Utility of pHH3 Immunohistochemistry in Risk Stratification of Gastrointestinal Stromal Tumors: An Inter-Observer Variability Study
Ahmad Alkhasawneh, Lizette Vila Duckworth, Xiaomin Lu, Thomas George, John Reith, Tania Zuluaga Tora. University of Florida, Gainesville, FL
Background: Gastrointestinal stromal tumors (GIST) are graded based on mitotic count (low grade ≤5 mitoses/5mm2, high grade >5 mitoses/5mm2). However, mitotic counts are subject to interobserver variability. Phospho-histone H3 (pHH3) is a recently described mitosis-specific antibody which may decrease the interobserver variability. Our aim was to evaluate the mitotic count interobserver variability of GIST by comparing counts on H&E slides vs pHH3 immunostains. To our knowledge, no previous studies have evaluated mitotic activity of GIST using pHH3.
Design: 70 GIST cases (≥2 cm) resected between 2000-2011 were identified. All cases with mitotic rates >5 mitosis/5mm2 (10 cases) were included. 40 of the remaining 60 cases with ≤5 mitoses/5mm2 were randomly selected for a total of 50 cases. All slides from each case were reviewed to identify the slide with most "hotspots" (areas with readily identifiable mitotic figures). In cases lacking prominent mitoses, slides with hypercellular areas and/or nuclear pleomorphism were selected. pHH3 immunohistochemistry (IHC) was performed on the corresponding block. Quantitative analysis of mitotic activity was performed independently by four pathologists (1 soft tissue pathologist, 1 gastrointestinal pathologist, 1 gastrointestinal pathology fellow, and 1 pathology resident) on H&E and pHH3-stained slides by counting the number of mitoses/5mm2 area. An interclass correlation analysis was performed to test the interobserver variability for all cases.
Results: Kappa values showed very good interobserver variability for mitotic counts on both H&E and pHH3-stained slides. Of the 200 total mitotic counts on H&E slides, 141 had ≤5 mitoses/5mm2 (low grade) and 59 had >5 mitoses/5mm2 (high grade). However, using pHH3 IHC, 30 (21%) of the low grade counts had >5 mitoses and 6 (10%) of the high grade counts had ≤5 mitoses. Overall, at least one observer's mitotic count changed in 25 of 50 cases.
Conclusions: The current risk stratification for GIST is based on site, tumor size, and accurate mitotic count. In this study, interobserver variability was very good for both methods used. However, pHH3 IHC may alter mitotic counts in a significant number of GIST, and may be a valuable adjunct in the assessment of mitoses in GIST.
Monday, March 4, 2013 9:30 AM
Poster Session I Stowell-Orbison/Surgical Pathology/Autopsy Awards Poster Session # 88, Monday Morning