Limited Role of Ki67 Proliferative Index in Predicting Overall Survival in Patients with Pulmonary Carcinoid Tumors.
Ann E Walts, Catherine Bresee, Delma Ines, Alberto M Marchevsky. Cedars-Sinai Medical Center, Los Angeles, CA
Background: The current WHO classification defines pulmonary carcinoid tumors (PC) as typical (TC) or atypical (ATC) based on mitotic index (2 per 10 hpf) and/or the presence of necrosis. Ki67 index (%) has been incorporated into the classification of GI carcinoid tumors but there are currently no established criteria for interpreting Ki67 index in PC.
Design: 101 PC diagnosed as TC (78) and ATC (23) by WHO criteria were retrieved from our surgical pathology files. Slides were reviewed, diagnoses were confirmed, and survival information was obtained. A representative section of each tumor and appropriate controls were immunostained with Ki67 antibody (Ventana Inc. Tucson AZ) and the percentages of positive tumor nuclei were determined using the Ariol SL-50 Image Analyzer and imaging software (Genetix Corp, Boston MA) in accordance with the manufacturer's recommendations. Kaplan-Meier methodology was used to determine rates of overall survival and Cox proportional hazard models were used to determine the prognostic value of histology-based diagnoses and Ki-67 indices. ROC curves were used to establish cut-points of various Ki67 indices on overall survival.
Results: The mean Ki67 indices were significantly different between TC (mean 3.8, range 0.5 to 22) and ATC (mean 17.9, range 0.5 to 56; p=0.002) although the frequency distributions of Ki67 indices in the two groups showed considerable overlap. There was no significant difference in the mean length of follow-up between the TC and ATC groups. Histological diagnosis and Ki67 index were each independently strong predictors of survival (p<0.001 and p=0.003, respectively). However, when considered together in multivariate analysis, histological diagnosis was a strong predictor of survival (p<0.001) while Ki67 index was not (p=0.678). ROC curve analysis indicated that Ki67 cutoffs of 5%, 4%, and 3% yielded specificities of 69%, 61%, and 44% and sensitivities of 71%, 76%, and 88%, respectively in predicting overall survival. However, none of these Ki67 cutoff values provided a significant difference in survival within either the TC or the ATC group.
Conclusions: -Ki67 index is a significant predictor of overall survival in PC patients. -Histologic diagnosis utilizing WHO criteria is a significant predictor of overall survival in PC patients. -Taken together, histological diagnosis is a stronger predictor of overall survival and Ki67 index does not provide additional significant predictive information on overall survival within either TC or ATC patients.
Wednesday, March 2, 2011 9:30 AM
Poster Session V # 252, Wednesday Morning