Search

The levels of sPD-1 and sPD-L1 were analyzed in blood serum of 132 patients (age 14-70 years) with primary bone tumors: osteosarcoma (N=39), chondrosarcoma (N=42), Ewing sarcoma (N=9), chordoma (N=12), giant-cell bone tumor (GCBT) (N=16), benign neoplasms (N=14) and in and practically healthy subjects (age 19-58 years; N=27). sPD-L1 levels in all studied bone neoplasms were significantly higher than in the control. Serum sPD-1 level in GCBT patients was significantly higher than in the control, benign neoplasms, chondrosarcoma, and chordoma patients, but did not differ from osteosarcoma group. sPD-1 concentration in Ewing sarcoma was significantly higher than in chordoma and chondrosarcoma, but did not differ from the control. sPD-1 level in chondrosarcoma patients was also lower than in osteosarcoma, Ewing sarcoma, and in the control. Both sPD-1 and sPD-L1 concentrations were not significantly associated with the type of affected bone, process localization, disease stage, tumor histological grade, patients' age and sex. These results suggest the possibility of using these biological markers for preliminary assessment of the character of the process in the bone.

Chordomas are rare malignant tumours, which typically occur in the axial skeleton and skull base. They arise from embryonic remnants of the notochord. They constitute less than 5 % of primary bone tumours. They are characterised by their locally aggressive potential with high frequency of recurrences and a median overall survival of 6 years. The initial therapeutic strategy must be discussed in an expert centre and may involve surgery, preoperative radiotherapy, exclusive radiotherapy or therapeutic abstention. Despite this, more than 50 % of patients will be facing recurrences with few therapeutic options available at this advanced stage. This review aims to outline current treatment options available in chordomas, as well as discussing potentiality of new therapeutic approaches through their molecular characterization and the comprehension of their immunological environment., (Copyright © 2019 Société Française du Cancer. Published by Elsevier Masson SAS. All rights reserved.)

Thirty-five chordomas and more than 100 other tumors that have to be considered in the differential diagnosis, were immunohistochemically analyzed using a panel of antibodies including those to subsets of keratins (K), HBME-1, a monoclonal antibody recognizing an unknown antigen on mesothelial cells, and neuroendocrine markers. The patterns of immunoreactivities in chordoma were compared with those in renal cell carcinoma, colorectal mucinous adenocarcinoma, pituitary adenoma, skeletal chondrosarcoma, and extraskeletal myxoid chondrosarcoma (ESMC). Chordomas were consistently positive for keratin cocktail AE1/AE3, and for the individual keratins K8 and K19, and nearly always positive for K5, but they showed negative or only sporadic reactivity for K7 and K20. The keratin K8 and K19 reactivity was retained in those chordomas showing solid sheets of epithelioid, spindle cells, or cartilaginous metaplasia, and in one of two cases showing overtly sarcomatous transformation. In comparison, keratins were never present in skeletal chondrosarcoma, although K8 and to a lesser extent K19 were seen in occasional cases of ESMC with chordoid features. HBME-1 reacted strongly with chordoma and skeletal chondrosarcoma but was almost never positive in renal or colorectal carcinoma. These carcinomas lacked K5-reactivity, in contrast to chordoma. Chordomas were also consistently positive for neuron-specific enolase and occasionally focally for synaptophysin, but never for chromogranin. In contrast, pituitary adenomas regularly expressed the full spectrum of neuroendocrine markers and differed from chordoma by having a narrower repertoire of keratins, often showing negative or focal keratin 8- or AE1/AE3 reactivity and being almost always K19-negative. These findings indicate that chordoma can be immunohistochemically separated from tumors that can resemble it. Immunohistochemistry is especially useful in the diagnosis of small biopsy specimens that offer limited material for morphological observation.

A 67-year-old white man with a remote history of a chordoma of the clivus presented with myasthenia gravis. We investigated the possibility that these conditions were related immunologically. Tissue sections of various chordoma specimens were reacted with dilutions of patient serum and control serum by an indirect immunoperoxidase method. In addition, sections were reacted with antibodies to muscle antigens. Of six chordomas, five reacted positively to patient serum. None reacted to control serum. One chordoma reacted positively to desmin, and all six reacted positively to myoglobin. We propose that the patient may have produced antibody to muscle-like antigens of the chordoma that subsequently cross-reacted with acetylcholine receptor and led to clinical myasthenia gravis.

Seven chordomas, chondrosarcomas, and mucinous colorectal carcinomas, five liposarcomas, and five ependymomas were evaluated for the presence of epithelial membrane antigen (EMA) and carcinoembryonic antigen (CEA) by using an immunoperoxidase technique. All of the chordomas and mucinous carcinomas were cytoplasmic EMA positive, while the chondrosarcomas, liposarcomas, and ependymomas were negative. Among the tumors that were studied, only the mucinous carcinomas were positive for CEA. The EMA positivity of a chordoma reflects its epithelial nature and is a valuable aid in making the differential diagnosis between a chordoma and nonepithelial tumors, while the absence of CEA in a chordoma aids in making the distinction between a chordoma and a mucinous carcinoma.