotherapy vs adjuvant radio-chemotherapy vs neo-adjuvant chemotherapy plus adjuvant radiotherapy) reached a significant level of difference. Cox regression analysis showed that only immunohistochemical triple combination of high COX-2 and low p21 and p27 expression influenced survival (p=0.0001, hazard ratio 4.7, 95% confidence intervals 3-11) regardless of type of treatment. CONCLUSIONS: At Cox regression analysis, a combination of high COX-2 and low p21 and p27 expression resulted in the only negative prognosticator of malignant pleural mesothelioma. With this combination, less aggressive surgical options might be preferred. Copyright 2010 European Association for Cardio-Thoracic Surgery. Published by Elsevier B.V. All rights reserved.
OBJECTIVES: The expression of cyclooxygenase-2 (COX-2) and cell-cycle proteins (p21 and p27) proves useful in predicting prognosis and orientating therapy in many malignant tumours. Malignant pleural mesothelioma is an uncommon and lethal cancer for which there are limited treatment options. In this study, we evaluated the impact on prognosis and the influence on therapeutic strategy of immunohistochemical expression of COX-2, p21 and p27 in specimens from patients treated for malignant pleural mesothelioma. METHODS: We retrospectively reviewed immunohistochemical expression of COX-2, p21 and p27 dichotomised into high and low expression from specimens of 77 consecutive patients undergoing biopsy-plus-pleurodesis (n=6), pleurectomy-decortication (n=44) or extrapleural pneumonectomy (n=27) operations for malignant pleural mesothelioma between 1987 and 2007. Histology was of epithelioid (n=50), biphasic (n=17) and sarcomatoid (n=10) subtypes. Tumour node metastasis (TNM)-stage was I (n=21), II (n=36) and III (n=20). Therapies used were sole adjuvant radiotherapy (n=17), adjuvant radio-chemotherapy (n=56) and neo-adjuvant chemotherapy plus adjuvant radiotherapy (n=4). From 2005 on, preoperative maximal standard uptake value (SUV(MAX)) was also measured by fluorodeoxyglucose positron-emission-tomography. Significance was investigated by Kaplan-Meier survival and Cox regression analysis. RESULTS: The median survival was 10 months. Survival was negatively influenced by histology (epithelioid vs biphasic vs sarcomatoid) (p=0.009), positive macroscopic resection margins (p=0.016), metastatic mediastinal lymph nodes (p=0.016), high COX-2 (p=0.0001) expression, low p21 (p=0.0001) expression and low p27 (p=0.001) expression. Conversely, neither the type of surgery (biopsy-plus-pleurodesis vs pleurectomy-decortication vs extrapleural pneumonectomy), nor preoperative SUV(MAX) (> or = 6.0 vs <6.0), or combined therapies (sole radiotherapy vs adjuvant radio-chemotherapy vs neo-adjuvant chemotherapy plus adjuvant radiotherapy) reached a significant level of difference. Cox regression analysis showed that only immunohistochemical triple combination of high COX-2 and low p21 and p27 expression influenced survival (p=0.0001, hazard ratio 4.7, 95% confidence intervals 3-11) regardless of type of treatment. CONCLUSIONS: At Cox regression analysis, a combination of high COX-2 and low p21 and p27 expression resulted in the only negative prognosticator of malignant pleural mesothelioma. With this combination, less aggressive surgical options might be preferred. Copyright 2010 European Association for Cardio-Thoracic Surgery. Published by Elsevier B.V. All rights reserved.