Chordoma is a rare malignant tumor with notochordal differentiation, usually affecting the axial skeleton of young patients. We report a case of a high-grade epithelioid tumor involving the synovium and soft tissues of the knee in a 74-year-old male patient. The preliminary biopsy was inconclusive, but a diagnosis of metastatic clear-cell carcinoma of unknown origin was suggested. However, imaging studies did not reveal any primary lesions. The resection specimen consisted of nests and sheets of oval to polygonal cells with discernible cell borders, clear or lightly amphophilic cytoplasm, and round to oval nuclei with occasional well-visible eosinophilic nucleoli. Rare atypical mitoses, necrotic areas, and bizarre nuclei were noted. The biopsy and resection specimens underwent a wide molecular genetic analysis which included methylation profiling. The DKFZ sarcoma classifier assigned the methylation class chordoma (dedifferentiated) with a calibrated score of 0.96, and additionally, a loss of SMARCB1 locus was noted in the copy number variation plot. To verify these findings, T-brachyury and SMARCB1 immunostaining was performed afterward, showing diffuse nuclear positivity and complete loss in the tumor cells, respectively. To assess the prevalence of T-brachyury immunopositivity among SWI/SNF-deficient tumors and to evaluate its specificity for poorly differentiated chordoma, we analyzed a series of 23 SMARCB1- or SMARCA4-deficient tumors, all of which were negative. After incorporating all the available data, including the absence of any morphological features of conventional chordoma, the case was diagnosed as poorly differentiated chordoma. As illustrated herein, the utilization of methylation profiling in the diagnostic process of some carefully selected unclassifiable soft tissue neoplasms may lead to an increased detection rate of such extremely rare soft tissue tumors and enable their better characterization.
Keywords: Brachyury; Extra-axial chordoma; INI1; Knee; Methylation; Poorly differentiated chordoma; SMARCB1.
© 2023. The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.