Faculty Bibliography

In the age of precision medicine, extensive research has investigated tumour biomarkers to predict the behaviour of cancer and/or response to treatment in order to better understand the prognosis and treatment of disease. In breast cancer, significant progress has been made to categorise a common disease into subtypes defined by intrinsic tumour biology, measured by tumour biomarkers. This review encompasses the established biomarkers within breast cancer with the most up-to-date information regarding their understanding and clinical use as predictive and/or prognostic markers of breast cancer.

Low-stage, low-grade endometrioid endometrial carcinoma (EEC), the most common histologic type of endometrial cancer, typically has a favorable prognosis. A subset of these cancers, however, displays an aggressive clinical course with early recurrences, including distant relapses. All statistical tests were 2-sided. Using a combination of whole-exome and targeted capture sequencing of 65 FIGO stage IA and IB grade 1 EECs treated with surgery alone, we demonstrate that chromosome 1q gain (odds ratio [OR] = 8.09, 95% confidence interval [CI] = 1.59 to 54.6; P = .02), PIK3CA mutation (OR = 9.16, 95% CI = 1.95 to 61.8; P = .01), and DNA mismatch repair-deficient molecular subtype (OR = 7.92, 95% CI = 1.44 to 87.6; P = .02) are independent predictors of early recurrences within 3 years in this patient population. Chromosome 1q gain was validated in an independent dataset of stage I grade 1 EECs subjected to whole-exome sequencing. Our findings expand on the repertoire of genomic parameters that should be considered in the evaluation of patients with low-stage, low-grade EEC.

Herpes viruses are known for infecting epithelial cells and manifesting as vesicles. However, herpes viruses can also infect stromal cells. While established in the ocular setting, cutaneous stromal herpes (deep herpes) is previously unreported and may evade clinical and microscopic detection. We searched for skin biopsies with herpes stromal disease. Clinical information was retrieved via electronic medical records and pathology records system. Hematoxylin and eosin slides, immunohistochemical staining, and polymerase chain reaction detection of viral DNA was performed. We identified 12 specimens from 10 patients with cutaneous stromal herpes simplex virus 1/2 (n=7) or varicella-zoster virus infection (n=5). The most common site involved was the buttocks/perianal region (n=6). Ulceration was a frequent dermatologic finding (n=8). Pyoderma gangrenosum was clinically suspected in 6 specimens (50%). Eight patients (80%) were immunosuppressed. Biopsies frequently demonstrated a dense dermal mixed inflammatory infiltrate with subcutaneous extension and enlarged cells with viral cytopathic changes confirmed by herpes simplex virus 1/2 or varicella-zoster virus immunohistochemistry (n=10) or polymerase chain reaction (n=2). Most specimens (67%) lacked evidence of characteristic epidermal keratinocyte infection. This study presents the first known report of the ability of herpes virus to infect deep stromal cells of the dermis. We raise awareness of cutaneous stromal herpes in patients presenting with atypical clinical lesions, particularly while immunocompromised. Establishing the correct diagnosis is critical for initiating therapy.

Papillary neoplasms of the breast are a heterogeneous group of epithelial tumors nearly entirely composed of papillae. Their classification rests on the characteristics of the epithelium and the presence and distribution of the myoepithelial cells along the papillae and around the tumor. Papillary neoplasms of the breast can be diagnostically challenging, especially if only core needle biopsy (CNB) material is available. This review summarizes salient morphological and immunohistochemical features, clinical presentation, and differential diagnoses of papillary neoplasms of the breast. We include a contemporary appraisal of the upgrade rate to carcinoma (invasive carcinoma and ductal carcinoma in situ [DCIS]) and atypical hyperplasias in surgical excision specimens obtained following CNB diagnosis of papilloma without atypia, and a review of the available follow-up data in cases without immediate surgical excision.

EWSR1 is a 'promiscuous' gene that can fuse with many different partner genes in phenotypically identical tumors or partner with the same genes in morphologically and behaviorally different neoplasms. Our study set out to examine the EWSR1 fusions identified at our institution over a 3-year period, using various methods, their association with specific entities and possible detection of novel partners and associations. Sixty-three consecutive cases investigated for EWSR1 gene fusions between 2015 and 2018 at our institution were included in this study. Fusions were identified by either break-apart fluorescence in-situ hybridization (FISH), our clinical RNA-based assay for fusion transcript detection or both. Twenty-eight cases were concurrently tested by FISH and NGS, 24 were tested by FISH alone and 11 by NGS alone. Of the 28 cases with dual testing, 24 were positive by both assays for an EWSR1 gene fusion, 3 cases were discordant with a positive FISH assay and a negative NGS assay, and 1 case was discordant with a negative FISH assay but a positive NGS assay. Three novel fusions were identified: a complex rearrangement involving three genes (EWSR1/RBFOX2/ERG) in Ewing sarcoma, a EWSR1/TCF7L2 fusion in a colon adenocarcinoma, and a EWSR1/TFEB fusion in a translocation-associated renal cell carcinoma. Both colonic adenocarcinoma and renal cell carcinoma had not been previously associated with EWSR1 rearrangements to our knowledge. In a subset of cases, detection of a specific partner had an impact on the histological diagnosis and patient management. In our experience, the use of a targeted NGS-based fusion assay is superior to EWSR1 break-apart FISH for the detection of known and novel EWSR1 rearrangements and fusion partners, particularly given the emerging understanding that distinct fusion partners result in different diseases with distinct prognostic and therapeutic implications.

PURPOSE/OBJECTIVE:Human epidermal growth factor receptor 2 (HER2)-positive breast cancers are known to have significant clinical and pathological response to neoadjuvant systemic therapy (NST). The aim of this study was to identify factors associated with pathological complete response (pCR), defined as no residual invasive carcinoma in the breast and axillary lymph nodes (ypT0/is ypN0), among patients with HER2-positive breast cancer and to compare pCR rates between breast cancers with HER2 protein overexpression by immunohistochemistry (IHC) versus HER2 gene amplification by fluorescence in situ hybridization (FISH) in the absence of protein overexpression by IHC. METHODS:We conducted a retrospective review of HER2-positive breast cancer patients treated with NST and surgery at Memorial Sloan Kettering Cancer Center between January 2013 and May 2018. Estrogen receptor (ER), progesterone receptor (PR), and HER2 status were assessed according to the 2018 ASCO/CAP guidelines. RESULTS:During the study period, 560 patients were identified. Of 531 patients with IHC results available, 455 patients had HER2 IHC 3+, and 76 had IHC < 3+ but HER2 amplification detected by FISH. The overall pCR rate was 59% (330/560). The pCR rate among patients with HER2 protein overexpression (IHC 3+) was 67%, compared to 17% among patients with HER2 amplification by FISH (IHC < 3+). On univariate and multivariate analyses, HER2 protein overexpression by IHC (IHC 3+) was a significant predictor of pCR, along with grade 3 histology, PR-negative status, and dual anti-HER2 therapy. CONCLUSION/CONCLUSIONS:Although both HER2 IHC and FISH are standard HER2 testing methods in breast cancer, achievement of pCR is associated with HER2 IHC expression level, among other factors.