Tumor Sampling Using Tissue Microarray Can Alter the Ki-67 Proliferation Index and Cause Misclassification of Breast Tumors.
Cary Chisholm, Courtney Yau, Christopher Ruud, V O Speights. Scott & White Memorial Hospital, Temple, TX
Background: Metastatic breast cancer is a devastating diagnosis that can be difficult to adequately treat. Breast cancers have been recently divided into several subgroups: Luminal A (ER+/Her2-, Ki-67 <14%), Luminal B (ER+/Her2-, Ki-67 >14%), ER+/Her2+, ER-/Her2+, and triple negative (ER-/PR-/Her2-). These subgroups have different clinical behavior and response to hormonal or chemotherapy. Lymphocyte infiltration of breast tumors has been shown very recently to potentially predict an increased response to chemotherapy. Many studies are performed using tissue microarray. We evaluated whether the leading edge of the tumor has a higher proliferation index than the center and whether tumor sampling for Ki-67 could lead to misclassification of tumors.
Design: 110 patients were identified with breast cancer metastases diagnosed between 1999-2007. Pertinent demographic, clinical, and pathological data were collected after permission was granted from the institutional review board. The primary breast tumors were retrieved and sampled in the center and leading edge using a 3 mm microarray punch. The microarray tissue was arranged into the blocks in such a manner as to provide no reliable way to predict which tissue samples were taken from the same tumor. These were then stained with Ki-67 and the proliferation index was visually estimated.
Results: Ki-67 proliferation indexes were significantly higher at the leading edge (31.53) than in the center (28.08) of the tumor (p=0.0061). When classified according to the Ki-67 from the center, 16 patients were identified as Luminal A and 36 were classified as Luminal B. Overall survival was 57 months and 56 months, respectively. Using the Ki-67 from the leading edge reclassified 2 (12.5%) Luminal A patients as Luminal B such that 14 were now Luminal A and 38 were Luminal B. Overall survival data for Luminal A and Luminal B changed dramatically: 67 months and 50 months, respectively, suggesting that these two tumors were indeed Luminal B based on clinical behavior.
Conclusions: The leading edge of the tumor was conclusively shown to proliferate at a higher rate than the center of the tumor. Approximately 13% of tumors were classified as Luminal A when the center of the tumor was sampled. Sampling the periphery of the tumor reclassified these as Luminal B consistent with their clinical behavior. When performing breast cancer research using tissue microarray methods, the leading edge must be sampled to avoid inappropriate classification.
Tuesday, March 1, 2011 9:30 AM
Poster Session III # 35, Tuesday Morning