Significance and reliability of pathologic marker Ki-67 in patients with neuroendocrine cancers.

Abstract

264 Background: Molecular markers, especially Ki-67, in neuroendocrine cancers (NETs) have become increasingly important. Debate exists as to the optimal role of ki-67 in the treatment of this uncommon cancer. This study was designed to examine the significance of ki-67 as a clinical predictor and its reliability in the treatment of NETs. Methods: A retrospectively assembled integrated database with prospectively collected data of patients undergoing multidisciplinary management for NETs between 2000 to 2009 was analysed. Clinical and outcomes data were collected. Ki-67 was then categorized to Group A (≤2%), B (3-15%) and C (>15%). We compared the original Ki-67 value to the reviewed value. We then used a kappa statistic to calculate the inter-rater agreement between the original and reviewed determination of Ki-67. Fisher's exact test was used to compare categorical variables. Survival was evaluated using the log-rank test. All analyses were carried out using SAS 9.1.3. Results: A total of 184 patients were seen at our clinic. Ki-67 correlated with metastases at presentation (36, 52, 63% for Groups A, B, C respectively, p<0.05), while influencing treatment with chemotherapy even in the absence of metastases (14, 29, 57% for Groups A, B, C respectively, p< 0.002). Ki-67 predicted overall survival (p=0.0005) in favor of Group A. 99 patients had an original Ki-67 reported from the referring center and then a review by an expert pathologist at the multidisciplinary clinic. In Group A, there was 94.4% agreement, with 3.7% of cases upgraded at review to Group B and 1.9% of cases upgraded to Group C. In Group B, there was 94.3% agreement, with 5.7% of cases downgraded to Group A, and 0% upgraded. In Group C, there was 90% agreement, with 10% of cases downgraded to Group B and none to Group A (kappa = 0.89). Conclusions: Our previous report had demonstrated that Ki-67 influenced decisions regarding treatment options. In our updated population of NETs patients, Ki-67 continued to predict biology, influence treatment, and predict survival. In addition we demonstrated high reproducibility of Ki-67. It appears that with modern techniques Ki-67 can be highly reproducible and reliable tool in improving outcomes in this patient population. No significant financial relationships to disclose.