Larger Cohort Of Chronic Myeloid Leukemia Patients Research Articles

Cytogenetics-Based Risk Prediction of Blastic Transformation of CML Treated with Tyrosine Kinase Inhibitors

Background:Blast phase (BP) of chronic myeloid leukemia (CML) remains a major therapeutic challenge in the era of tyrosine kinase inhibitor (TKI) therapy. Given the lack of effective treatment for BP, a potential key to improve patient outcome lies in identifying patients at high risk of progression and preventing onset of BP. Additional chromosomal abnormalities (ACAs) are important determinants of patient outcome. Despite its rarity in TKI era, BP develops in most CML patients in pre-TKI era regardless of ACAs. It is unknown how the development of BP is differently affected by ACAs in the TKI era.Design:We investigated the value of ACAs in prediction of BP in a large cohort of CML patients. Inclusion criteria included: 1, received TKIs as part of therapy; 2, age at diagnosis > 18 years; 3, presence of t(9;22) or variant translocations detected by conventional karyotyping. All cytogenetic abnormalities in Ph+ cells except -Y were considered ACAs. We examined the latencies from initial CML diagnosis to ACA emergence (Interval 1), from ACA emergence to BP (Interval 2) and survival after BP (Interval 3).Results:A total of 2,326 patients were included in the study, including 1,322 men and 1,004 women with a median age at diagnosis of 49 years (range, 18-88 years). In total, 570 had ACAs, including 333 (58.4%) with a single ACA and 237 (41.5%) with multiple ACAs. The most common single ACAs were: +8, +Ph, 3q26.2 rearrangement, i(17q), and -7/7q-, with a frequency of 13.5%, 12.6%, 7.5%, 6.6%, 3.9%, respectively. Of 237 patients with complex ACAs, 18 had 3q26.2 rearrangement, 19 had -7/7q-, and 30 had i(17q) as a component of complex ACAs. Seven patients had two of these three ACAs in different combinations. These complex ACAs were grouped together as high-risk complex ACAs. Complex ACAs without these three chromosomal abnormalities were designated as other complex ACAs.Patients with 3q26.2 rearrangement, -7/7q-, i(17q) and high-risk complex ACAs had a significantly faster blastic transformation, with a median Interval 2 of 2.8, 0, 32.6, 0 months, respectively, compared with 19.2 months in patients with other complex ACAs, and unreached in patients with +8, +Ph, or other single ACAs, and (Fig 1A). Based on the frequency and velocity of blastic transformation, patients were stratified into four risk groups. Patients without ACA formed the standard-risk (SR) group, patients with isolated 3q26.2 rearrangement, -7/7q-, or i(17q) and high-risk complex ACAs formed the high-risk (HR) group, patients with +8, +Ph or other single ACAs formed the intermediate-1 risk (IR-1) group, and patients with other complex ACAs formed the intermediate-2 risk (IR-2) group. By risk group, the median Interval 2 was unreached for IR-1group, 19.2 months for IR-2 group, and 1.9 months for HR group. (Fig 1B), and the interval from CML diagnosis to BP (Interval 1 + Interval 2) was unreached for SR and IR-1, 83.7 months for IR-2 patients, and 58.2 months for HR patients (Fig 1C). This risk of BP correlated well with patient survival from CML diagnosis (Fig 1D).The disease course of CML patients with ACAs follows a triphasic course, that is, Interval 1, 2 and 3. There was no or little variation in Interval 1 among three ACA groups (Fig 1E). No significant difference in Interval 3 was observed among three ACA groups neither (Fig 1F). SR patients had only slightly better survival of 17.0 months, compared with 9.3 months in other 3 groups combined (Fig 1F). The lack of difference in Interval 1 and 3 indicate that the predominant determinant of CML disease course in TKI era lies in the duration of Interval 2, which is in turn highly ACA-dependent.Conclusions:Chromosomal alterations predict the risk of blastic transformation in CML patients treated with TKIs. Patients can be stratified into four different cytogenetic risk groups based on ACA-dependent risk of blastic transformation. This risk stratification highly correlates with patient survival. By prolonging the duration of Interval 2, TKI therapy mitigates the risk of blastic transformation associated with low-risk ACAs or no ACAs but does not alter the natural course of CML associated with high-risk ACAs: 3q26.2 rearrangement, -7/7q- and i(17q). Thus, we identify a group of patients who are at a high risk of rapid disease progression. These patients may benefit from alternative treatment approach including early allogeneic hematopoietic stem cell transplantation to prevent blastic transformation. [Display omitted] DisclosuresCortes:ARIAD: Consultancy, Research Funding; ImmunoGen: Consultancy, Research Funding; Teva: Research Funding; Novartis Pharmaceuticals Corporation: Consultancy, Research Funding; Sun Pharma: Research Funding; Pfizer: Consultancy, Research Funding; BMS: Consultancy, Research Funding. Kantarjian:Bristol-Meyers Squibb: Research Funding; Amgen: Research Funding; Pfizer: Research Funding; ARIAD: Research Funding; Novartis: Research Funding; Delta-Fly Pharma: Research Funding.

Concomitant Administration of Direct Oral Anticoagulants in Chronic Phase Chronic Myeloid Leukemia Patients Treated with Tyrosine Kinase Inhibitors.

In the last decades, the chronic myeloid leukemia (CML) therapeutic landscape has changed dramatically with the introduction of tyrosine kinase inhibitors (TKIs), with 10-year survival rates improving to up to 80%. Long-lasting TKI treatment, in particular with second-generation TKIs, has enabled clinicians to manage the onset of several side effects and other co-morbidities, such as atrial fibrillation or venous thromboembolism (VTE). We retrospectively evaluated nine CML patients treated with TKIs between 2017 and 2020 who experienced atrial fibrillation or VTE and received concomitant administration of TKIs and direct oral anticoagulants (DOACs) outside clinical trials, to evaluate the efficacy and safety of this combination. Median age was 66 years at CML diagnosis (range 52-73 years) and 69 years at the time of starting DOACs. A female predominance was observed. The median follow-up of concomitant DOAC and TKI administration was 8.5 months; edoxaban was administered in six patients and apixaban in two patients, and one patient received rivaroxaban. Regarding CML treatment, four patients received imatinib, two patients bosutinib, and three nilotinib. In eight patients DOACs were started for atrial fibrillation and in one patient for VTE. In none of the patients treated with the combination were additional symptomatic thrombotic adverse events or major bleedings reported. In this small case series, the use of DOACs in CML patients seemed feasible. Additional data on long-term outcomes including a larger cohort of CML patients treated with DOACs are, however, needed.

High Incidence of Hepatitis B Viral Reactivation in Chronic Myeloid Leukemia Patients Treated with Tyrosine Kinase Inhibitors

Background: Reactivation of hepatitis B virus (HBV) is a well-recognized complication in patients with chronic HBV infection undergoing cytotoxic or immunosuppressive therapy. Currently, BCR-ABL1 tyrosine kinase inhibitors (TKIs) have been a standard of first-line treatment for chronic myeloid leukemia (CML) about last two decades. As there have been several case reports of HBV reactivation in CML patients treated with TKIs, the test for hepatitis B infection before initiating TKIs is recommended. Therefore, this study aimed to evaluate the risk of HBV reactivation in a large cohort of CML patients on TKIs treatment.Methods: We retrospectively reviewed the medical records of 1817 patients, who were diagnosed with CML and had available HBV serology between 2001 and 2017 from St. Mary's Hospital, Seoul, Korea.Results: Among 1817 patients, 76 patients (4.2%) were HBs-antigen (HBs-Ag) positive and 1741 (95.2%) were HBs-Ag negative. 302 patients (17.3%) of 1741 HBs-Ag negative patients were HBc-antibody positive indicating past infection of HBV. After exclusion of 7 patients who did not receive TKIs treatment, 69 patients were analyzed for HBV reactivation. The median age at diagnosis of CML was 42 (range 21-73) years and 44 (64%) were males. Median duration of follow-up from the recognition of HBs-Ag was 66.7 months (range, 2.4 - 192.7 months). Of 69 patients, a patient had been on antiviral therapy for chronic HBV infection before the diagnosis of CML. Eighteen patients (26% of 68) received antiviral prophylaxis prior to or concomitantly with the initiation of TKIs. Additional 4 patients were simultaneously initiated antiviral treatment with TKIs due to the evidence of active HBV infection such as the high titer of HBV DNA or elevated level of Alanine aminotransferase. Among 46 patients who did not receive antiviral prophylaxis, 12 patients (26%; 95% CI, 12-36%) eventually required antiviral treatment due to the development of active HBV hepatitis while on TKI therapy. HBV reactivation according to each TKI treatment was developed in 7 patients with imatinib, 2 patients in dasatinib, 1 patient in nilotinib, and 1 patient in radotinib therapy. One patient developed HBV reactivation permanently discontinued TKI therapy. Median time to HBV reactivation from CML diagnosis was 22.1 (range, 6.3-124.6 months) months. Median duration of antiviral therapy of 35 patients who received antiviral therapy was 38.2 months (95% CI, 25.7 ~ 50.7 months). 9 patients were off antiviral therapy at the last follow-up, 5 patients from prophylaxis group and 4 patients from treatment group. There was no difference in the duration of antiviral treatment between prophylactic and treatment groups. The initial choice of antiviral therapy as prophylaxis or treatment were entecavir for 14 (41% of 34), lamivudine for 8 (24%), tenofovir for 8 (24%), telbivudine for 3 (9%), and adefovir for 1 (3%).Conclusion: We evaluated HBV reactivation in a large cohort of HBs-Ag positive patients with CML from a single center, who received TKIs treatment. The reactivation rate of HBV was high at 26% without antiviral prophylaxis, which strongly support the routine screening of HBV serology prior to initiation of TKI therapy to identify HBs-Ag positive patients and the close monitoring of hepatic functions and HBV serology during TKI therapy. Given the high incidence of HBV reactivation with TKI treatment, concomitant antiviral prophylaxis with TKIs should be considered in HBs-Ag positive patients receiving TKI treatment. [Display omitted] DisclosuresKim:Ilyang: Research Funding; Novartis: Research Funding; BMS: Research Funding; Pfizer: Research Funding.

Experimental and integrative analyses identify an ETS1 network downstream of BCR-ABL in chronic myeloid leukemia (CML)

The BCR-ABL oncogene, the hallmark of chronic myeloid leukemia (CML), has been shown to activate several signaling pathways in leukemic cells. The natural history of this disease has been radically modified by tyrosine kinase inhibitors (TKIs). However, resistance to several lines of TKI therapies and progression to blast crisis (BC) remain significant concerns. To identify novel signaling pathways induced by BCR-ABL, we performed a transcriptome analysis in a BCR-ABL-expressing UT-7 cell line. More than 2000 genes differentially expressed between BCR-ABL-expressing and parental UT-7 cells were identified and ETS1 was found to be the most upregulated. ETS1 protein expression was also shown to be highly increased in UT-7 cells expressing BCR-ABL either constitutively or under the control of TET-inducible promoters. ETS1 expression is tyrosine-kinase dependent because it was reduced by TKIs. A significant increase of ETS1 messenger RNA (mRNA) expression was observed in blood cells from CML patients at diagnosis compared with healthy controls. Integration of publicly available chromatin immunoprecipitation sequencing and transcriptomic data with our results allowed us to identify potential ETS1 targets, some of which are involved in the progression of CML. The messenger RNA expression of two of these genes (DNM3 and LIMS1) was found to be associated with the absence of major cytogenetic response after 1 year of imatinib therapy. The present work demonstrates for the first time the involvement of the ETS1 transcriptional program in the experimental UT-7 model and a large cohort of CML patients.

Evaluation of the Risk of Hepatitis B Reactivation Among Patients with Chronic Myeloid Leukemia Treated with Tyrosine Kinase Inhibitors

▪Background: Immunosuppressive therapy is a known risk factor for hepatits B reactivation. The highest risk is reported in hematologic patients treated with anti-CD20 monoclonal antibodies, glucocorticoids and hematopoietic stem cell transplantation. Currently, treatment with tyrosine kinase inhibitors (TKIs) in the vast majority of chronic myeloid leukemia (CML) patients is life-long. Recently healthcare professionals were advised to test for hepatitis B infection before initializing therapy with TKIs and closely monitor HBV carriers. The risk is considered class effect with all TKIs and recommendations are based on case reports.Aim: The aim of the current study was to evaluate the risk of hepatitis B reactivation in patients with CML treated with TKIs.Methods: The records of patients with CML treated with TKIs from a single center were systematically reviewed for hepatitis B serology and serum biochemistry.Results: One hundred eighty one patients diagnosed with CML between the years 1983 and 2016 were evaluated. The median age at diagnosis was 49.7 years (range 18-89), 117/181 (65%) males, with median duration of follow up with TKI therapy of 5.3 years (range 0.4 to 33 years). Over a total of 1195 years of therapy with TKIs no cases of HBV reactivation were identified. Among 114 patients with hepatitis serology, 11 patients (10%) had evidence of prior resolved HBV infection (HBsAg negative/anti-HBc positive). Two of them had anti-HBs positive serology, one had negative PCR for HBV DNA and other two patients received lamivudine prophylaxis. Only one patient had HBsAg positive serology. None of the patients with positive hepatitis serology had clinical or biochemistry evidence of hepatitis B reactivation. The 67 patients without available hepatitis serology had normal liver transaminases at 6 months of TKI therapy and at last day of follow up, confirming that no overt hepatitis B reactivation occurred.Conclusions: We evaluated hepatitis B reactivation in a rather large cohort of CML patients, treated with TKIs. Although there were no cases of HBV reactivation during long term follow up, it should be emphasized that even a low incidence may exert a significant risk due to the long duration of treatment in a chronic disease with lifelong therapy. Our study infers that patients with serology of prior resolved HBV infection are at low risk for hepatitis B reactivation. Larger cohorts of patients with positive hepatitis B serology in a multicenter long term evaluation should be performed in order to address current recommendations for patients' safety and concerns. DisclosuresNo relevant conflicts of interest to declare.

Superoxide Dismutase 2 (Sod2) Expression Is Highly Decreased In Chronic Myeloid Leukemia (CML): Contribution To Genetic Instability In Bcr-Abl-Expressing Leukemic Cells

A increased oxidative stress is one of the hallmarks of chronic myeloid leukemia (CML). Manganese Superoxide dismutase 2 (SOD2) is a key scavenger enzyme against reactive oxygen species (ROS) in eukaryotic cells, playing a crucial role in antioxidant defense but its potential implication in the CML pathophysiology is not determined. We have analyzed the expression of SOD2 at the RNA level in a large cohort of CML patients (n=49) compared to a cohort of controls with non-malignant hyperleucocytosis (n= 19). This analysis showed a major decrease of SOD2 mRNA, inversely correlated with leucocytosis (p< 0.0001) and with Sokal Score, high-risk patients showing a lower expression of SOD2 (p= 0.019). To determine the effects of the silencing of SOD2 gene in an experimental CML model, we analyzed SOD2 expression in the human UT7 cell line expressing either native or BCR-ABL carrying T315I mutation. SOD2 protein levels in BCR-ABL expressing cells were similar to parental UT7 cells, suggesting that BCR-ABL has no direct effect on SOD2 expression. We have then silenced SOD2 expression in UT7 cells as well as UT7-BCR-ABL cells. Interestingly, the expression of H2O2 levels measured with flow cytometry was not changed after SOD2 inhibition, suggesting activation of compensatory anti-oxidant pathways. A global analysis of 27 anti-oxidant gene expression showed a reduction of PRDX2 only in BCR-ABL-expressing cells. To analyze same cells at the genomic level, we have performed CGH arrays in SOD2 silenced UT7, UT7-BCR-ABL and UT7-T315I cells as compared to their counterparts expressing SOD2. These experiments showed a major genetic instability with deletions and duplications in several chromosomal loci only in native or T315I-mutated BCR-ABL expressing cells silenced for SOD2. The 13q31.3-q34 locus, containing GPC5 and GPC6 genes, was found to be duplicated in non-mutated BCR-ABL cells deficient in SOD2 expression. GPC5 and GPC5 belong to the Glypican family, which are cell surface heparan sulfate proteoglycans anchored to the cell membrane via glycosyl-phosphatidyl-inositol-structures. The expression of several GPC is increased in some malignancies, including glioma, pancreatic carcinoma, breast cancer (GPC1), hepatocellular carcinoma (GPC3) and rhabdomyosarcoma (GPC5). Furthermore, the 13.q31.3 region, duplicated in BCR-ABL-expressing cells upon SOD2 deletion, contains the polycistronic mir-17-92 cluster (mir-17, mir-18a, mir19a, mir-20a, mir-19b1, mir-92a1), which is highly increased in CML CD34+ cells. Interestingly, the inhibition of SOD2 expression did not lead to the same genomic modifications detected by CGH arrays; the duplication of the 13q31.3 and 13q31.3q24 regions was not found in UT7 BCR-ABL cells harboring the T315I mutation. T315I cells silenced for SOD2 expression exhibited multiple abnormalities among which a large deletion of the 20q11.21 region which encodes the defensin gene family (DEFB115, DEFB116, DEFB118, DEFB119, DEFB121, DEFB122, DEFB123, DEFB124; REM1, DUSP15). Beta-defensins are antimicrobial peptides, which are part of the innate immunity against pathogens. The loss of beta-defensin 1 gene has already been reported in solid tumors. In conclusion, we demonstrate in a large cohort of CML patients, a major decrease of SOD2 expression, involved in antioxidant defense. We further show for the first time a link between loss of SOD2 expression and genetic instability in BCR-ABL expressing cells as demonstrated by CGH arrays. Genetic consequences of reduced SOD2 expression in CML should be further analyzed in prospective studies as patients with low SOD2 expression could be more prone to develop a mutator phenotype under TKI therapies. Disclosures:Guilhot:Novartis, BMS, Ariad, Pfizer: Honoraria. Turhan:Bristol Myers Squibb, Novartis: Consultancy, Honoraria.

The e13a2 BCR-ABL1 Fusion Transcript Is a Candidate Adverse Prognostic Factor In Chronic Myeloid Leukemia Patients Treated Frontline With Imatinib Mesylate

Background and aimsThe e13a2 and the e14a2 are the most frequent BCR-ABL1 transcripts in chronic myeloid leukemia (CML) patients. The aim of the present study was to assess the prognostic value of the e13a2 (b2a2) or the e14a2 (b3a2) fusion transcripts on the long-term outcome of early chronic phase (ECP) CML patients treated frontline with imatinib mesylate (IM). Few studies, mostly based on a small number of patients or with a relatively short observation, have been published, particularly in ECP setting. Due to partially conflicting data, the transcript type is not actually considered an established baseline prognostic factor. To our knowledge, this is the first long-term analysis conducted in the context of large prospective studies. MethodsA retrospective analysis of 3 GIMEMA CML WP prospective clinical studies (NCT00514488, NCT00510926, observational study CML/023) was performed. All the 559 enrolled patients were analyzed. A qualitative RT-PCR for BCR-ABL1 transcript was routinely performed at enrollment. Patients expressing rare transcripts or with both e13a2-e14a2 transcripts were excluded: 493 out of 559 patients were included, 203 (41%) with e13a2 transcript and 290 with e14a2 transcript (59%). Definitions. Major molecular response (MMR): BCR-ABL1IS ratio <0.1%. Deep molecular response (MR4.0): detectable disease ≤ 0.01% BCR-ABL1IS or undetectable disease with ≥10,000 ABL1 transcripts. Progression: transformation to accelerated or blastic phase. Failure: according to 2013 ELN recommendations. Event: failure or permanent discontinuation for any reasons. All the deaths, at any time and for any reasons, were included. ResultsThe median follow-up was 76 months; 95% of the patients had at least a 5-year observation. The 2 groups were comparable for demographic and hematologic characteristics, except for the proportion of male patients, that was higher in the e13a2 group (p = 0.05), and for the baseline percentages of eosinophils, that was lower in the e13a2 group (p = 0.009). The Sokal, Hasford and EUTOS risk score distributions and the proportion of patients with CCA in Ph+ cells were comparable. The complete cytogenetic response (CCyR) rate at 12 month was not significantly different (75% and 79% in e13a2 e14a2 patients, respectively, p=0.274), but the MMR rate at 18 months and the MR4.0 at 36 months were significantly lower in e13a2 patients (52% and 67%, p = 0.001; 20% and 30%, p = 0.013, respectively). As far as the rapidity of response, the time to MMR and the time to MR4 were significantly slower for patients with e13a2 transcript (12 and 6 months; 54 and 37 months, respectively), with an inferior overall estimated probability of MMR (85% and 92%, p < 0.001) and MR4 (57% and 71%, p = 0.002). The overall survival (83% and 86%, p = 0.023), the progression-free survival (81% and 86%, p = 0.007), the failure-free survival (54% and 71%, p < 0.001) and the event-free survival (46% and 61%, p = 0.003) were significantly lower in patients with e13a2 transcript (figure 1). In a multivariate analysis (Cox model), the transcript type retained its prognostic significance, when adjusted for other relevant variables. Analyzing separately the patients treated with IM 400 mg (n = 371; e13a2 and e14a2 transcript: 208 and 163 patients, respectively) or IM 800 mg (n = 122; e13a2 and e14a2 transcript: 82 and 40 patients, respectively), all the above mentioned response and outcome differences were confirmed, with one exception: in the high-dose IM group the overall survival probability was inferior for the e13a2 patients, but, probably due to the small number of patients, the difference was not statistically significant (83% and 86%, p = 0.268). [Display omitted] ConclusionsA long-term analysis in a large cohort of CML patients suggests that the e13a2 transcript is an adverse prognostic factor in ECP CML patients treated frontline with IM. An independent evaluation is required for confirmation. AcknowledgmentsUniversity of Bologna, BolognaAIL, COFIN, Fondazione Carisbo. Disclosures:Castagnetti:Novartis: Consultancy, Honoraria; Bristol Myers Squibb: Consultancy, Honoraria. Gugliotta:Novartis: Consultancy, Honoraria; Bristol-Myers Squibb: Consultancy, Honoraria. Abruzzese:BMS, Novartis: Consultancy.

Chromosomal Abnormalities in the Early Stem Cell Population of CML Patients in Complete Cytogenetic Remission.

Clonal cytogenetic abnormalities, most commonly involving chromosomes 7 and 8, are detectable by conventional karyotyping in Ph-negative metaphases of some chronic myeloid leukemia (CML) patients with a major cytogenetic response (MCyR) to imatinib. It is unknown whether these abnormalities involve the primitive progenitor cell compartment, and whether their frequency in this compartment may exceed the frequency detected by karyotyping. To answer these questions we analyzed lineage-negative CD34+/CD38− and CD34+/CD38+ cells from CML patients in complete cytogenetic response (CCyR), using by fluorescent in situ hybridization (FISH) for chromosome 7 and 8 abnormalities and BCR-ABL.Methods: Mononuclear cells (MNC) were selected from the bone marrow of patients with CCR by Ficoll-Hypaque density gradient centrifugation and enriched for lineage-negative cells using an immunomagnetic column. Lineage-negative cells were further sorted into CD34+/38− and CD34+/38+ cells by multicolor FACS. Interphase FISH analysis was performed using 7 LSI D7S522 Spectrum Orange / CEP7 Spectrum Green (chromosome 7), CEP8 Spectrum Aqua (chromosome 8) and LSI BCR/ABL +9q34 TriColor Dual Fusion Probe. Thus far, 5 CML patients with CCR (3 with a normal karyotype and 2 with trisomy 8) and 1 normal control have been analyzed.Results: Of the three CML patients in CCR with normal karyotype, one had 9% deletion 7q (internal cut off of 0.5%) and the second 1.2% trisomy 8 (just over the internal cut off of 1%) cells in the CD-34+/38− population, while the CD-34+/38+ cell population did not show abnormalities. The third CCR patient had no abnormalities in the CD-34+/38− and CD-34+/38+ cell populations. Two CML patients in CCR with 45% trisomy 8 abnormal cells by conventional cytogenetics had 44% and 60% trisomy 8 positive cells in the CD34+/38+ population (two few CD34+/CD38− cells were available for analysis). No BCR-ABL signal was detected in any cell. In the healthy control, the CD34+/38+ cells were normal, but 4.1% of CD34+/38− showed a deletion of 7q.Conclusion: Clonal chromosomal of chromosomes 7 and 8 in Ph-negative primitive hematopoietic progenitor cells may be more common than suggested by conventional karyotyping. A larger cohort of CML patients in CCR and healthy individuals is under study to determine if this phenomenon is indeed related to CML or occurs also in normals. Results will have implications for the interpretation of karyotypes in patients with hematologic malignancies.