Search

Background: Advances in molecular genetics of sarcoma have enabled the identification of type-specific aberrations. We aimed to assess the clinical effect of systematic implementation of molecular assays to improve sarcoma misdiagnosis., Methods: In this multicentre, observational study, we recruited patients from 32 centres of the French Sarcoma Group/Reference Network in Pathology of Sarcomas. Eligibility criteria included: biopsy or surgical resection; suspicion of: dermatofibrosarcoma protuberans (cohort 1), dedifferentiated liposarcoma (cohort 2), Ewing's sarcoma family of tumours (cohort 3), synovial sarcoma (cohort 4), alveolar rhabdomyosarcoma (cohort 5), and myxoid or round cell liposarcoma (cohort 6); review by one sarcoma-expert pathologist; availability of frozen material (except for cohort 1 of patients with dermatofibrosarcoma protuberans because anti-CD34 immunohistochemistry is performed on paraffin-embedded tissue); and patient information. For each case, the pathologist made one primary diagnosis followed by up to two differential diagnoses, based on histological characteristics only. Each diagnosis was classified as certain, probable, or possible. For each case to determine the molecular classification, we did fluorescence in-situ hybridisation on paraffin-embedded samples. We also did comparative genomic hybridisation and quantitative PCR (cohort 2) or reverse transcriptase PCR (cohorts 3-6) on frozen and paraffin-embedded samples. We made a final diagnosis based on the molecular results. The clinical effect of diagnosis correction was assessed by a board of experts., Finding: Between June 22, 2009, and Oct 30, 2012, 395 patients were enrolled in the study, of which 384 were eligible for inclusion. The diagnosis was eventually modified by molecular genetics for 53 patients: eight (16%) of 50 patients with dermatofibrosarcoma (cohort 1), seven (23%) of 30 patients with dedifferentiated liposarcoma (cohort 2), 13 (12%) of 112 with Ewing's sarcoma family of tumours (cohort 3), 16 (16%) of 97 patients with synovial sarcoma (cohort 4), seven (15%) of 46 patients with alveolar rhabdomyosarcoma (cohort 5), and two (4%) of 49 patients with myxoid or round cell liposarcoma (cohort 6), with an effect on primary management or prognosis assessment in 45 cases., Interpretation: Molecular genetic testing should be mandatory for diagnostic accuracy of sarcoma and appropriate clinical management, even when histological diagnosis is made by pathologist experts in this field., Funding: French National Cancer Institute and Nice University Hospital., (Copyright © 2016 Elsevier Ltd. All rights reserved.)

HMGA2, CDK4, and JUN genes have been described as frequently coamplified with MDM2 in atypical lipomatous tumor, well-differentiated liposarcoma, and dedifferentiated liposarcoma. We studied the frequency of amplification of these genes in a series of 48 dedifferentiated liposarcomas and 68 atypical lipomatous tumors/well-differentiated liposarcomas. We correlated their amplification status with clinicopathological features and outcomes. Histologically, both CDK4 (P=0.007) and JUN (P=0.005) amplifications were associated with dedifferentiated liposarcoma, whereas amplification of the proximal parts of HMGA2 (5'-untranslated region (UTR) and exons 1-3) was associated with atypical lipomatous tumor/well-differentiated liposarcoma (P=0.01). CDK4 amplification was associated with axial tumors. Amplification of 5'-UTR and exons 1-3 of HMGA2 was associated with primary status and grade 1. Shorter overall survival was correlated with: age >64 years (P=0.03), chemotherapy used in first intent (P<0.001), no surgery (P=0.003), grade 3 (P<0.001), distant metastasis (P<0.001), node involvement (P=0.006), and CDK4 amplification (P=0.07). In multivariate analysis, distant metastasis (HR=8.8) and grade 3 (HR=18.2) were associated with shorter overall survival. A shorter recurrence-free survival was associated with dedifferentiated liposarcoma (P<0.001), grade 3 (P<0.001), node involvement (P<0.001), distant metastasis (P=0.02), recurrent status (P=0.009), axial location (P=0.001), and with molecular features such as CDK4 (P=0.05) and JUN amplification (P=0.07). Amplification of 5'-UTR and exons 1-3 (P=0.08) and 3'-UTR (P=0.01) of HMGA2 were associated with longer recurrence-free survival. Distant metastasis was associated with shorter recurrence-free survival (HR=5.8) in multivariate analysis. Dedifferentiated liposarcoma type was associated with axial location, grade 3 and recurrent status. In conclusion, we showed that the amplification of HMGA2 was associated with the atypical lipomatous tumor/well-differentiated liposarcoma histological type and a good prognosis, whereas CDK4 and JUN amplifications were associated with dedifferentiated liposarcoma histology and a bad prognosis. In addition, we also provided the first description of the molecular evolution of a well-differentiated liposarcoma into four successive dedifferentiated liposarcoma relapses, which was consistent with our general observations.

Rare cancer issues have not been much explored yet because of their low incidence. That is why epidemiological studies have difficulties in identifying indisputable etiological risk factors. An expert opinion, mainly concentrated in some establishments, is required for these cancers' management. However, on account of the potential remoteness of these therapeutic resources, the patients' way of care remains also unstudied. By means of a geographical analysis of a regional exhaustive cohort of sarcoma, diagnosed in 2006 and 2007 and followed during five years at least, we can make progress on these different issues. Gastro-Intestinal and Stromal Tumors (GIST) occur more frequently in privileged territories while liposarcomas arise in more deprived areas. The association between liposarcomas and areas deprivation is significant (P=0.05). Moreover, pre-operative biopsy and some clinical patient characteristics, age, grade or tumor localization, are associated with an increase in the distance covered by patients for the first-line treatment (p ≤ 0,001). In the scope of an interdisciplinary collaboration, the geographical approach develops some hypothesis for rare cancers research, which must be tested by other larger scale studies.