Context To overcome BCR::ABL inhibition, alternative signaling pathways may be activated in Ph-positive cells to compensate for the loss of BCR::ABL kinase activity. Therefore, Ph-positive cells will be able to proliferate and survive despite effective BCR::ABL inhibition. Various mutations in participating genes can lead to the activation of alternative signaling pathways. Objective Using NGS data annotation, identify BCR::ABL-independent genes with mutations of unknown clinical significance (VUS) and evaluate their impact on the development of resistance to tyrosine kinase inhibitor therapy. Patients The study included 55 patients divided into two groups. Group 1 – 44 patients with resistance and group 2 – 11 patients who responded to treatment. The clinical significance of the detected mutations was assessed using Franklin. Main Outcomes Measures The myeloid panel of 116 genes with an average reading depth of 200x or 1000x on MiSeq (Illumina) was used. The clinical significance of mutations was evaluated using Franklin. KEGG PATHWAY was used for functional annotation of genes. Results VUS mutations were detected both in group 1 (81%) and in group 2 (100%). Mutations in the TET2, KMT2D and NF1 genes were found in both groups. In group 1, mutations in 29 genes were identified. 29% of genes - genes of the FOXO and EGFR signaling pathways: PTEN, BRAF, ATM, KLF2, PIK3CB, STAT3. In 60% of patients with VUS in the FOXO or EGFR signaling pathways, MMR was never achieved despite multiple lines of therapy (p=0.023). It has also been noted that patients with FOXO or EGFR mutations have more than two additional chromosomal aberrations (ACAs). Median event-free survival in patients with VUS with FOXO or EGFR and ACAs is reduced compared with patients with VUS or ACAs alone (p=0.0072). This fact indicates the possible joint influence of these two factors on the development of resistance. Conclusions The largest number of mutations was found in genes involved in the FOXO and EGFR signaling pathways. Their association with ACAs emphasizes their importance in the development of resistance to TKI therapy.

Context The genetics of Ph-negative myeloproliferative diseases include not only driver mutations, but also mutations in a wide range of genes, as well as chromosomal abnormalities. Objective To assess the role of NGS profiling and cytogenetic analysis in evaluating the prognosis of the disease and the effectiveness of treatment for patients with Ph-negative myeloproliferative neoplasms. Patients or Other Participants The study included 118 patients (44 men and 74 women) with a diagnosis of polycythemia vera (30/118), essential thrombocythemia (29/118), or primary myelofibrosis (59/118). Cohort median age was 54 years (IQR 19-85). Main Outcomes Measures A panel of 118 genes with an average reading depth of 1000x on MiSeq was used. The clinical significance of mutations was determined using a limit of 3% variant allele frequency and COSMIC/Franklin databases. Chromosome analysis was performed using fresh bone marrow aspirates and G-banding with trypsin stain. For statistical analysis, the Kaplan-Meier method was used, with statistical significance assessed using the Cox-Mantel and chi-square tests. Results 47% of patients (56/118) had 1 to 5 pathogenic mutations (Me=1). The presence of any pathogenic mutation was associated with a decrease in EFS and OS (P=0.0001). The number of pathogenic mutations also affected the prognosis: in patients with ≥2 mutations, EFS and OS were significantly reduced (P<0.0001) compared to patients with fewer mutations. The cohort of patients (42/118) ever treated with ruxolitinib showed decreased OS and EFS for those with ≥2 pathogenic mutations (P=0.0376 and P=0.0046). Cytogenetic analysis was performed for 62/118 patients. 11/62 patients presented with cytogenetic abnormalities: 3/11 had del(13) and 1/11 each had +8; +8,del(9);del(11); del(7); +9,+t(1;9)+t(1;3); 1-2dup(1); der(22)t(1;22); -7; and inv(3)(q21q26),+6,+8,+9,-17,+3 with variations. The presence of a chromosomal aberration was associated with EFS and OS decrease (P=0.028 and P=0.0244), but it did not affect the treatment effectiveness of ruxolitinib (P=0.51 and P=0.58 for OS and EFS). The presence of cytogenetic abnormalities was significantly more common in patients with ≥1 pathogenic mutation (P=0.0061). Conclusions Combined cytogenetic studies and mutation analysis make it possible to predict the course of the disease and evaluate the effectiveness of therapy.

Context NPM1 mutations define the largest distinct genetic subset, 30% of AML, and is considered a favorable risk marker in the absence of FLT3-ITD mutation. Despite this, 30–50% of patients experience relapse, and the factors that drive relapse are still not fully understood. Objective To analyze the genomic heterogeneity of AML with NPM1 mutations using NGS and assess the impact of the most frequent aberrations on prognosis. Methods The study included 40 AML patients with mutated NPM1 and normal karyotype. Samples were analyzed by high-throughput sequencing using a custom panel consisting of 118 genes. Results All patients (40/40, 100%) had at least one mutation in addition to mutations in NPM1, with an average of 5 mutations per patient (1-9). The most frequently mutated genes were DNMT3A (55.0%), IDH1/2 (40.0%), TET2 (32.5%), FLT3-ITD (27.5%), PTPN11 (30%), FLT3-TKD (17.5%), N/KRAS (17.5%), RYR1 and GATA2 (12.5%). Patients with FLT3-ITD mutations had a 5-year overall survival (OS) of 34%, compared to patients without FLT3-ITD, 55% (p=0.001). We found a trend toward lower OS in patients with DNMT3AR882 mutations (p=0.153). We identified a small group of patients who lived more than 5 years after the diagnosis of AML, where there was a trend toward a decreased presence of myelodysplasia-related gene mutations (p=0.211). In the group of 29 patients without FLT3-ITD, 14 patients (48.3%) had relapse after consolidation chemotherapy. We compared mutations at diagnosis and relapse in 10 patients- only 2 patients relapsed with exactly the same mutations. Mutations in FLT3 and IDH1/2 were both acquired and lost at relapse. Mutations in NRAS were more frequently lost at relapse. Various cytogenetic abnormalities were acquired at relapse (4/10,40%). The mean number of mutations at diagnosis in relapse group was 5.5, compared to 3.9 in patients without relapse (p=0.041). Patients with NPM1/DNMT3A/IDH1/2mut combination showed a trend toward lower OS (p=0.09). Conclusions NPM1mut AML patients have a highly heterogeneous molecular genetic profile. The highest mutational load was found in patients with relapsed disease. The presence of FLT3-ITD and DNMT3A R882 mutations negatively affects OS. The combination of DNMT3A+IDH1/2 worsens OS in the NPM1mut/ FLT3-ITDwt group.

Topic: 3. Acute myeloid leukemia - Biology & Translational Research Background: Core-binding factor (CBF) acute myeloid leukemia (AML) encompasses AML with inv(16) and t(8;21). Despite sharing a common pathogenic mechanism involving rearrangements of the CBF transcriptional complex, there is growing evidence for considerable genotypic heterogeneity. Aims: To analyze the genomic heterogeneity of CBF-AML using NGS and assess the impact of the most frequent aberrations on prognosis. Methods: The study included 46 CBF-AML pts - 22 pts with inv(16) and 24 pts with t(8;21). Samples were analyzed by high-throughput sequencing on MiSeq platform (Illumina, USA). A panel consisting of 118 genes was developed based on literature data. Results: 97 mutations were found in 48 of 118 target genes: 60 mutations in AML pts with t(8;21) and 37 mutations in those with inv(16). In 21 genes, mutations were detected repeatedly (≥2 times). At least 1 mutation was detected in 95.8% of pts with t(8;21) and 95.5% of pts with inv(16). On average, significantly more mutations were found in AML pts with t(8;21) than in AML pts with inv(16), 2.6 versus 1.9 mutations, respectively (p=0.04). In AML pts with t(8;21), 41 different genes were mutated; with inv(16), 19 genes contained mutations. Of the 48 affected genes, mutations in 12 genes were common in both CBF-AML subgroups, whereas pts with t(8;21) had a higher proportion of private mutations (70.7% (29) versus 36.8% (7), respectively). Mutations in RTK/RAS signaling pathways were the most common events in the two types of CBF-AML, with the highest incidence in NRAS+KRAS (26.1%; t(8;21), 12.5%; inv(16), 40.9%), KIT (19.6%; t(8;21), 20.8%; inv(16), 18.2%) and FLT3 (4.3%; t(8;21), 0%; inv(16), 9.1%). Characterization of the molecular genetic profile of pts with t(8;21) AML: mutations in epigenetic regulation genes ASXL1 (16.7%) and EZH2 (4.2%), DNA methylation - DNMT3A (4.2%), cohesion complex - RAD21 (8.3%) and SMC3 (8.3%), MGA (16.7%), involved in MYC signaling pathway, SH2B3 (8.3%), SETBP1 (8.3%) that were absent in AML with inv(16) were found. Molecular genetic profile of pts with inv(16) AML: aberrations of genes involved in chromatin modification were rare events (in the ASXL1 gene - 0%), as were mutations related to the cohesion complex, which were absent in all cases. A mutation in the transcriptional corepressor BCORL1 was detected in one pnt. Approximately half (45.7%) of pts with CBF-AML had mutations in genes that activate signaling pathways, and the prognostic significance of such mutations was examined (Fig.1). A significant increase in the risk of relapse was found in pts with these mutations compared to pts without them (p=0.046). Pts with mutations in KIT gene had a significantly worse recurrence-free survival rate than pts without the mutation (p=0.031).Fig. 1. RFS of CBF-AML pts with and without mutations in genes that activate signaling pathways Summary/Conclusion: Patients with CBF-AML have a highly heterogeneous molecular genetic profile. Two CBF-AML groups hold their own unique genetic features, what makes it appropriate to analyze them separately in the research. The most frequent molecular events in CBF-AML are mutations in N/KRAS, KIT and FLT3 genes, which activate intracellular signaling pathways involving tyrosine kinases. Mutations in KIT gene and other mutations in genes involved in the activation of signaling pathways negatively affect the relapse-free survival of pts with CBF-AML. The use of the latest molecular genetic technologies, such as NGS, makes it possible to obtain a detailed characterization of the mutational profile for each pt, which is critical for CBF-AML pts, given the high frequency of mutations in them and the development of effective targeted therapy. Keywords: Core binding factor leukemia, Prognosis, Mutation analysis

S344dL/49%.This retrospective study was approved by the Ethics Committees of ANMRC in accordance with the Declaration of Helsinki.All patients signed informed consent.Patients: At the time of erythrocytosis onset median age was 46 (range, 20-55) yrs, median time of TKI treatment was 159 (range, 67-214) mos.Number of previous TKIs ranged from 0 to 4. Patients were under following TKIs: ponatinib (n=6), nilotinib (n=2) and bosutinib (n=1) with the best responses: complete hematologic response (n=2), molecular response 2 (n=2), at least major molecular response (n=5) at erythrocytosis onset.There was no history of using anabolic steroids, blood/erythropoietin (EPO) doping, or diuretics; 5/9 patients were smokers.Results: Median time from last TKI to erythrocytosis was 11 (range, 3-61) mos.At the last visit erythrocytosis was sustained in all patients except 1 who underwent stem cell transplantation.Median time of follow-up was 41 (range, 17-103) mos.All patients had a mean increase of Hgb 1.8 (range, 0.4-4.2) g/dL during TKI therapy.Median peak Hgb and Hct was 17.9 (range, 16.6-20.4)g/dL and 54.1 (range, 50.7-62.4)%, respectively.Phlebotomies were performed in 2 cases.No skin or microvascular symptoms, nor major thrombotic events were observed.All patients were negative for V617F and exon 12 JAK2 mutations.No overexpression of HIF1a or VHL were observed in 6/9 evaluable patients.All patients except one had EPO level within reference range.Testosterone levels were unremarkable in all patients.There was no evidence of EPOproducing tumors, renal artery stenosis or another renal disorders, as well as cardiac or pulmonary diseases.Conclusion: We reported erythrocytosis of unknown origin in CP-CML patients.Whether erythrocytosis might be related to TKI therapy is disputable and should be further investigated.

Background: Chronic myeloid leukemia (CML) is a myeloproliferative neoplasm associated with the presence of the chimeric BCR::ABL gene, which is treated with tyrosine kinase inhibitors (TKIs). The presence of mutations in the kinase domain of the BCR::ABL causes the development of resistance to TKIs. As mutations are not detected in 100% of resistant patients, the question arises about the causes of resistance to therapy in these patients. Next generation sequencing (NGS) makes it possible to analyze mutations in a wide range of genes, evaluate their impact on resistance to TKI therapy, and select the appropriate therapy for each patient. Aims: To study BCR::ABL-independent pathways of resistance to TKI therapy using modern molecular genetic technologies. Methods: The study included two groups of patients. 32 patients with resistant CML (18 men and 14 women) with a median age of 44 (IQR 14-74 years). The control group was 11 patients who responded to TKI therapy (5 men and 6 women) with a median age of 58 (IQR 33-75 years). The absence of mutations in the kinase domain of the BCR::ABL gene in patients of both groups was confirmed by direct Sanger sequencing. The presence of resistance to TKI therapy was assessed according to the ELN 2021 criteria. For the NGS study, a myeloid panel of 118 genes was used with an average reading depth of 1000x on a MiSeq instrument (Illumina). The clinical significance of the mutations found was assessed using the COSMIC, ClinVar, and VarSome databases. Survival was analyzed using the Kaplan-Meier method with statistical significance assessed using the Mantel-Cox test. Results: During the NGS study in the group with resistance to TKI therapy, genetic abnormalities were found in all patients: an average of 5 mutations (1 to 10 per patient). 95% (144/152) mutations had unclear clinical significance: most often such mutations occur in the tumor suppressor NF1 (14), transcription regulator genes TET2 (14) and ATRX (11), and cohesin complex gene STAG2 (8). Pathogenic mutations were found in the gene involved in the activation of the Ras/MAPK cell signaling pathway PTPN11 (1), epigenetic regulation genes EZH2 (1) and ASXL1 (4), proto-oncogene RHOA (1), transcription factor RUNX1 (1), and a gene involved in DNA methylation DNMT3A (1). In the control group, genetic abnormalities were detected in 81% (9/11) of patients: an average of 3 mutations (0 to 4 per patient). All variants are nonpathogenic: the most frequent mutations were found in the NF1 (4) and TET2 (3) genes. Pathogenic mutations in RUNX1 and ASXL1 genes have been noted to occur with resistant CML and are associated with an unfavorable course of the disease. The presence of pathogenic mutations reduces the overall survival of therapy-resistant patients with CML (Fig. 1).Fig.1. Survival of patients with resistance to TKI therapy depending on the presence of pathogenic mutations In two patients with pathogenic mutations in ASXL1 gene, despite several lines of therapy, a large molecular response was not achieved. In the future, these patients showed progression of the disease to the blast phase. Summary/Conclusion: The latest generation sequencing method makes it possible to detect pathogenic mutations in the PTPN11, EZH2, ASXL1, RHOA, DNMT3A, RUNX1 genes in patients with resistant CML, which significantly worsen survival. Application of this method can reveal additional BCR::ABL-independent resistance pathways and evaluate the course of the disease. Keywords: Chronic myeloid leukemia, Resistance

Topic: 15. Myeloproliferative neoplasms - Biology & Translational Research Background: Determination of “driver” mutations in the JAK2, CALR and MPL genes in patients with myeloproliferative neoplasms not associated with the Philadelphia chromosome (Ph-MPN) is the gold standard in diagnosis. However, the genomic landscape of such patients is very complex and is characterized by the presence of mutations not only in the driver genes. Such mutations can cause a more aggressive course of the disease. The next generation sequencing (NGS) method allows to perform simultaneous comprehensive analysis of an extensive panel of genes, which can significantly reduce the execution time and cost of the study. In addition, an important advantage of NGS is the ability to read the entire gene, including areas not studied by the standard Sanger sequencing method. This fact makes NGS an important tool in detecting pathogenic mutations and predicting the course of the disease. Aims: To evaluate the possibilities of using NGS technology in the diagnosis and determination of prognostic features of the course of the disease in Ph-MPN patients. Methods: The study included 31 patients (12 men and 19 women) aged from 27 to 85 years (Me=55 years). The Ph-MPN diagnosis was previously established in all patients: PMF (18/31), PV (3/31), ET (7/31), unspecified MPN – in 3/31 patients. All patients were analyzed for the presence of mutations in the driver genes – in 19/31 (61.2%) cases a mutation was detected in the JAK2 (V617F), 5/31 (16.2%) CALR, 3/31 (9.7%) MPL, 4/31 (12.9%) had no mutations in any of the driver genes (“triple-negative status”). In all patients, sequencing was performed using a myeloid panel of 121 genes with an average reading depth of 200x or 1000x on a MiSeq (Illumina) device. When analyzing the data obtained, a 3% threshold of allele frequency (VAF) was used. The clinical significance of mutations was established using COSMIC and ClinVar databases. To analyze the survival rate, the Kaplan–Meyer method was used with an assessment of statistical significance using the Cox-Mantel test. Results: During the NGS analysis, genetic abnormalities were detected in all patients. At the same time, mutations of somatic nature were detected in 90% of cases (28/31), on average 5 mutations in one patient (1-18 in one patient). In 94% of the studied samples (29/31), from 1 to 5 pathogenic mutations were detected (Me=2). Somatic mutations were found for 3 out of 4 patients with triple-negative status, which made it possible to confirm the clonality of the disease with the help of NGS and establish a diagnosis. Pathogenic mutations in “non-driver” genes were found in 16/27 patients. These genes perform various functions – epigenetic regulation (ASXL1 (7/31), TET2 (4/31), IDH1 (2/31), EZH2 (1/31), SETBP1 (1/31)), RNA splicing (SRSF2 (2/31), U2AF1 (2/31), DDX3X (1/31)), signal transmission (CBL (1/31), EP300 (1/31), KRAS (1/31), APC (2/31)) and chromatin remodeling (ATRX (2/31). Mutant variants of ASXL1 (7/31) were found in more than 10% of patients. The presence of mutations in the ASXL1 gene is associated with a deterioration in the median overall survival compared to patients without mutation (7,6 years and 12 years, respectively, p=0.043) (Fig. 1).Figure 1. Survival of patients with Ph-MPN depending on the presence of mutations in the ASXL1 gene, p=0.043 Summary/Conclusion: The study of the mutational profile of patients with Ph-MPN by NGS allows us to confirm the clonal nature of the disease. The use of the NGS method makes it possible to predict the course of the disease and determine indications for allogeneic HSCT, including Ph-MPN patients without driver mutations. Keywords: Thrombocythemia, Idiopathic myelofibrosis, Polycythemia vera, Myeloproliferative disorder

NGS data can confirm disease clonal nature, assess disease prognosis, and select target therapy for patients with Ph-negative myeloproliferative neoplasms. To get new information it is possible to expand the panel of sequenced genes, as well as study other genetic mechanisms underlying disease pathogenesis (e.g., changes in gene expression level, microRNAs, or transcription factors).

The detection of multiple mutations in patients with CML may explain resistance to TKI therapy; however, to better understand the pathogenesis of CML, it is important to know not only the presence of certain mutations but also their level of expression and to study the effects of various regulatory elements, such as transcription factors and miRNAs.

The use of NGS could find the additional targets for treatment of CML resistant patients besides TKI.

Splenectomy is frequently associated with thrombocytosis, its management and long-term outcomes in connection with diminished clonal hematopoiesis clearance are not yet well established.