Clinical and genetic risk factors define two risk groups of extracranial malignant rhabdoid tumours (eMRT/RTK)

Karolina Nemes, Susanne Bens, Denis Kachanov, Margarita Teleshova, Peter Hauser, Thorsten Simon, Stephan Tippelt, Wilhelm Woessmann, Olaf Beck, Christian Flotho, Lorenz Grigull, Pablo H. Driever, Paul-Gerhardt Schlegel, Claudia Khurana, Kathrin Hering, Reinhard Kolb, Alfred Leipold, Floor Abbink, Maria J. Gil-da-Costa, Martin BeneschKornelius Kerl, Stephen Lowis, Carmen H. Marques, Norbert Graf, Karsten Nysom, Christian Vokuhl, Patrick Melchior, Thomas Kröncke, Reinhard Schneppenheim, Uwe Kordes, Joachim Gerss, Reiner Siebert, Rhoikos Furtwängler, Michael C. Frühwald*

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

Introduction: Extracranial rhabdoid tumours are rare, highly aggressive malignancies primarily affecting young children. The EU-RHAB registry was initiated in 2009 to prospectively collect data of rhabdoid tumour patients treated according to the EU-RHAB therapeutic framework. Methods: We evaluated 100 patients recruited within EU-RHAB (2009–2018). Tumours and matching blood samples were examined for SMARCB1 mutations by sequencing and cytogenetics. Results: A total of 70 patients presented with extracranial, extrarenal tumours (eMRT) and 30 with renal rhabdoid tumours (RTK). Nine patients demonstrated synchronous tumours. Distant metastases at diagnosis (M+) were present in 35% (35/100), localised disease (M0) with (LN+) and without (LN−) loco-regional lymph node involvement in 65% (65/100). SMARCB1 germline mutations (GLM) were detected in 21% (17/81 evaluable) of patients. The 5-year overall survival (OS) and event-free survival (EFS) rates were 45.8 ± 5.4% and 35.2 ± 5.1%, respectively. On univariate analyses, age at diagnosis (≥12 months), M0-stage, absence of synchronous tumours, absence of a GLM, gross total resection (GTR), radiotherapy and achieving a CR were significantly associated with favourable outcomes. In an adjusted multivariate model presence of a GLM, M+ and lack of a GTR were the strongest significant negative predictors of outcome. Conclusions: We suggest to stratify patients with localised disease (M0), GTR+ and without proof of a GLM (5-year OS 72.2 ± 9.9%) as ‘standard risk’. Patients presenting with one of the features M+ and/or GTR− and/or GLM+ belong to a high risk group (5-year, OS 32.5 ± 6.2%). These patients need novel therapeutic strategies such as combinations of targeted agents with conventional chemotherapy or novel experimental approaches ideally within international phase I/II trials.
Original languageEnglish
Pages (from-to)112-122
Number of pages11
JournalEuropean Journal of Cancer
Volume142
DOIs
Publication statusPublished - 1 Jan 2021

Cite this