Primary Ewing's sarcoma/primitive neuroectodermal tumor of the kidney: a clinicopathologic and immunohistochemical analysis of 11 cases.

Ewing's sarcoma/primitive neuroectodermal tumor (ES/PNET) is an extraordinarily rare primary tumor in the kidney and can be mistaken for a variety of other round cell tumors, including blastema-predominant Wilms' tumor (WT). Approximately 90% of ES/PNET have a specific t(11;22), which results in a chimeric EWS-FLI-1 protein. Immunohistochemistry for the carboxy-terminus of FLI-1 is sensitive and highly specific for the diagnosis of ES/PNET. WT-1, the WT-associated tumor suppressor gene, is overexpressed in WT but not in ES/PNET. No study has examined FLI-1 or WT-1 expression in renal ES/PNET. The clinicopathologic features of 11 renal ES/PNET were studied along with immunohistochemistry for cytokeratin, desmin, CD99, FLI-1, and WT-1. WT were also immunostained for CD99 (5 cases), FLI-1 (10 cases), and WT-1 (9 cases). The patients (6 men, 5 women) ranged from 18 to 49 years of age (mean, 34 yr). The mean tumor size was 11.8 +/- 3.8 cm (mean +/- standard deviation). Presenting symptoms included abdominal/flank pain and/or hematuria. Grossly, all tumors showed necrosis and hemorrhage, and 4 had cystic change. Microscopically, all tumors showed vaguely lobular growth, primitive round cells, and variable rosette formation. Epithelial, myogenous, or cartilaginous differentiation was not seen. Immunohistochemical results on the renal ES/PNET were cytokeratin (2/8 focal), desmin (0/9), CD99 (8/8), FLI-1 (5/8), and WT-1 (0/8). In comparison, the WT only rarely expressed CD99 (1/5) and did not express FLI-1 (0/10), but were usually WT-1-positive (7/9). Follow-up on 8 cases (mean, 28 mo; range, 6-64 mo) showed 4 lung and pleural metastases, 1 bone metastasis, liver metastasis, 2 local recurrences, and 5 deaths from disease (median time to death, 16.8 mo). No case had distant metastatic disease at presentation. Adjuvant therapy included chemotherapy (8 cases), radiation (3 cases), and bone marrow transplantation (1 case). Our study affirms a unique proclivity of renal ES/PNET for young adults and that it is a highly aggressive neoplasm, with rapid death in many cases, usually after the development of treatment-resistant lung metastases. These tumors must be distinguished from blastema-predominant WT and other primitive renal tumors that require different therapy. FLI-1 and WT-1 immunohistochemistry may be valuable in this differential diagnosis, given the known immunophenotypic overlap between ES/PNET and blastema-predominant WT with regard to CD99, cytokeratin, and desmin. The accurate distinction between these two entities has clear prognostic and therapeutic implications.

A call to standardize preanalytic data elements for biospecimens.

CONTEXT: Biospecimens must have appropriate clinical annotation (data) to ensure optimal quality for both patient care and research. Clinical preanalytic variables are the focus of this study. OBJECTIVE: To define the essential preanalytic variables (data fields) that should be attached to every collected biospecimen and to provide a complete list of such variables, along with their relative importance, which can vary, depending on downstream use, institutional needs, and information technology capabilities. DESIGN: The College of American Pathologists Diagnostic Intelligence and Health Information Technology Committee sponsored a Biorepository Working Group to develop a ranked list of the preanalytic variables for annotating biospecimens. Members of the working group were experts in anatomic, clinical, and molecular pathology; biobanking; medical informatics; and accreditation. Several members had experience with federal government programs, such as the National Cancer Institute's Biospecimens and Biorepository Branch and the National Cancer Institute's Community Cancer Center Program. Potential preanalytic variables were identified and ranked along with available supporting evidence, definitions, and potential negative effects if the variable was not attached to the biospecimen. Additional national and international stakeholders reviewed the draft manuscript. RESULTS: The ranked listing of 170 preanalytic variables produced can be used as a guide for site-specific implementation into patient care and/or research biorepository processes. Conclusions.-In our collective experience, it is often difficult to choose which of the many preanalytic variables to attach to any specific set of biospecimens used for patient care and/or research. The provided ranked list should aid in the selection of preanalytic variables for a given biospecimen collection.