No abstract

Merkel cell carcinoma (MCC) is an aggressive neuroendocrine carcinoma of the skin. The main etiological agent is Merkel cell polyomavirus (MCPyV), detected in 80% of cases. About 5% of cases, called combined MCC, feature an admixture of neuroendocrine and non-neuroendocrine tumor cells. Reports of the presence or absence of MCPyV in combined MCC are conflicting, most favoring the absence, which suggests that combined MCC might have independent etiological factors and pathogenesis. These discrepancies might occur with the use of different virus identification assays, with different sensitivities. In this study, we aimed to determine the viral status of combined MCC by a multimodal approach. We histologically reviewed 128 cases of MCC and sub-classified them as "combined" or "conventional." Both groups were compared by clinical data (age, sex, site, American Joint Committee on Cancer [AJCC] stage, immunosuppression, risk of recurrence, and death during follow-up) and immunochemical features (cytokeratin 20 and 7, thyroid transcription factor 1 [TTF1], p53, large T antigen [CM2B4], CD8 infiltrates). After a first calibration step with 12 conventional MCCs and 12 cutaneous squamous cell carcinomas as controls, all eight cases of combined MCC were investigated for MCPyV viral status by combining two independent molecular procedures. Furthermore, on multiplex genotyping assay, the samples were examined for the presence of other polyoma- and papillomaviruses. Combined MCC differed from conventional MCC in earlier AJCC stage, increased risk of recurrence and death, decreased CD8 infiltrates, more frequent TTF1 positivity (5/8), abnormal p53 expression (8/8), and frequent lack of large T antigen expression (7/8). With the molecular procedure, half of the combined MCC cases were positive for MCPyV in the neuroendocrine component. Beta papillomaviruses were detected in 5/8 combined MCC cases and 9/12 conventional MCC cases. In conclusion, the detection of MCPyV DNA in half of the combined MCC cases suggests similar routes of carcinogenesis for combined and conventional MCC.

The HOXB13 gene is a member of the homeobox gene family, and prostate development depends on HOXB13 function. HOXB13 is a very sensitive and specific marker of prostate tissue and prostate cancer. When the origin of a tumor in a resection specimen or in biopsy material is unclear, it allows determining the prostate as the primary. Our aim was to determine whether HOXB13 has similar sensitivity for determining prostate origin of lymph node and bone metastases. We retrieved cases of lymph node and bone metastases of histologically confirmed prostate cancer (PCa) and selected lymph node metastases of urothelial carcinoma (UCa). A panel of antibodies against HOXB13, PSA, ERG, Androgen receptors, p504S, p63, GATA-3, CK7, and Uroplakin 2 and 3 was tested on these tissue samples. Two pathologists analysed and scored staining as either 0 (negative) or + (positive). The selected cohort consisted of 74 cases of lymph node and 15 of bone metastases of PCa and 15 of lymph node metastases of UCa. HOXB13 was expressed in 93 % of lymph node and in 33 % of bone metastases of PCa. All lymph node metastases of UCa were negative. Sensitivity of HOXB13 as a marker for prostate origin in lymph node metastases was 93 % and for bone metastases 33 %. Inter-observer variability in assessment of staining was good, as only two (1.9 %) of lymph node metastasis of PCa were discordant. HOXB13 is a useful marker for prostate origin when doubt exists regarding the site of the primary of a metastatic lesion. On bone metastases, HOXB13 immunohistochemistry performed less well, probably due to the use of tissue decalcification.

Immune thrombocytopenia (ITP) is an acquired antibody-mediated disease, for which splenectomy remains a curative treatment. We analyzed histology and phenotypes of ITP-splenectomy specimens from 41 adult patients, with different previous ITP-specific treatments, including B-cell-depleting rituximab (RTX) or not, in an attempt to predict splenectomy success or failure on the basis of day 56 postoperative platelet counts. RTX-naive ITP-spleen samples, compared with those from a 20-patient control trauma cohort, contained the following nonspecific, ITP-evocative, white-pulp lesions: follicular helper T-cell (programmed death-1 and inducible T-cell COStimulator) expansion in reactive follicles (P=0.01 and 0.03, respectively) and regulatory T-cell (FOXP3) expansion in the T-cell zone (P=0.049). On comparing ITP-splenectomy samples that would be successful with those that would be failures, only marginal zone hyperplasia differed (P=0.017). Indeed, 13/21 (61.9%) successful splenectomy samples exhibited marginal zone hyperplasia, as opposed to 1/9 (11.1%) failed splenectomy specimens. RTX impact on ITP-splenectomy samples was characterized by white-pulp (P=0.03) and marginal zone atrophies (P=0.01), and periarteriolar T-cell-zone hyperplasia (P<0.0001). The results of this novel comparative study of the histologic patterns of 41 ITP patients' evocative splenic lesions enabled clear description of different ITP morphologies and phenotypes, as a function of prior treatment and splenectomy success or failure.