[1968] Impact of Different Antibody Clones on Ki-67 Labeling Index Assessment in Breast Carcinoma

Beth Z Clark, Mohamed M Desouki, David J Dabbs, Kim McManus, Rohit Bhargava. Magee-Womens Hospital of University of Pittsburgh Medical Center, Pittsburgh, PA

Background: Ki-67 proliferative index (LI) by immunohistochemistry is used in invasive breast carcinoma as a prognostic indicator, especially in estrogen receptor-positive breast carcinoma. Variability in Ki-67 LI may occur due to inter-observer variability, area of the tumor counted, and tumor heterogeneity. Although guidelines for interpreting Ki-67 results have been published, the impact of various clones on Ki-67 LI is still not well documented.
Design: In this quality assurance study of 46 invasive breast carcinomas, we determined Ki-67 LI using three antibody clones (30-9, SP-6, and MIB-1) to Ki-67. All samples were routinely fixed for 8-48 hours. In addition, we performed the staining with the same clones on paired tumor samples fixed for 96 hours to determine the impact of prolonged fixation on each clone. A single pathologist determined the Ki-67 LI for all clones and all cases, in order to minimize variability. The difference in Ki-67 LIs between different clones was determined using paired t-tests.
Results: Following routine fixation for 8-48 hours, the mean Ki-67 LI was highest for SP6 and lowest for MIB1 (mean Ki-67 LI of 36% for SP6, 34% for 30-9, and 30% for MIB1; p-value 0.3700 for 30-9 and SP6, 0.0124 for 30-9 and MIB1, and 0.0002 for SP6 and MIB1). Following 96-hour formalin fixation, mean Ki-67 LI remained highest for SP6 and lowest for MIB1 (mean Ki-67 LI of 37% for SP6, 30% for 30-9, and 27% for MIB1; p value <0.0001 for 30-9 and SP6, 0.0006 for 30-9 and MIB1, and <0.0001 for SP6 and MIB1). For the 2 clones showing the largest difference among each other, SP6 showed >10%age point higher Ki-67 LI than MIB1 in 18% of routinely fixed cases and in 39% of cases fixed for 96 hours.
Conclusions: The antibody clones for Ki-67 show statistically significant differences in mean LI, which could account for minor clinical differences in Ki-67 results from different institutions. The SP6 antibody clone was not affected by prolonged fixation and the 2 other clones (30-9 and MIB1) showed slight reduction in mean LI in the cases fixed for 96 hours. In a separate analysis, however, we found the difference (between routine fixation and 96 hours fixation) for these 2 clones to be clinically insignificant. The different clones used for Ki-67 analysis add to variability in LI, in addition to tumor heterogeneity, method of analysis, and inter-observer variability.
Category: Quality Assurance

Tuesday, March 5, 2013 9:30 AM

Poster Session III # 277, Tuesday Morning

 

Close Window