HIGH SENSITIVITY MUTATION SCREENING AND CLONAL ANALYSIS ALLOWED BY ULTRA-DEEP AMPLICON SEQUENCING UNCOVER THE COMPLEXITY OF BCR-ABL MUTATION STATUS IN PATIENTS TREATED WITH TYROSINE KINASE INHIBITORS

Sborník

Background and Aims. point mutations within the Bcr-Abl kinase domain (KD) have been implicated in resistance to tyrosine kinase inhibitors (TKI) in chronic myeloid leukemia (CML) and Ph+ acute lymphoblastic leukemia (ALL) patients. Sanger sequencing (SS) is the recommended method for mutation screening, but it cannot identify mutant subpopulations <20%, nor can it discriminate polyclonal from compound mutations, unless it is preceded by cloning. The majority of patients positive by SS have a single mutant clone detectable, more rarely two or more; multiple mutations are thought to accumulate mainly after multiple lines of therapy. The recent development of ultra-deep amplicon sequencing protocols based on the Roche 454 next-generation-sequencing (NGS) technology has opened the way to a more accurate qualitative and quantitative characterization of the mutant clones that survive TKI therapy. Methods. longitudinal analysis of 92 samples collected at regular timepoints during treatment was retrospectively performed in 25 patients with TKI-resistant and 5 patients with TKI-sensitive CML or Ph+ ALL. NGS allowed to screen the Bcr- Abl KD with a lower detection limit of 0.1% and to reconstruct the clonal architecture of the mutated subpopulations, following quantitatively their evolution over time. SS was performed in parallel. Results. NGS revealed an unexpected complexity in the Bcr-Abl mutation status. In 94% of the samples, the higher sensitivity allowed to identify a variety of minor subclones (n=2-12 per sample; abundance,1%-18.8%) harbouring point mutations and more rarely insertions or deletions, alone or in addition to the dominant mutations detectable by SS. In all the cases, complexity was found to be high since diagnosis. The analysis of patients who later showed evidence of a single TKI-resistant mutation by SS (n=12) showed that, in all cases, additional low-level mutations were present - either in a small subfraction of the dominant mutated clone, or in an independent one, or both. The analysis of patients who accumulated multiple TKI-resistant mutations as assessed by SS showed that the newly acquired mutations could arise in the pre-existing mutated subclone(s) (2/13 patients), or in a previously wild-type subclone (2/13 patients), but they more often arose in parallel in both wild-type and mutated subclones (9/13 patients) - generating a complex jigsaw of multiple, competing populations harbouring different combinations of mutations. However, quantitative follow-up of these subclones showed that only one or a few take over, and some specific compound mutants (T315I+F359V; F317L+M351T..) were found to have greater ‘fitness’ over individual mutants, while others (T315I+E355G; Y253H+E255V..) have lower. Conclusions. mutation(s) detectable by SS are often the ‘tip of the iceberg’. A mosaic of small mutated subclones are present since diagnosis. Depending on their absolute and relative ‘fitness’, some may survive TKI therapy, although only one or a few will be capable to achieve dominance; acquisition of additional mutations dictates further dynamics of shrinkage/expansion. The level of heterogeneity is reduced only transiently when a highly-resistant subclone takes over. Reasoning on the basis of mutation(s) detectable by SS may not always be sufficient to predict responsiveness to a TKI or substantiate clinical decisions.

Supported by: PRIN; IGA-NT11155.

Haematologica, 2012; 97(s1):  453