Ependymosarcomas are rare tumors, with less than 20 cases reported in the literature [21
]. Reports include 2 cases of subependymoma with sarcomatous change [11
], 1 case of mixed subependymoma-rhabdomyosarcoma [20
], and 15 cases of sarcomatous change in ependymoma [7
]. The ependymomas described have been classified as WHO Grade II or Grade III [2
]. Seven of the 15 ependymosarcomas were described in the original resection specimens [7
], while 8 were described in recurrent tumor specimens [2
]. Seven of the 15 cases were described after the patient had received radiation therapy [2
]. Whether these tumors are radiation induced depends on a number of factors, including defining the time interval required between the radiation and the development of the sarcomatous component [18
Multiple studies over the years have attempted to elucidate the pathogenesis of gliosarcomas. Initially, the glial and mesenchymal components were thought to arise from different cells of origin as per the “polyclonal hypothesis”. The mesenchymal element was thought to develop from fibroblasts, pluripotent cells of the vascular adventitia or perivascular spaces, vascular smooth muscle cells, or monohistiocytic cells [1
]. According to the “monoclonal hypothesis”, the mesenchymal component develops from glial precursors during tumor progression [1
]. More recent molecular genetic studies have shown that both glial and mesenchymal components share common genetic aberrations in most instances, therefore supporting the monoclonal hypothesis [1
]. Chromosomal imbalances identified in gliosarcomas include gains on chromosomes 7, X, 9q, 12q, and 20q, and losses on chromosomes 10, 9p, 13q, and 17 [1
]. Genetic imbalances are similar in both glial and sarcomatous components in most cases studied.
It is of interest that molecular cytogenetic study of this example of ependymosarcoma demonstrated increased ploidy changes in the sarcomatous component. The sarcomatous component of this tumor also had polysomies of chromosomes 11 and 12 [18
]. In addition, another case of ependymosarcoma (of 4 cases tested) in the series by Rodriguez et al. (Case 2) also displayed polysomies of chromosomes 11 and 12 restricted to the sarcomatous component. This combination of findings is intriguing, and raises the possibility that ploidy changes may in part explain the development of this unique morphologic variation at the genetic level.
Rearrangements involving chromosomes 1 and 19 have been reported in multiple cancers [14
]. A t(1;19)(q12;p13) has been detected in 3 cases of malignant melanoma [15
]. The breakpoint in chromosome 19 appeared similar to that reported in precursor B-cell leukemia, and at the time the human insulin receptor gene, grossly mapping to this region, was felt to be the involved. However, subsequently it was discovered that precursor B-cell acute lymphoblastic leukemia has a PBX1
fusion gene characterized by either a der(19)t(1;19)(q23;p13.3) or t(1;19)(q23;p13.3).The probe set designed to identify these translocations [19
] was used in the FISH analysis of our case. A t(1;19)(q23;q13) has also been reported in a medulloblastom aand glioblastoma . Another case involving 19p13 is seen in t(1;19)(p22;p13.1), identified for the first time in a mixed epithelial stromal tumor (MEST) of the kidney [6
]. Not unlike ependymosarcomas, MEST are biphasic tumors composed of epithelial and mesenchymal elements. Lui et al. [13
] reported similar cytogenetic findings in a highly malignant primitive soft tissue sarcoma, suggestive of a rhabdomyosarcoma variant, with a hypertetraploid karyotype and t(1;19)(q12;q13.2). The increased ploidy and t(1;19) involving 1q12 are similar to the findings we report in the sarcomatous component in our case, although detailed mapping is lacking.
The relationship of these genetic alterations with prognosis is unclear at the present time. Sarcomatoid change occurring in tumors outside the nervous system portends more aggressive behavior. The prognosis in the prior ependyomosarcoma series was variable, not uniformly dismal [18
], and the patient in this report remains disease free 55 months after resection. The genetic alterations described in this case of ependymosarcoma are intriguing. Future studies should be of value in clarifying their clinical and/or biological significance, and identifying the specific genes involved in these alterations.