DOI: 10.1200/jco-26-00280 ISSN: 0732-183X

Circulating Tumor DNA Profiling Across Phase II/III Ewing Sarcoma Trials Identifies High-Risk Patients With TP53 , STAG2 , and CDKN2A

Lieke Mous, Stelly Ballet, Sylvain Dureau, Camille Benoist, Eleonore Frouin, Victor Renault, Laura Kirton, Peter Peneder, Nina Jehanno, Anja L. Frei, Stéphanie Reynaud, Eve Lapouble, Anne-Sophie Plissonnier, Marie-Emmanuelle Legrier, Ellen Blanc, Aurélie Guinot, Jennifer Anderton, Piers Gaunt, Abigail Evans, Sandra J. Strauss, Manuela Krumbholz, Markus Metzler, Elizabeth A. Roundhill, Susan A. Burchill, Samuel Abbou, Sarah Winter, Martin G. McCabe, Nathalie Gaspar, Birgit Geoerger, Gudrun Schleiermacher, Eleni M. Tomazou, Nadège Corradini, Valerie Laurence, Bernadette Brennan, Perrine Marec-Berard, Olivier Delattre, Gaelle Pierron, Didier Surdez

PURPOSE

Ewing sarcoma is an aggressive bone cancer driven by FET::ETS fusions, with few recurrent secondary genomic alterations (most commonly involving STAG2 , TP53 , and CDKN2A ) and copy-number variants. Currently, diagnosis relies exclusively on tumor biopsies and imaging, and despite a well-defined genomic landscape, risk stratification remains based solely on clinicoradiologic parameters. Furthermore, tumor evolution and intrapatient heterogeneity remain poorly characterized, and prospective evidence supporting molecular risk stratification is limited.

MATERIALS AND METHODS

We prospectively integrated serial circulating tumor DNA (ctDNA) sampling into the first-line phase II/III Euro Ewing 2012 (EE2012) and CombinaiR3 trials, complemented by relapse studies. Across 347 patients, 991 blood, tumor, and bone marrow samples were profiled using custom targeted sequencing and shallow whole-genome sequencing to enable integrated biologic characterization and clinical correlative analyses.

RESULTS

ctDNA recapitulated established Ewing sarcoma genomic landscapes while also revealing intrapatient tumor heterogeneity when compared with matched tumor biopsy profiles. Blood and bone marrow plasma profiles were highly concordant. Longitudinal sampling identified selection of TP53 -altered clones at relapse. Clinically, detection of TP53 and/or STAG2 loss-of-function mutations and/or CDKN2A focal deletions in ctDNA at diagnosis predicted poor outcomes and, in a multivariable model adjusting for metastatic status, tumor volume, age, and treatment, defined an independent molecular high-risk subgroup with inferior event-free survival (hazard ratio [HR], 3.6 [95% CI, 2.2 to 5.8], P < .001) and overall survival (HR, 4.3 [95% CI, 2.5 to 7.3], P < .001) in EE2012 patients.

CONCLUSION

Prospective, trial-integrated ctDNA profiling provides comprehensive molecular characterization, reveals clinically relevant heterogeneity and clonal evolution, and identifies an independent molecular high-risk subgroup at diagnosis, supporting ctDNA-guided risk stratification in Ewing sarcoma.

More from our Archive