A novel BRD4-NUT fusion in an undifferentiated sinonasal tumor highlights alternative splicing as a contributing oncogenic factor in NUT midline carcinoma.

NUT midline carcinoma (NMC) is a fatal cancer that arises in various tissues along the upper midline of the body. The defining molecular feature of NMC is a chromosomal translocation that joins (in the majority of cases) the nuclear testis gene NUT (NUTM1) to the bromodomain protein family member 4 (BRD4) and thereby creating a fusion oncogene that disrupts cellular differentiation and drives the disease. In this study, we report the case of an adolescent NMC patient presenting with severe facial pain, proptosis and visual impairment due to a mass arising from the ethmoid sinus that invaded the right orbit and frontal lobe. Treatment involved radical resection, including exenteration of the affected eye with the view to consolidate treatment with radiation therapy; however, the patient experienced rapid tumor progression and passed away 79 days post resection. Molecular analysis of the tumor tissue identified a novel in-frame BRD4-NUT transcript, with BRD4 exon 15 fused to the last 124 nucleotides of NUT exon 2 (BRD4-NUT ex15:ex2Δnt1-585). The partial deletion of NUT exon 2 was attributed to a mid-exonic genomic breakpoint and the subsequent activation of a cryptic splice site further downstream within the exon. Inhibition of the canonical 3' acceptor splice site of NUT intron 1 in cell lines expressing the most common NMC fusion transcripts (PER-403, BRD4-NUT ex11:ex2; PER-624, BRD4-NUT ex15:ex2) induced alternative splicing from the same cryptic splice site as identified in the patient. Detection of low levels of an in-frame BRD4-NUT ex11:ex2Δnt1-585 transcript in PER-403 confirmed endogenous splicing from this alternative exon 2 splice site. Although further studies are necessary to assess the clinical relevance of the increasing number of variant fusions described in NMC, the findings presented in this case identify alternative splicing as a mechanism that contributes to this pathogenic complexity.

Fascin and cyclin D1 immunoreactivity in non-neoplastic vulvar squamous epithelium, vulvar intraepithelial neoplasia and invasive squamous carcinoma: correlation with Ki67 and p16 protein expression.

AIMS: To investigate cyclin D1 and fascin immunoreactivity in normal, reactive and neoplastic vulvar skin correlating the findings with p16 protein and Ki67 expression. METHODS: 66 vulvar biopsy or resection specimens demonstrating normal appearances, reactive epidermal changes, usual-type vulvar intraepithelial neoplasia (uVIN), differentiated-type VIN (dVIN), p16-positive squamous cell carcinoma (SCC) and p16-negative SCC were examined immunohistochemically for cyclin D1, fascin, Ki67 and p16 protein. Where applicable, expression patterns were compared in microanatomically distinct areas, particularly at the invasive front (deep tumour margin) of SCC. RESULTS: Normal epidermis showed parabasal Ki67 and cyclin D1 staining while fascin labelled cells in the lower one-third of the epithelium. Reactive and dVIN specimens demonstrated mildly increased Ki67 and cyclin D1 expression that maintained parabasal polarity, whereas uVIN and p16-positive SCC were characterised by loss of cyclin D1 staining. However, in 14 of 20 p16-positive SCC small infiltrative tumour groups and single infiltrating cells at the invasive front showed a cyclin D1-positive/ Ki67-negative phenotype. In contrast, p16-negative SCC generally showed diffuse and concordant cyclin D1 and Ki67 labelling, including at the invasive margin. Fascin expression was increased in all VIN and SCC lesions. CONCLUSIONS: Variations in cyclin D1 and Ki67 expression between p16-positive and p16-negative vulvar SCCs suggest different mechanisms of invasion in these tumour subgroups. Fascin is upregulated in vulvar squamous neoplasia but immunostaining does not discriminate in situ from invasive lesions nor putative human papilloma virus (HPV)-associated and HPV-independent SCCs.

CD147 (EMMPRIN) and matrix metalloproteinase-2 expression in uterine endometrioid adenocarcinoma.

Low-grade uterine endometrioid adenocarcinomas (EACs) may exhibit a distinct pattern of myometrial invasion characterized by the presence of microcystic, elongated, and fragmented ('MELF') glands but the factors influencing this pattern of invasion are not known. Immunohistological expression of CD147 (extracellular inducer of matrix metalloproteinase, EMMPRIN) and matrix metalloproteinase-2 (MMP2) was studied in 22 EAC and the results compared between the 'conventional' neoplastic glands, foci of MELF-type invasion, and in the tumor-associated stroma. The conventional tumor areas showed strong membranous expression of CD147, and staining was more consistently present toward the central (luminal) aspect of the larger glands. In contrast, the tumor cells showed only focal expression of MMP2, and in some cases CD147 and MMP2 staining was inversely correlated. The neoplastic epithelium within MELF areas usually was negative for both CD147 and MMP2. However, there was consistent MMP2 expression in the reactive stroma that characteristically surrounds foci of MELF pattern invasion. This micro-anatomical variation in protein expression within EAC suggests a functional alteration in the neoplastic epithelium during the invasive process. It is likely that stromal MMP2 plays a role in potentiating invasion in these tumors but this may be regulated by factors other than CD147.

Prognostic value of mitotic counts and Ki-67 immunoreactivity in adult-type granulosa cell tumour of the ovary.

AIMS: The identification of reliable prognostic factors in patients with ovarian stage 1 adult-type granulosa cell tumour (GCT) has proved problematic. Some reports have suggested that proliferation indices may be of value, but the data are conflicting and the methods of assessment often poorly defined. In this study the mitotic activity and Ki-67 immunohistochemistry was assessed in a series of GCT using carefully specified methodology, and the findings were correlated with clinicopathological findings. METHODS: Tumour proliferation was assessed in 38 primary GCT by counting mitotic figures in 50 high-power fields (x500 magnification) with results expressed as a mean count per 2 mm(2) standardised area. The number of mitotic figures and Ki-67 immunoreactive cells per 10,000 tumour cells was also assessed using an ocular cell counting graticule. The results were correlated with tumour stage at presentation and with the development of tumour recurrence. RESULTS: Twenty-nine patients were stage 1 at presentation, and nine patients had high-stage disease (extra-ovarian spread). Nine patients with initial stage 1 disease developed metastases, and 20 patients had no evidence of recurrence over a mean follow-up period of 11.1 years. There was no significant correlation between any of the proliferation indices or with clinical outcomes. CONCLUSIONS: These results suggest that proliferation assessment is of limited value in the pathological assessment of GCT. Future studies should carefully specify the methods of assessing cell proliferation to ensure a reliable comparison of results.

Value of elastin staining in the assessment of peritoneal implants associated with ovarian serous borderline tumours.

AIMS: To determine whether elastin stains aid in classifying peritoneal implants associated with ovarian serous borderline tumours (SBT). METHODS AND RESULTS: The study group comprised 80 implants (nine invasive and 71 non-invasive) from 28 patients with ovarian SBT. Elastin stains were performed using histochemical and immunohistochemical methods to demonstrate the peritoneal elastic lamina (PEL), and evaluated with regard to assessment of the subtype of implant. The elastin stains demonstrated the PEL in most anatomical sites other than the omentum and the bladder and were considered helpful in 44/80 (55%) cases. The stains were most useful in the assessment of poorly oriented or traumatized biopsy specimens and in confirming the superficial distribution of non-invasive implants. The staining was non-contributory in most of the remaining biopsies, because the PEL was not identified. CONCLUSIONS: Demonstration of the PEL using elastin stains can be useful in the subclassification of implants associated with ovarian SBT and is of most value in confirming the superficial distribution of non-invasive lesions. However, evaluation is limited by the absence of a defined elastic layer in a proportion of biopsy specimens, possibly reflecting their superficial location, as well as absence of a distinct PEL in sites such as the omentum.