The Effect of Prolonged Cold Ischemia Time on Estrogen Receptor Detection in Breast Cancer
Xiaoxian Li, Michael T Deavers, Ming Guo, Lavinia P Middleton, Ping Liu, Lei Huo. The University of Texas, MDACC, Houston
Background: Recent guidelines provided by ASCO/CAP recommended that cold ischemia time (time from tumor removal to fixation) be limited to within 1hr in order to obtain accurate ER IHC results. While this is done routinely for core biopsy specimens, it may be difficult for resection specimens due to comprehensive intra-operative evaluations. Data is limited on the effect of prolonged cold ischemia time on ER IHC. In this study, we compare ER expression levels in resection specimens that have a cold ischemia time of >1hr to their corresponding biopsy specimens with the aim to guide our clinical practice.
Design: 78 resection specimens of invasive breast carcinoma with a cold ischemia time of >1hr identified between February and May 2011 in our department computer database were tested in this study. One section of the invasive carcinoma from each case was stained for ER (6F11, Novocastra Laboratories), then scored following ASCO/CAP guidelines and divided into 3 categories (10% and above, positive; 1-9%, low positive; less than 1%, negative). ER results from the corresponding core biopsy specimens were recorded from the pathology reports. Concordance was evaluated by weighted kappa statistics considering any change in category as an event. Any decrease in detected expression from biopsy to resection resulting in a categorical change or any decrease of more than 30% within the positive group in association with cold ischemia time was analyzed with Wilcoxon rank-sum test.
Results: The cold ischemia time ranged from 64 to 357 min (mean, 124). Based on the biopsy specimens, 70 tumors were ER positive, and 8 were ER negative. In the resection specimens, 2 of the 8 negative tumors on biopsy stained low positive, and 1 of the positive tumors stained low positive and 1 stained negative. The weighted kappa value was 0.83, indicating good concordance. Deceased expression resulting in a change in category (2 of 78 cases, 3.8%) was not associated with cold ischemia time (p=0.75). Within the group that was positive in both biopsy and resection specimens (68 cases), a lower expression level by over 30% in the resection was found in 4, and a higher expression level by over 30% in the resection was found in 5. This decrease in expression level was not associated with cold ischemia time (p=0.43).
Conclusions: In the majority of cases, cold ischemia time up to a few hours does not affect ER IHC. The decreased ER expression in the resection specimens in this study may not be due to prolonged cold ischemia time and warrants further investigation.
Monday, March 19, 2012 1:00 PM
Poster Session II # 71, Monday Afternoon