Nicolas Duployez, Romane Joudinaud, Augustin Boudry, Mathilde Hunault, Laurène Fenwarth, Emmanuelle Tavernier, Cécile Pautas, Sarah Bertoli, Suzanne Tavitian, Emmanuel Raffoux, Tony Marchand, Maël Heiblig, Sylvain Chantepie, Martin Carré, Pierre Péterlin, Maria Pilar Gallego-Hernanz, Romain Guieze, Célestine Simand, Pascal Turlure, Anne Huynh, Émilie Lemasle, Ludovic Gabellier, Juliette Lambert, Felipe Suarez, Samy Chraïbi, Laurence Sanhès, Karine Celli‐Lebras, Ariane Mineur, Claude Gardin, Norbert Ifrah, Norbert Vey, Régis Peffault de Latour, Lucie Rigolot, Isabelle Luquet, Dominique Penther, Raphaël A. Itzykson, Éric Delabesse, Jean‐François Hamel‐Broza, Stephane de Botton, Arnaud Pigneux, Hervé Dombret, Christian Récher, Claude Preudhomme, Pierre-Yves Dumas
ABSTRACT: FLT3 internal tandem duplications (FLT3-ITD) are major genetic events in acute myeloid leukemia (AML). Although the clinical impact of FLT3-ITD "macroclones" (allelic ratio [AR] ≥0.05) is well established, the significance of low-level FLT3-ITD subclones ("microclones") remains uncertain. We conducted a post hoc analysis of 1733 patients with newly diagnosed AML enrolled in the Backbone Intergroup 1 trial (ClinicalTrials.gov identifier: NCT02416388). Using next-generation sequencing (NGS), we detected FLT3-ITD microclones (AR between 0.0004 and 0.05) in 17.4% of patients without FLT3-ITD macroclones. Microclones and macroclones (low and high AR) were independently associated with increased relapse risk (cause-specific hazard ratio, 1.50 [95% confidence interval (CI), 1.18-1.91]; 1.98 [1.50-2.62]; and 2.33 [1.69-3.22], respectively) after adjustment for age, white blood cell count, other gene mutations, midostaurin treatment, and allogeneic hematopoietic stem cell transplantation. At 2 years, the cumulative incidence of relapse reached 42.5% (95% CI, 37.0-47.9) in patients with macroclones, 45.1% (38.3-51.6) in patients with microclones, and 29.4% (26.6-32.3) in patients without FLT3-ITD. In NPM1-mutated AML, both microclones and macroclones were associated with higher levels of measurable residual disease (MRD) and increased relapse risk, without independent impact on overall survival after adjustment for MRD. An analysis of paired samples further revealed that 41.8% of relapses in patients with FLT3-ITD microclones at diagnosis were associated with a macroclone at relapse. These findings challenge current risk stratification models and support the integration of NGS-based FLT3-ITD detection into the diagnostic and prognostic workflow for AML. Prospective trials addressing the management of patients with FLT3-ITD microclones are warranted, as is their consideration in future European LeukemiaNet guidelines.