- © 2012 by American Society of Clinical Oncology
Reply to E.A. Rakha et al
- University of Texas MD Anderson Cancer Center, Houston TX
- Okayama University Hospital, Okayama City, Japan
- Corresponding author: Lajos Pusztai, MD, D.Phil., Professor of Medicine, UT MD Anderson Cancer Center, Department of Breast Medical Oncology, PO Box 301439, Houston TX 77230-1439; e-mail: lpusztai{at}mdanderson.org.
We thank Rakha et al1 for their thoughtful comments and the additional dimension that they have provided for our report on borderline estrogen receptor (ER)–positive cancers.2 We studied gene-expression profiles of breast cancers that were 1% to 9% ER-positive by routine immunohistochemistry (IHC) and found that only a minority of these cancers had molecular features of ER-positive, endocrine-sensitive tumors. Three different types of assessment, including ESR1 mRNA expression,3 PAM50 molecular class,4 and sensitivity to endocrine therapy gene signature5 consistently indicated that the majority of these cancers showed molecular features of ER-negative, basal-like cancers. We concluded that in this rare subset of patients a second RNA-based ER assessment may help to identify the minority of ESR1 mRNA-positive, luminal type cancers, and as a result of the substantial uncertainty about endocrine sensitivity in this population with borderline ER-positive cancer, the safest clinical approach may be to consider both adjuvant endocrine and chemotherapy.
Rakha et al1 point out an alternative method to gauge confidence in the ER status of 1% to 9% IHC-positive cancers through a repeat ER staining on the surgical resection specimen, particularly when the initial hormone-receptor determination was performed on a core needle biopsy. Their own IHC-based reassessment of borderline ER-positive cancers agrees with our finding that only approximately one quarter of these cases showed greater than 10% ER staining on repeat IHC.6 On the basis of their report, we support the conclusion of Rakha et al1 that a repeat ER IHC on the surgical specimen is a reasonable approach to clarify the receptor status of borderline ER-positive cancers. We could not have possibly made a statement about the value of repeat ER IHC on the basis of our work because we did not study the value of repeat IHC.
AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
The author(s) indicated no potential conflicts of interest.
