b iframe width height src https www
GATA3 is a transcription factor that plays a role in mam-mary epithelial cell differentiation  and is linked to the ER signaling pathway [19,20]. In surgical pathology, GATA3 is a sensitive marker of both breast and urothelial carcinomas; however, its λ-Carrageenan has also been observed in tumor types including salivary gland, gestational trophoblastic, some pan-creatic carcinomas, and many others [21,22]. GATA3 has su-perior sensitivity compared with older, established breast lineage markers such as GCDFP15 and mammaglobin , and it is a sensitive marker for identifying MBC [23,24]. How-ever, the main drawback to GATA3 as a breast lineage marker is that its sensitivity is highly correlated with ER status. Although GATA3 positivity in ER-positive PBC ranges from 91% to 100%, it has a comparatively lower sensitivity in TNBC, with most studies reporting the sensitivity of GATA3 between 43% and 66% in these tumors [25,26].
AR is a steroid hormone receptor and nuclear transcription factor that plays a role in normal breast development and mammary cell proliferation . AR is the most widely expressed nuclear hormone receptor in breast cancer, and it is positive in 85% to 95% of ER+ breast cancers, as well in a large number of carcinomas and TNBCs [28,29]. AR has been the focus of recent attention as a potential candidate for therapeutic targeting, particularly for patients with TNBC. Several clinical trials are currently investigating AR antago-nists and modulators for patients with TNBC and treatment-refractory ER-positive disease [30,31]. Meanwhile, the significance of AR expression as a prognostic biomarker in TNBC has been controversial, with several reports describing conflicting results . Similar to GATA3, the expression of AR in breast cancer is highly correlated with ER status and tumor subtype, with AR showing higher expression in ER+ and lower-grade tumors . There is significant variability in the reported prevalence of AR in TNBC, possibly related to differences in primary antibody and cutoff used for positivity. A wide range of 6% to 75% positivity rate has been reported, with most positivity rates reported between 25% and 35% for AR in TNBC [34,35]. To date, no studies have compared the utility of AR as a breast lineage marker in TNBC to GATA3 and SOX10. Few publications have addressed the stability and concordance of AR between PBC and matched MBC.
There are several strengths in the design of our study. Our analysis represents the largest series of cases comparing SOX10 expression in PBC and triple-negative MBC. In con-trast to other reports in which only a small subset of series of MBC was TNBC , our investigation exclusively focused on triple-negative MBC, which is an important diagnostic dilemma and the most relevant clinical scenario. Sampling bias can be marked with tissue microarray–based analyses; we minimized this potential source of bias by performing IHC on whole tissue sections, which often represented half or the entire tumor in the MBC cases. Also, our present work is the only comparative analysis of SOX10 to both GATA3 and AR with regard to marker concordance and stability in MBC.
In summary, we conclude that using both GATA3 and SOX10 is recommended for confirming a breast site of origin in putative breast cancer metastases that lack ER, PR, and HER2 expression, whereas the addition of AR is not useful. In analyzing PBC, SOX10 could be a useful adjunct to support a mammary primary in a poorly differentiated, triple-negative tumor that lacks a recognizable in situ component, and in which the differential diagnosis includes a nonmammary me-tastasis to the breast. SOX10 is a sensitive and stable lineage marker in TNBC and can be used to support a mammary pri-mary site of origin in metastases from patients with a clinical history of TNBC. In the appropriate clinical context and in conjunction with IHC results such as positive cytokeratin staining, SOX10 can also suggest a mammary primary if a metastatic lesion is triple negative and melanoma has been excluded.