Repeat of Estrogen Receptor on Breast Resection Specimens with “Low ER” on Core Biopsy – A Quality Assurance Study
Nika C Gloyeske, David J Dabbs, Abbie Mallon, Kim McManus, Rohit Bhargava. Magee-Womens Hospital of UPMC, Pittsburgh, PA
Background: In breast cancer, the level of estrogen receptor (ER) expression as determined by immunohistochemistry (IHC) is an important prognostic and predictive marker. Therefore, the ASCO/CAP guidelines on hormone receptor testing suggest reporting ER IHC results semi-quantitatively. Usually hormone receptor studies are performed on the core biopsy specimen and if ER is positive (H-score ≥1), the testing is not repeated on the resection. However, there is some concern if semi-quantitative results for “low ER positive cases” on core biopsy are accurate and representative of the entire tumor.
Design: We selected patients that had breast cancers with a low level of ER expression (H-score 1-100) on core biopsies performed in 2010. We compared the core biopsy result on each case with repeat ER H-score on one tumor block from the resection. The cold ischemic time was calculated for each of the resection specimens when it was available.
Results: Of the 731 total ER positive breast cancers in 2010, 53 (7.3%) cases had H-scores of 1-100. Of these 53 cases, 37 had paired core biopsy and resection specimens containing invasive tumor which formed the basis of this quality assurance study. The mean H-score for the core biopsies was 28, and 18 for the resection (p=0.1252). The H-scores on repeat testing ranged from 0-100 with 8 of 37 cases becoming negative (i.e. H-score of 0). Eight patients received neoadjuvant therapy of which 4 (50%) became negative on repeat testing. However only 4 of 29 (14%) cases not treated with neoadjuvant became negative (p=0.0487). These 8 patients had similar H-scores on core biopsy in both groups (mean H-score of 26 in 4 tumors treated with neoadjuvant, and mean H-score of 26.3 in 4 cases not treated with neoadjuvant). The most significant difference between tumors that became negative versus those that remained positive was the prolonged cold ischemic time in the former group. The mean cold ischemic time for tumors that became negative was 129 minutes (range: 20-337 minutes) compared to 52 minutes (range: 17-179 minutes) for tumors that remained positive (p=0.0078).
Conclusions: Breast cancers expressing ER at a low level constitute a minority of ER positive cases. Core biopsies are better suited than resection specimens for determining low level ER expression because of negligible cold ischemic time. Although the difference in clinical behavior for low ER expressing tumors versus ER negative tumors is sometimes debatable, this study reiterates the importance of maintaining a cold ischemic time of one hour or less for IHC result accuracy.
Category: Quality Assurance
Tuesday, March 5, 2013 9:30 AM
Poster Session III # 274, Tuesday Morning