Search

Pulmonary carcinoid tumors are currently classified as typical or atypical based on the mitotic index (2 per 10 hpf) and/or the presence of necrosis. Following incorporation of the Ki-67 index into the classification of GI carcinoid tumors, our oncologists have also been requesting this test as part of the work-up of pulmonary carcinoid tumors although there are currently no established criteria for interpreting Ki-67 index in these neoplasms. We utilized the Ariol(®) SL50 Image Analyzer system to measure the Ki-67 index in 101 pulmonary carcinoid tumors (78 typical and 23 atypical) and then correlated the Ki-67 index and the histological diagnoses in univariate and multivariable analysis with overall survival. The mean Ki-67 indices for the typical carcinoids (3.7 s.d.± 4.0) and the atypical carcinoids (18.8 s.d.± 17.1) were significantly different (P<0.001) although the frequency distributions of Ki-67 indices in the two groups overlapped considerably. Receiver operating characteristic curve analysis showed that a Ki-67 index cutoff value of 5% provided the best fit for specificity and sensitivity in predicting overall survival. Histological diagnosis and the Ki-67 index cutoff of 5% were each independently strong predictors of survival (P<0.001 and P=0.003, respectively). When considered together in multivariable analysis, histological diagnosis was the stronger predictor of overall survival and a Ki-67 index cutoff of 5% did not provide additional significant predictive survival information within either the typical carcinoid or the atypical carcinoid patient group. A few typical carcinoid patients with Ki-67 indices of 5% appeared to have worse survival after 5 years than those with Ki-67 indices <5%, but the data set was insufficiently powered to further analyze this. These findings do not provide best evidence for the routine use of Ki-67 index to prognosticate overall short-term survival in patients with pulmonary carcinoid tumors.