Faculty Bibliography

Background: Traditionally, the American Joint Committee on Cancer (AJCC) staging uses Tumor size, Node status and distant Metastasis (TNM) to determine breast cancer prognosis, known as anatomic stage (AS). In the eighth edition, new parallel prognostic stage (PS) groups incorporate tumor biomarkers such as estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor 2 (HER2), tumor grade and Recurrence Score (RS). The new PS system attempts to refine the prognostic groups tailored to the tumor biology. Starting 2018, both AS and PS will be available to patients. In anticipation of the change, we performed a one-year retrospective study to look at how the breast cancer cases would be staged according to the new PS compared to the traditional AS. Design: Patients diagnosed with primary breast cancer in 2016 were identified using pathology laboratory software system. Only the cases with positive ER, PR and negative HER2 and concurrent Oncotype Dx RS were included. The cases were reviewed and stratified using the traditional TNM AS groups and also into the new PS groups. The change of each case's AS and PS was analyzed. Results: 106 cases were reviewed; 27 cases had Oncotype Dx RS less than 11 and 79 had equal to and greater than 11. Most common AS and PS were IA (67 and 65 cases, respectively). PS remained unchanged in 58.5% of cases, of which 91% were stage IA. Seventeen (16%) cases were upstaged and 27 (25%) cases were downstaged (see table 1). Conclusions: The new PS staging will bring about significant change to the traditional AS system-nearly half (42%) of cases reviewed at our institution would see a change in new prognostic stage grouping. We, as pathologists, need to reach a consensus as to how to integrate the results of the ancillary tests into our pathology reports in preparation for the upcoming PS to better serve our breast cancer patients

Background: Microglandular adenosis (MGA) is a rare borderline lesion of the breast characterized by an infiltrative collection of small glands that characteristically lack a myoepithelial cell layer. MGA is associated with invasive carcinoma in 20-30% of cases, and has been proposed as a non-obligate precursor to basal-like breast cancers. Somatic TP53 mutation of MGA and its associated carcinoma has been previously reported. We identified a case of triple negative metaplastic carcinoma with mesenchymal differentiation with morphologic and immunohistochemical evidence of progression from MGA to atypical MGA (AMGA), carcinoma in situ (CIS) and invasive carcinoma. We performed laser microdissection and whole exome sequencing of each four components (MGA, AMGA, CIS and cancer) with a matched benign sample to characterize the mutational landscape of these foci. Design: We selected a case of a metaplastic carcinoma with mesenchymal differentiation in juxtaposition to foci of MGA, AMGA and CIS. Immunohistochemically, all four foci were negative for estrogen receptor, progesterone receptor and human epidermal growth factor receptor 2 (her2) and positive for S-100. The distinct morphologic components and a matched control lymph node were separately microdissected from formalin-fixed paraffin embedded blocks. DNA extraction was performed and subjected to whole-exome sequencing on Illumina HiSeq platform. The results were demultiplexed and converted to FASTQ format using Illumina bcl2fastq software. Singlenucleotide and small indel somatic variants were called with MuTect2. Copy number profiles were calculated using Control-FREEC. Clonal populations were identified and quantified using PyClone. Results: Sequencing data resulted in mean coverage of 96-134X of targeted exome regions. Our results found a recurrent stop-gain R213* TP53 mutation in MGA, atypical MGA, CIS and metaplastic carcinoma. In addition, through variant allele frequency analysis, we identified two putative clonal clusters shared by all foci indicating a common molecular signature that is preserved in the morphologic spectrum of MGA-AMGA-CIS-metaplastic carcinoma. Conclusions: MGA is a molecularly advanced lesion with somatic mutation of TP53. We postulate that TP53 is an early event in the progression of MGA through AMGA, CIS and its associated metaplastic carcinoma. We report significant genetic overlap between MGA and its associated cancer

Background: Over 250,000 breast cancer cases are diagnosed annually in the US, of which 15% (n=37,500) are triple negative breast cancers (TNBC). TNBC has an aggressive clinical course with limited therapeutic options. Some studies estimate that a small proportion of TNBCs defined by immunohistochemical (IHC) stains alone, i.e. estrogen receptor (ER), progesterone receptor (PR) and human epidermal growth factor receptor 2 (HER2)-negative, can, in fact, show HER2 amplification by fluorescent in situ hybridization (FISH) and, therefore, be reclassified as HER2-enriched. If properly classified, these patients can benefit from anti-HER2 therapy. To capture this subset of patients, we performed a now four-year-long prospective study, in which TNBCs were reflexed to HER2 by FISH. Herein, we present our findings with detailed clinicopathologic analysis of the reclassified cases. Design: TNBC cases and ER/PR-positive, HER2-negative breast cancers from 2014 to 2017 with concurrent IHC and FISH results were analyzed. The pathology slides were reviewed by two pathologists. Stromal tumor infiltrating lymphocytes (sTILs) were defined as the percentage of all mononuclear cells within tumor stroma not in direct contact with tumor cells. Results: Of the 253 TNBCs in the past 4 years, 13 tumors (5%) showed HER2 amplification by FISH (6 surgical excisions and 7 core biopsies). Two patients have previously identified BRCA1 mutation. Twelve of 13 tumors (92%) were grade 3. All tumors were invasive ductal carcinoma of no special type. No apocrine morphology was noted. Notably, five tumors had >=50% sTILs, which can be classified as lymphocyte-predominant breast cancer based on some studies. In the same time period, 41 ER/PR-positive/HER2-negative breast cancers were reflexed to HER2 by FISH, none of which showed amplification. All tumors were fixed in formalin for comparable amount of time (6-72 hours). After reclassification, approximately 40% of patients received anti-HER2 therapy. (Table Presented) Conclusions: Pathologists may consider reflexing HER2 evaluation to FISH in TNBCs with grade 3, high ki67 and high sTILs. If 5% of TNBCs can be reclassified as HER2-enriched tumors, based on the national statistics, annually approximately 1875 patients with TNBCs by IHC may actually benefit from anti-HER2 therapy

Myxoid lesions of the breast can be diagnostically challenging entities. We report 4 cases of CD34+ fibromyxoid lesion that have been previously diagnosed as "benign myxoid lesion," "nodular mucinosis," or "mammary myofibroblastoma, myxoid type" on the basis of CD34-positivity. The lesions were microscopically well circumscribed and composed of a paucicellular spindle cell proliferation in a background of myxoid stroma. No epithelial component was identified. The spindle cells showed immunohistochemical reactivity for CD34 and smooth muscle actin. Based on morphologic and immunohistochemical similarities between these cases and myxoid myofibroblastoma, we compared 4 myxoid lesions with cases of typical myofibroblastoma, utilizing retinoblastoma (Rb) antibody and fluorescent in situ hybridization for 13q14 gene rearrangement (encoding the Rb gene). The myxoid lesions showed retention of Rb protein by immunohistochemistry, whereas Rb expression was lost in cases of myofibroblastoma. We identified loss of 13q14 in 3 of 4 cases of myofibroblastoma. Notably, 13q14 gene rearrangement was not observed in any of the myxoid lesions. Our data show that there is at least a subset of CD34+ fibromyxoid lesions that, despite overlapping morphologic and immunohistochemical phenotype and proposed common histogenesis with myofibroblastomas, is genetically distinct from the latter based on Rb analysis.

Bromodomain and extraterminal domain inhibitors (BETi) represent promising therapeutic agents for metastatic melanoma, yet their mechanism of action remains unclear. Here we interrogated the transcriptional effects of BETi and identified AMIGO2, a transmembrane molecule, as a BET target gene essential for melanoma cell survival. AMIGO2 is upregulated in melanoma cells and tissues compared to human melanocytes and nevi, and AMIGO2 silencing in melanoma cells induces G1/S arrest followed by apoptosis. We identified the pseudokinase PTK7 as an AMIGO2 interactor whose function is regulated by AMIGO2. Epigenomic profiling and genome editing revealed that AMIGO2 is regulated by a melanoma-specific BRD2/4-bound promoter and super-enhancer configuration. Upon BETi treatment, BETs are evicted from these regulatory elements, resulting in AMIGO2 silencing and changes in PTK7 proteolytic processing. Collectively, this study uncovers mechanisms underlying the therapeutic effects of BETi in melanoma and reveals the AMIGO2-PTK7 axis as a targetable pathway for metastatic melanoma.