Search

Several studies have reported that chronic myeloid leukaemia (CML) patients expressing e14a2 BCR::ABL1 have a faster molecular response to therapy compared to patients expressing e13a2. To explore the reason for this difference we undertook a detailed technical comparison of the commonly used Europe Against Cancer (EAC) BCR::ABL1 reverse transcriptase quantitative polymerase chain reaction (RT-qPCR) assay in European Treatment and Outcome Study (EUTOS) reference laboratories (n = 10). We found the amplification ratio of the e13a2 amplicon was 38% greater than e14a2 (p = 0.015), and the amplification efficiency was 2% greater (P = 0.17). This subtle difference led to measurable transcript-type dependent variation in estimates of residual disease which could be corrected by (i) taking the qPCR amplification efficiency into account, (ii) using alternative RT-qPCR approaches or (iii) droplet digital PCR (ddPCR), a technique which is relatively insensitive to differences in amplification kinetics. In CML patients, higher levels of BCR::ABL1/GUSB were identified at diagnosis for patients expressing e13a2 (n = 67) compared to e14a2 (n = 78) when analysed by RT-qPCR (P = 0.0005) but not ddPCR (P = 0.5). These data indicate that widely used RT-qPCR assays result in subtly different estimates of disease depending on BCR::ABL1 transcript type; these differences are small but may need to be considered for optimal patient management.

Introduction: The clonal hematopoiesis with somatic mutations is age-related phenomenon with a frequency around 10% for population older than 65 years in contrast to population younger than 50 years with frequency of 1%. Mutations in genes involved in epigenetic modification and RNA splicing, which are recurrently mutated in myeloid neoplasms and associated with increased risk of hematologic cancer, seem to represent a premalignant condition. Generally, ASXL1 mutations are frequently found in myeloid malignancies. Patients with chronic myeloid leukemia (CML) diagnosed at the age of 15 to 39 years, also called adolescent and young adults (AYAs), have a worse prognosis and response to tyrosine kinase inhbitors (TKIs) compared to elderly patients. Little is known about the molecular background differing AYA from the common group of CML patients. Objectives: To determine, whether the worse prognosis and response to therapy of CML AYAs is associated with the clonal hematopoiesis with somatic mutations. Methods: Samples from 22 AYAs were retrospectively analyzed at the time of diagnosis (aged 18-37; Table 1). Of them, 20 patients failed on TKI or relapsed after allo-HSCT (allogeneic hematopoietic stem cells transplantation). In 6/20 AYAs, mutations in the kinase domain of BCR-ABL1 were detected at the time of TKI failure (M244V, T315I, E255K/V + Q252H, F317L + M351T, V379I, L284S). Two responders were included for comparison. Sequencing of custom myeloid panel (Roche), partly or fully covering 36 genes frequently mutated in myeloid malignancies, was performed on MiSeq (Illumina). Data was analyzed in NextGENe software (Softgenetics). The detected variants were characterized by open-source databases (VarSome, Ensembl, COSMIC, NCBI - dbSNP) and confirmed by Sanger sequencing and/or ASO-ddPCR. Results: At the time of diagnosis, somatic mutations were identified in ASXL1 (n=4), CSF3R (n=1), TET2 (n=1), PCDHA12 (n=1), SETD2 (n=1), ATRX (n=1), and SIRT1 (n=1) in 10/20 AYAs, who subsequently failed on treatment (Table 1). Overall, 6 missense, 3 frameshift mutations and one nonsense mutation were detected. In patients #21 and #22 with optimal response to TKIs, no mutations were detected at diagnosis. In patient #10, ASXL1 mutation E773X was confirmed at the time of TKI failure and also at the allo-HSCT relapse. In patient #6, G645delinsGWfs was found at the diagnosis and on the 3rd line nilotinib treatment. Another ASXL1 mutation, S795delinsCLfs, was found in a patient #1 only at diagnosis. In patient #19, ASXL1 mutation T1372delinsTCfs found at diagnosis will be followed during the TKI treatment. In patient #3, the CSF3R mutation A593V was found at diagnosis and confirmed 14 months after the imatinib initiation. In patient #8, who relapsed after 2nd allo-HSCT, the RUNX1 D198N was found in the same clone bearing BCR-ABL1 T315I, both confirmed by ASO-ddPCR also before 1st allo-HSCT. This clone was, in the follow-up treatment, responsible for the relapse to CNS and also the relapse even after 3rd allo-HSCT and patient died. Conclusions: The preliminary data of this work outlined that somatic mutations in the myeloid genes are frequently found in CML AYAs, who failed on the TKI or relapsed after allo-HSCT, alone or together with mutated BCR-ABL1. The most frequently mutated gene was ASXL1, which is in line with the work by Ernst et al. (2018) even though on younger patients including children. Despite the clonal hematopoiesis with somatic mutations is considered as age-related phenomenon, in AYA CML patients, it may represent a critical problem in achieving sustained molecular response on solo TKI therapy, or even worse, it may result in higher risk of therapy failure and disease progression. Supported by MZCR 00023736 Table Disclosures Stoklosa: Janssen: Honoraria. Machova:Incyte: Consultancy; Angelini: Consultancy.

This work investigated patient-specific genomic BCR-ABL1 fusions as markers of measurable residual disease (MRD) in chronic myeloid leukaemia, with a focus on relevance to treatment-free remission (TFR) after achievement of deep molecular response (DMR) on tyrosine kinase inhibitor (TKI) therapy. DNA and mRNA BCR-ABL1 measurements by qPCR were compared in 2189 samples (129 patients) and by digital PCR in 1279 sample (62 patients). A high correlation was found at levels of disease above MR4, but there was a poor correlation for samples during DMR. A combination of DNA and RNA MRD measurements resulted in a better prediction of molecular relapse-free survival (MRFS) after TKI stop (n = 17) or scheduled interruption (n = 25). At 18 months after treatment cessation, patients with stopped or interrupted TKI therapy who were DNA negative/RNA negative during DMR maintenance (green group) had an MRFS of 80% and 100%, respectively, compared with those who were DNA positive/RNA negative (MRFS = 57% and 67%, respectively; yellow group) or DNA positive/RNA positive (MRFS = 20% for both cohorts; red group). Thus, we propose a “traffic light” stratification as a TFR predictor based on DNA and mRNA BCR-ABL1 measurements during DMR maintenance before TKI cessation.

PURPOSE: Here, we studied whether amplicon next-generation deep sequencing (NGS) could improve the detection of emerging BCR-ABL1 kinase domain mutations in chronic phase chronic myeloid leukemia (CML) patients under tyrosine kinase inhibitor (TKI) treatment and discussed the clinical relevance of such sensitive mutational detection. METHODS: For NGS data evaluation including extraction of biologically relevant low-level variants from background error noise, we established and applied a robust and versatile bioinformatics approach. RESULTS: Results from a retrospective longitudinal analysis of 135 samples of 15 CML patients showed that NGS could have revealed emerging resistant mutants 2-11 months earlier than conventional sequencing. Interestingly, in cases who later failed first-line imatinib treatment, NGS revealed that TKI-resistant mutations were already detectable at the time of major or deeper molecular response. Identification of emerging mutations by NGS was mirrored by BCR-ABL1 transcript level expressed either fluctuations around 0.1 %(IS) or by slight transcript level increase. NGS also allowed tracing mutations that emerged during second-line TKI therapy back to the time of switchover. Compound mutants could be detected in three cases, but were not found to outcompete single mutants. CONCLUSIONS: This work points out, that next-generation deep sequencing, coupled with a robust bioinformatics approach for mutation calling, may be just in place to ensure reliable detection of emerging BCR-ABL1 mutations, allowing early therapy switch and selection of the most appropriate therapy. Further, prospective assessment of how to best integrate NGS in the molecular monitoring and clinical decision algorithms is warranted.

Introduction: In chronic myeloid leukemia (CML) resistant to tyrosine kinase inhibitors (TKI), detection of mutations in the BCR-ABL1 kinase domain (KD) is routinely performed on transcript level. To determine the level of BCR-ABL1 KD mutation is important to follow kinetics of resistant CML cells and therapeutically prevent progression. However, the mutation types and levels are not always reliable predictors of subsequent dynamics of mutation-bearing clones and of corresponding clinical consequences (Willis, 2005; Khorashad, 2006; Preuner, 2012). DNA analysis enables more precise quantification of (sub)clonal levels and thus might be more reliable approach to monitor dynamics of BCR-ABL1 KD mutations. Aim: To study clonal evolution of resistant CML cells using genomic quantification of mutated BCR-ABL1 KD by droplet digital PCR (ddPCR). Methods: BCR-ABL1 mutation analysis on transcript level was performed using next generation sequencing (NGS) (Nextera XT; Illumina) and on DNA level using allele-specific ddPCR assays detecting T315I, E255K and Y253H (Bio-Rad). The level of genomic BCR-ABL1 mutation was determined as a copy number of mutation divided by a copy number of genomic BCR-ABL1 fusion. Quantification of genomic BCR-ABL1 was performed by ddPCR using patient-specific primers and probes designed to detect individual fusions. ALB (albumin) quantification was used as a control of DNA load/cell numbers. For analyses, mRNA and DNA extracted from KCL-22 cell line resistant to imatinib (IM) and from leukocytes of a patient who developed T315I during TKI therapy were used. Results: KCL-22 cell line is characterized by 2 Ph chromosomes and by ability to develop resistance by acquisition of BCR-ABL1 mutations early after the exposure to IM. We repeatedly found, that during early cultivation in the presence of IM, BCR-ABL1-T315I transcript increased up to maximum of 50%. Subsequently, after 2 months, BCR-ABL1-E255K transcripts became detectable and increased over time to 100%, while T315I decreased to un-detectable levels. To study the observed kinetics, we isolated 4 clones resistant to 4 µM IM that expressed 1) 50% of T315I, 2) 50% of E255K and 3) 30% of Y253H. In the fourth clone, no BCR-ABL1 mutation was detected, but mutation acquisition was found in KRAS, RUNX1 and ATRX. The levels of mutated BCR-ABL1 transcripts in mutation bearing clones remained stable over time. DNA analyses confirmed the same level of mutated BCR-ABL1 and revealed that in all resistant clones, only 1 Ph chromosome carried the BCR-ABL1 mutation (T315I, E255K or Y253H). Based on quantification of genomic BCR-ABL1 fusion and albumin we found, that the un-mutated BCR-ABL1 fusion was duplicated in Y253H clone, explaining the 30% level of Y253H. To follow a clonal evolution, we mixed the 4 KCL-22 resistant clones and analyzed BCR-ABL1 KD mutations at both mRNA and DNA levels during exposition to IM. We found that T315I clone overgrew other 3 clones in the mixture over time and 1 Ph chromosome remained mutated. These data confirm the T315I mutation being the most resistant; however, the data from the original cell culture, where the 100% E255K clone overgrew the 50% T315I cells, demonstrate, that a less resistant mutation might dominate the culture if present on both Ph chromosomes (as revealed by DNA analysis). We compared mRNA and DNA approach in 14 samples collected during individualized treatment management of a CML patient, who developed T315I during TKI therapy. The first mutation detection was during warning response preceded by eight samples negative by mRNA-NGS approach; DNA ddPCR analysis reliably detected T315I mutation in 7 of these 8 samples. Six mRNA positive samples were positive by DNA approach, which showed the same level of T315I. Conclusions: Allele-specific ddPCR together with quantification of BCR-ABL1 genomic fusion represents highly sensitive and reliable method providing fast and precise quantification of BCR-ABL1 mutations. A single DNA analysis is able to uncover clinically relevant events including BCR-ABL1 amplification or additional mutation acquisition, which presumably influence fitness of leukemic cells and clonal evolution during therapeutic interventions. The information provided by DNA mutational analysis may thus refine prediction of mutation kinetics and consequently improve management of progressed CML and Ph+ ALL. Support GACR 18-18407S, MZCR 00023736, AZV 15-31540, AZV 16-30186A Disclosures Klamova: Novartis: Consultancy, Honoraria; Bristol-Myers Squibb: Consultancy, Honoraria. Ernst:Novartis: Research Funding. Soverini:Incyte Biosciences: Consultancy; Novartis: Consultancy; Bristol Myers Squibb: Consultancy. Machova:Bristol-Myers Squibb: Consultancy, Other: Educational grant funding; Incyte: Consultancy; Novartis: Consultancy.