Despite extensive characterization of the role of the EWS-ETS fusions, little is known about secondary genetic alterations and their clinical contribution to Ewing sarcoma (ES). It has been demonstrated that the molecular structure of EWS-ETS lacks prognostic value. Moreover, CDKN2A deletion and TP53 mutation, despite carrying a poor prognosis, are infrequent. In this scenario identifying secondary genetic alterations with a significant prevalence could contribute to understand the molecular mechanisms underlying the most aggressive forms of ES. We screened a 67 ES tumor set for copy number alterations by array comparative genomic hybridization. 1q gain (1qG), detected in 31% of tumor samples, was found markedly associated with relapse and poor overall and disease-free survival and demonstrated a prognostic value independent of classical clinical parameters. Reanalysis of an expression dataset belonging to an independent tumor set (n ¼ 37) not only validated this finding but also led us to identify a transcriptomic profile of severe cell cycle deregulation in 1qG ES tumors. Consistently, a higher proliferation rate was detected in this tumor subset by Ki-67 immunohistochemistry. CDT2, a 1q-located candidate gene encoding a protein involved in ubiquitin ligase activity and significantly overexpressed in 1qG ES tumors, was validated in vitro and in vivo proving its major contribution to this molecular and clinical phenotype. This integrative genomic study of 105 ES tumors in overall renders the potential value of 1qG and CDT2 overexpression as prognostic biomarkers and also affords a rationale for the application of already available new therapeutic compounds selectively targeting the protein-ubiquitin machinery.

Level III, therapeutic study. See the Instructions for Authors for a complete description of levels of evidence.

The MTHFR C667T polymorphism is not predictive for toxicity or overall survival, but could be used for relapse risk stratification.

Background. Solitary fibrous tumor (SFT) is a rare variant of soft tissue sarcoma (STS). Materials and Methods. We reviewed SFT patients (pts) treated at our institution between 12/1990 and 09/2017. Results. We identified 94 pts with a median follow-up (mFU) of 4.7 years (range: 0.1–21.53). Primary sites were the chest (33%), abdomen (21.3%), brain (12.8%), and extremities (9.6%); 6.4% of pts presented with synchronous metastasis. Median overall survival (mOS) from the first diagnosis was 56.0 months (m) (0.3–258.3). Doege–Potter syndrome was seen in 2.1% of pts. Primary resection was performed in 86 pts (91.5%). Median progression-free survival was 34.1 m (1.0–157.1), and 43% of pts stayed SFT-free during FU. Local recurrence occurred in 26.7% after a mFU of 35.5 m (1.0–153.8), associated with an OS of 45.1 m (4.7–118.2). Metachronous metastasis occurred in 30.2% after a mFU of 36.0 m (0.1–157.1). OS in metastatic pts was 19.0 m (0.3–149.0). Systemic therapy was given to 26 pts (27.7%) with inoperable/metastatic disease. The most common (57.7%) upfront therapy was doxorubicin, achieving responses in 13.3% of pts with a PFS of 4.8 m (0.4–23.8). In second line, pts were treated with ifosfamide or pazopanib, the latter achieving the highest response rates. Third-line treatment was heterogeneous. Conclusion. SFT is an orphan malignancy with a highly variable clinical course and a considerable risk of local failure and metachronous metastasis. Surgery is the only curative option; palliative systemic therapy is used in inoperable/metastatic cases but achieves low response rates. The highest response rates are seen with pazopanib in second/third line.

Soft-tissue sarcomas (STS) represent a heterogeneous group of rare, malignant tumors of mesenchymal origin. Reliable in vivo sarcoma research models are scarce. We aimed to establish and characterize histologically and molecularly stable patient-derived xenograft (PDX) models from a broad variety of STS subtypes. A total of 188 fresh tumor samples from consenting patients with localized or advanced STS were transplanted subcutaneously in NMRI-nu/nu-immunodeficient mice. Once tumor growth was observed, the material was passaged to a next generation of mice. A patient-derived tumor sample was considered "successfully engrafted" whenever the sample was transplanted to passage 1. A PDX model was considered "established" when observing stable morphologic and molecular features for at least two passages. With every passage, histologic and molecular analyses were performed. Specific genomic alterations and copy-number profile were assessed by FISH and low coverage whole-genome sequencing. The tumor engraftment rate was 32% (61/188) and 188 patient samples generated a total of 32 PDX models, including seven models of myxofibrosarcoma, five dedifferentiated liposarcoma, five leiomyosarcoma, three undifferentiated pleomorphic sarcoma, two malignant peripheral nerve sheet tumor models, and single models of synovial sarcoma and some other (ultra)rare subtypes. Seventeen additional models are in early stages of engraftment (passage 1-2). Histopathologic and molecular features were compared with the original donor tumor and were stable throughout passaging. The platform is used for studies on sarcoma biology and suited for in vivo preclinical drug testing as illustrated by a number of completed and ongoing laboratory studies.