PURPOSE: Primary archival tumor tissues of 89 patients with clinical stage I nonseminomatous germ cell tumors were analyzed for MIB-1 expression and histological parameters such as percentage embryonal carcinoma and presence of vascular invasion to determine the value of these parameters to predict absence or presence of occult lymph node disease. MATERIALS AND METHODS: A monoclonal antibody (MIB-1) developed for application in paraffin-embedded tissues was used to measure immunohistochemical expression of Ki-67 for the overall tumor (total MIB-1) and for each malignant cell type present. In addition, the primary tumors were examined for the presence of vascular invasion and determination of quantitative histology. Univariate and multivariate logistic regression models were used for statistical analysis. RESULTS: Univariate analysis neither revealed total MIB-1 score nor MIB-1 score in the highest area of staining of the different cell types to significantly predict pathological stage I or stage II disease. However, the presence of vascular invasion (p < 0.0001) and the percentage of embryonal carcinoma (p < 0.0001) were significant risk factors for occult nodal disease. Multivariate logistic regression analysis revealed the combination of vascular invasion and the percentage of embryonal carcinoma to be the best model to predict pathological stage II correctly (86.5%). DISCUSSION: The determination of immunohistochemical MIB-1 expression did not correlate with pathological stage in clinical stage I nonseminomatous germ cell tumors (NSGCT). We were not able to define high risk or low risk groups for occult nodal disease based on MIB-1 staining results. However, percentage of embryonal carcinoma and presence of vascular invasion accurately predicted absence or presence of lymph node metastasis in clinical stage I NSGCT. Our study underlines that a prospective multicenter trial is urgently needed to accurately assess the role of MIB-1 staining in management of clinical stage I NSGCT.
PURPOSE: Primary archival tumor tissues of 89 patients with clinical stage I nonseminomatous germ cell tumors were analyzed for MIB-1 expression and histological parameters such as percentage embryonal carcinoma and presence of vascular invasion to determine the value of these parameters to predict absence or presence of occult lymph node disease. MATERIALS AND METHODS: A monoclonal antibody (MIB-1) developed for application in paraffin-embedded tissues was used to measure immunohistochemical expression of Ki-67 for the overall tumor (total MIB-1) and for each malignant cell type present. In addition, the primary tumors were examined for the presence of vascular invasion and determination of quantitative histology. Univariate and multivariate logistic regression models were used for statistical analysis. RESULTS: Univariate analysis neither revealed total MIB-1 score nor MIB-1 score in the highest area of staining of the different cell types to significantly predict pathological stage I or stage II disease. However, the presence of vascular invasion (p < 0.0001) and the percentage of embryonal carcinoma (p < 0.0001) were significant risk factors for occult nodal disease. Multivariate logistic regression analysis revealed the combination of vascular invasion and the percentage of embryonal carcinoma to be the best model to predict pathological stage II correctly (86.5%). DISCUSSION: The determination of immunohistochemical MIB-1 expression did not correlate with pathological stage in clinical stage I nonseminomatous germ cell tumors (NSGCT). We were not able to define high risk or low risk groups for occult nodal disease based on MIB-1 staining results. However, percentage of embryonal carcinoma and presence of vascular invasion accurately predicted absence or presence of lymph node metastasis in clinical stage I NSGCT. Our study underlines that a prospective multicenter trial is urgently needed to accurately assess the role of MIB-1 staining in management of clinical stage I NSGCT.
Authors: Duncan C Gilbert; Reem Al-Saadi; Khin Thway; Ian Chandler; Daniel Berney; Rhian Gabe; Sally P Stenning; Joan Sweet; Robert Huddart; Janet M Shipley Journal: Clin Cancer Res Date: 2015-10-09 Impact factor: 12.531