Project 5: BCR-ABL Mutationen
Prevalence, kinetics and clinical impact of BCR-ABL mutated subclones within ATP and myristyl binding sites in chronic myeloid leukemia
Marie-Therese Ott1, Jenny Rinke1, Anja Waldau1, Katerina Machova2, Simona Soverini3, Andreas Hochhaus1, Thomas Ernst1
- Abteilung Hämatologie/Onkologie, Klinik für Innere Medizin II, Universitätsklinikum Jena, Jena, Deutschland
- Institute of Hematology and Blood Tranfusion, Prag, Tschechische Republik
- Department of Hematology/Oncology „L.eA. Seràgnoli”, University of Bologna, Bologna, Italien
Introduction: Point mutations within the BCR-ABL kinase domain are the most common cause for resistance to tyrosine kinase inhibitors (TKI) in chronic myeloid leukemia (CML). BCR-ABL mutant subclones (also described as “low-level” mutations) were frequently detected in previous studies. However, the prevalence, kinetics and prognostic significance of such subclones in a clinically well-characterized patient population needs to be systematically investigated.
Methods: A long-range PCR was optimized to generate large amplicons of BCR-ABL in order to cover all important functional regions of the translocated ABL gene including the P-Loop, A-Loop, SH3-, SH2-contact and the myristoyl binding pocket. Long-range PCR products were analyzed by targeted deep next-generation sequencing (NGS). Interlaboratory tests of 22 CML-cell-line samples were performed to evaluate the sensitivity, reliability and robustness of the method. Patient samples from the German CML-V (TIGER) study were analyzed prior to TKI treatment (n=103) as well as during treatment for those who reached major molecular remission (MMR) late or lost it (n=30). Furthermore, 9 patients treated with asciminib were studied in order to evaluate mutation dynamics including the myristoyl-binding site.
Results: The ring trial indicated a detection limit of mutations down to a frequency of 0.7%. In order to safely avoid false positive mutations caused by PCR artefacts a cut-off at 3% was chosen. A total of 30 different point mutations were identified in 21 of 103 CML patients (21%) at diagnosis. The median mutation load was 4% (range, 3 to 99%) with most mutations spreading outside the ABL kinase domain. In follow-up samples, more mutations appeared within the kinase domain. In two cases the proportion of mutations increased and correlated with high BCR-ABL levels and loss of molecular response. One patient treated with asciminib developed a blast crisis after 9 months of treatment. NGS revealed the F359V mutation within the ATP binding site with a mutation proportion of 99% at this date and 4% two months earlier.
Conclusions: Low-level BCR-ABL mutations can be frequently found by NGS in CML patients at diagnosis. However, only a minority of these subclones were selected and associated with loss of molecular response. NGS represents a sensitive and reliable technology for monitoring BCR-ABL mutant subclones in all functional regions including the myristyl binding site.