Search

Background/Aim: Prediction of response to azacitidine (AZA) treatment is an important challenge in hematooncology. In addition to protein coding genes (PCGs), AZA efficiency is influenced by various noncoding RNAs (ncRNAs), including long ncRNAs (lncRNAs), circular RNAs (circRNAs), and transposable elements (TEs). Materials and Methods: RNA sequencing was performed in patients with myelodysplastic syndromes or acute myeloid leukemia before AZA treatment to assess contribution of ncRNAs to AZA mechanisms and propose novel disease prediction biomarkers. Results: Our analyses showed that lncRNAs had the strongest predictive potential. The combined set of the best predictors included 14 lncRNAs, and only four PCGs, one circRNA, and no TEs. Epigenetic regulation and recombinational repair were suggested as crucial for AZA response, and network modeling defined three deregulated lncRNAs (CTC-482H14.5, RP11-419K12.2, and RP11-736I24.4) associated with these processes. Conclusion: The expression of various ncRNAs can influence the effect of AZA and new ncRNA-based predictive biomarkers can be defined.

Simple Summary Although lncRNAs have been increasingly recognized as regulators of hematopoiesis, only several studies addressed their role in myelodysplastic syndrome (MDS). By genome-wide profiling, we identified lncRNAs deregulated in various groups of MDS patients. We computationally constructed lncRNA-protein coding gene networks to associate deregulated lncRNAs with cellular processes involved in MDS. We showed that expression of H19, WT1-AS, TCL6, and LEF1-AS1 lncRNAs associate with higher-risk MDS and proposed processes related with these transcripts. Abstract Background: myelodysplastic syndrome (MDS) is a hematopoietic stem cell disorder with an incompletely known pathogenesis. Long noncoding RNAs (lncRNAs) play multiple roles in hematopoiesis and represent a new class of biomarkers and therapeutic targets, but information on their roles in MDS is limited. Aims: here, we aimed to characterize lncRNAs deregulated in MDS that may function in disease pathogenesis. In particular, we focused on the identification of lncRNAs that could serve as novel potential biomarkers of adverse outcomes in MDS. Methods: we performed microarray expression profiling of lncRNAs and protein-coding genes (PCGs) in the CD34+ bone marrow cells of MDS patients. Expression profiles were analyzed in relation to different aspects of the disease (i.e., diagnosis, disease subtypes, cytogenetic and mutational aberrations, and risk of progression). LncRNA-PCG networks were constructed to link deregulated lncRNAs with regulatory mechanisms associated with MDS. Results: we found several lncRNAs strongly associated with disease pathogenesis (e.g., H19, WT1-AS, TCL6, LEF1-AS1, EPB41L4A-AS1, PVT1, GAS5, and ZFAS1). Of these, downregulation of LEF1-AS1 and TCL6 and upregulation of H19 and WT1-AS were associated with adverse outcomes in MDS patients. Multivariate analysis revealed that the predominant variables predictive of survival are blast count, H19 level, and TP53 mutation. Coexpression network data suggested that prognosis-related lncRNAs are predominantly related to cell adhesion and differentiation processes (H19 and WT1-AS) and mechanisms such as chromatin modification, cytokine response, and cell proliferation and death (LEF1-AS1 and TCL6). In addition, we observed that transcriptional regulation in the H19/IGF2 region is disrupted in higher-risk MDS, and discordant expression in this locus is associated with worse outcomes. Conclusions: we identified specific lncRNAs contributing to MDS pathogenesis and proposed cellular processes associated with these transcripts. Of the lncRNAs associated with patient prognosis, the level of H19 transcript might serve as a robust marker comparable to the clinical variables currently used for patient stratification.

Myelodysplastic syndromes (MDS) are a heterogeneous group of diseases with a high risk of transformation to acute myeloid leukemia (AML). One of the processes implicated in MDS pathogenesis is RNA splicing. Its alterations are caused by somatic mutations in splicing factor genes. Mutations in SF3B1 (Splicing Factor 3b Subunit 1) gene are the most frequently found mutations in MDS. Circular RNAs (circRNAs) are covalently closed RNAs that are produced by back-splicing process. CircRNAs can regulate multiple biological processes through various molecular mechanisms, such as microRNA sponging. Their deregulation is frequently found in cancer. It is likely that they also contribute to the development of MDS, however, their role in MDS has not been researched yet. Therefore, our aim was to explore circRNA levels in MDS and analyze their association with patient prognosis. We further hypothesized that mutations in splicing factor genes can affect production of circRNAs and thus, we examined circRNA levels with respect to the mutational status. We explored transcriptome of 78 MDS patients, 7 AML patients, and 13 healthy donors using Illumina RNA-seq of total RNA isolated from CD34+ bone marrow cells. To associate circRNA levels with mutational status, Illumina TruSight Myeloid Sequencing Panel Kit examining 54 genes was applied. Of 8,620 circRNAs identified by RNA-seq, 204 circRNAs were deregulated in MDS (e.g., MENTRNL, EBF1, and PPM1L-derived circRNAs) and 246 circRNAs were altered between lower- and higher-risk patients (e.g., CHST15, TMTC2, and PDE3B-derived circRNAs). Most of the progression-related circRNAs (n = 234) showed elevated levels in higher-risk patients, suggesting that the back-splicing process might be stimulated during the disease progression. In MDS patients with SF3B1 mutations, other 40 circRNAs were deregulated (e.g., ZNF91, ZEB1, and ZNF124-derived circRNAs). This circRNA profile was substantially different from the profiles associated with the rest of recurrently mutated splicing factor genes (SRSF2, U2AF1, and ZRSR2). To study alterations in forward- and back-splicing in SF3B1-mutated patients, we examined transcriptional differences on the levels of whole genes, transcript variants, and circRNAs and searched for specificities in transcription within individual gene loci. A set of circRNAs whose levels differed specifically without affecting expression of corresponding forward-spliced mRNAs included several oncology/hematopoiesis-relevant genes (e.g., ATM, CBL, ERCC5, ETV6, FLT3, and MAPK6). Gene loci with changed expression of both, alternative mRNA transcripts and circRNAs, but stable transcription on whole gene level included for example CDK14, KDM1A, and ZEB1. Because ZEB1 (Zinc Finger E-Box Binding Homeobox 1) serves as an essential hematopoietic transcription factor, we focused on ZEB1-derived circRNAs (hsa_circ_0000228 and hsa_circ_0003793) in more detail. We demonstrated that upregulation of these circRNAs is SF3B1-specific and not related to any other clinical or molecular characteristics. Finally, using RNA-seq data from CRISPR/Cas9 edited K562 cells (Liberante FG, et al., Sci Rep. 2019; 9:2678), we confirmed that SF3B1 K700E mutation leads to strong upregulation of ZEB1 circRNAs. To conclude, this is an early report showing for the first time that the levels of specific circRNAs are altered in MDS. We demonstrated that particular circRNAs may have potential to become markers that would contribute to more accurate prognosis of MDS patients. Further, we identified circRNAs with deregulated levels specifically in MDS patients with SF3B1 mutation, suggesting that this mutation affects circRNA production. Supported by GA CR (N20-19162S) and MH CZ-DRO (UHKT, 00023736). Disclosures No relevant conflicts of interest to declare.

Circulating microRNAs (miRNAs) are non-coding RNAs secreted into body fluids, and aberrant levels of these miRNAs correlate with diseases of various origins, making them highly potential clinical biomarkers. We investigated the spectrum of circulating miRNAs in the plasma of myelodysplastic syndrome (MDS) patients to identify miRNAs showing discriminatory levels in the patients with different prognosis. Plasma samples were analyzed with microarrays to define miRNA profiles, and the deregulated miRNAs were further studied using droplet digital PCR. With regard to the prognosis, the levels of miR-27a-3p, miR-150-5p, miR-199a-5p, miR-223-3p and miR-451a were reduced in higher-risk MDS. Multivariate analysis indicated miR-451a level as an independent predictor of progression-free survival (HR = 0.072, P = 0.006) and revealed a significant association of miR-223-3p level with overall survival (HR = 0.039, P = .032). Our data demonstrate that plasma levels of specific miRNAs are associated with MDS patient outcome and may add information beyond the currently used scoring systems.

Objectives Lenalidomide is a potent drug with pleiotropic effects in patients with myelodysplastic syndrome (MDS) with deletion of the long arm of chromosome 5 [del(5q)]. We investigated its effect on regulation of microRNA (miRNA) expression profiles in del(5q) patients with MDS in vivo. Methods We used miRNA expression microarrays to study changes in miRNA levels in peripheral blood CD14+ monocytes collected from patients before and during lenalidomide treatment and compared them with those from healthy donors. Results Before treatment, we observed strong upregulation of pro-apoptotic miR-34a and miR-34a* that diminished during lenalidomide exposure. Upregulation of HOX-related miR-196b and erythroid-specific miR-451 seen in untreated patients remained unchanged after the treatment. At the time of hematologic response, expression of several miRNAs clustering to the 14q32 locus was reduced. Additionally, we focused more deeply on miRNAs from the 5q commonly deleted region and found that levels of miR-378 and miR-378* followed haploinsufficiency trend. Conclusions This report describes changes in miRNA expression in del(5q) patients with MDS treated with lenalidomide, likely arising from deregulation of pathways implicated in lenalidomide action.

Introduction: Myelodysplastic syndromes (MDS) are hematopoietic stem cell disorders with a tendency to transform to acute myeloid leukemia. Several prognostic systems based almost entirely on clinicopathological characteristics have been developed. However, considering large heterogeneity of MDS, there is an ongoing substantial effort to identify new molecular markers, which would improve patient stratification. It has recently been discovered that various types of small noncoding RNAs (sncRNAs) are exported into blood circulation. Encapsulated in extracellular vesicles (EVs) or as a part of various molecular complexes, they play important roles in long distance cell-to-cell communication. Their secretion appears to be controlled and specific and thus may reflect (patho)physiological processes occurring in different cells. As diagnostic procedures move from bone marrow biopsies towards less invasive techniques, circulating sncRNAs have become of particular interest as potential novel semi-invasive biomarkers. Aims: We investigated circulating sncRNAs in MDS on the genome-wide level to better understand MDS pathogenesis and to search specific RNAs applicable as potential biomarkers. We analyzed paired samples from the whole plasma and EV fractions to define which of these two materials is better source of relevant sncRNA-based markers of MDS. Methods: We performed small RNA-seq in paired samples (plasma and EVs) in 42 patients and 17 healthy controls. To associate sncRNA levels with mutational status, Illumina TruSight Myeloid Sequencing Panel Kit examining 54 genes was applied. Overall survival (OS)‐associated sncRNAs were identified by performing univariate Cox regression along with a permutation test using BRB‐ArrayTools. Multivariate Cox regression analysis with backward variable selection was done to identify independent variables associated with patient survival. Results: Data analysis showed that the most abundant category of RNAs were miRNAs (51 %), followed by rRNAs, piRNAs, tRNAs, and mRNAs. Sample clustering revealed that RNA content of EV cargo is more homogenous than of plasma. Differential analysis identified a striking disbalance in RNA content between plasma and EVs in MDS patients (> 400 significantly deregulated sncRNAs) whereas these sncRNA profiles remained similar in controls. Such disproportion may suggest that mechanisms of sncRNA export into blood circulation may be specifically affected in MDS. As a result, these changes might substantially affect cell-to-cell communication of blood cells, contributing to the disease development. Further, we analyzed levels of individual sncRNAs and found that many hematopoiesis-related miRNAs were significantly increased in MDS patients compared to healthy controls, mostly both in plasma and EVs (e.g., miR-34a-5p, miR-125a-5p, and miR-150). Several miRNAs (e.g., miR-103b, miR-107, and miR-221) showed different levels between early and advanced MDS. Interestingly, the majority of miRNAs from the 14q32 miRNA cluster were specifically increased in early MDS. Additionally, we investigated possible impacts of somatic mutations on expression of circulating sncRNAs. However, only low numbers of circulating sncRNAs significantly associated with mutational status, suggesting no fundamental effects of somatic mutations on sncRNA export from MDS cells. Survival analysis identified sncRNAs with the highest level of association with OS (plasma: miR-1260b, miR-3191-3p, and miR-328-3p; EVs: miR-1237-3p, U33, hsa_piR_019420, and miR-548av-5p). To better stratify MDS patients, we defined risk prediction scores (separately for plasma and EVs) that combined effects of the above mentioned sncRNAs. The results showed that EV-combined score significantly increased the predictive power of the survival risk model (Figure 1). Moreover, Cox multivariate analysis revealed that this score is an independent variable the most significantly associated with OS (HR = 2.942, 95% CI 1.786 to 4.849, p < 0.001). Conclusions: Our data demonstrate that profile of circulating sncRNAs is specific in MDS and changes with the disease progression. Monitoring of sncRNA levels in EVs provides better prognostic performance for patient survival (compared to whole plasma), probably due to higher homogeneity of the EV fraction. Supported by grants 16-33617A and 00023736 from the Ministry of Health of the Czech Republic. Disclosures No relevant conflicts of interest to declare.

INTRODUCTION: Myelodysplastic syndromes (MDS) represent a heterogeneous spectrum of hematopoietic stem cell disorders characterized by inefficient hematopoiesis, peripheral blood cytopenia, dysplasia in one or more myeloid cell lineages, and a tendency to evolve into acute myeloid leukemia (AML). Hypomethylating agent azacitidine (AZA) is a standard therapy for patients with higher-risk MDS and AML with myelodysplasia-related changes (AML-MRC). The DLK1-DIO3 region is located on chromosome 14q32 and contains one of the largest miRNA clusters in the genome. Functions of these miRNAs have been linked to apoptosis and proliferation. Their expression is under the control of a differentially methylated region (DMR) located in the promoter of MEG3 gene. Significant overexpression of these miRNAs has previously been found in MDS and acute promyelocytic leukemia (APL). In APL, this upregulation has been linked to hypermethylation of MEG3-DMR. AIMS: We examined (i) expression of the 14q32 miRNAs and mRNAs (DLK1, RTL1, DIO3, MEG3, and MEG8) and (ii) DNA methylation of MEG3-DMR in patients with higher-risk MDS and AML-MRC before and during AZA treatment with the aim to reveal whether the overexpression of these miRNAs is epigenetically regulated in myelodysplasia. METHODS: The study comprised 24 paired samples of CD34+ bone marrow cells obtained from 12 patients with higher-risk MDS and AML-MRC before and during AZA therapy. miRNA profiling was done using Agilent microarrays and mRNA levels of the genes located within the 14q32 region were assessed by Illumina microarrays. Measurement of MEG3 level by RT-qPCR confirmed the microarray results on an independent cohort (81 MDS and 14 AML-MRC patients, and 13 controls). Methylation status of MEG3 promoter was investigated by amplicon bisulfite sequencing. RESULTS: We studied levels of miRNAs and mRNAs encoded in the DLK1-DIO3 region and found that 50 % of untreated patients showed strong upregulation of expression activity within the locus. Following AZA therapy, we observed a significant reduction in the miRNA/mRNA expression to nearly normal levels, suggesting that expression activity might be affected by hypomethylation. Interestingly, expression levels of miRNAs and mRNAs positively correlated, indicating that transcriptional activity was regulated uniformly throughout the region. Evaluation of expression data with clinical features in the pretreatment samples revealed that expression activity was related to BM blast count, patient diagnosis, and outcome. High expression was significantly associated with the diagnosis of AML-MRC and poor outcome, whereas low expression with MDS and favorable outcome. No chromosomal abnormality or cytogenetic risk category was linked to the expression changes. In the patients before treatment, we revealed a significant hypermethylation of a part of the MEG3-DMR that closely preceded the MEG3 start codon. This hypermethylation was seen in 50 % of the patients and negatively correlated with expression activity. Interestingly, the hypermethylation was associated with longer progression-free survival after initiation of AZA therapy (hazard ratio = 0.220, P = 0.03). Following AZA therapy, we identified an apparent decrease of methylation in another part of MEG3-DMR. This hypomethylation might be attributed to the hypomethylating effect of AZA and could therefore influence the deregulated expression of the miRNA cluster. Finally, we investigated the potential application of the deregulation observed in this locus for prognostic purposes regarding the responsiveness to AZA treatment. Although, we identified changes in expression and methylation levels before AZA therapy, these changes were not associated with future response status. CONLUSIONS: Here, we correlated changes in expression of 14q32 miRNAs and mRNAs with specific methylation changes in its regulatory sequences in higher-risk MDS and AML-MRC patients treated with AZA. We thoroughly investigated a complex regulatory mechanism within this region and based on our data, we assumed that epigenetic alterations affect expression activity in the locus. However, also other mechanisms such as influence of some transcription factors may play yet unrecognized roles in the resulting state. Supported by grants 17-31398A, 16-33617A, 16-33485A, and 00023736 (Ministry of Health of the Czech Republic). Disclosures No relevant conflicts of interest to declare.

INTRODUCTION: Myelodysplastic syndromes (MDS) are a heterogeneous group of clonal hematopoietic stem cell (HSC) disorders characterized by dysplastic and ineffective blood cell production and a tendency for transformation to acute myeloid leukemia (AML). Because of the heterogeneity of MDS, the development of additional molecular tools able to refine the Revised International Prognosis Scoring System (IPSS-R), predict outcome and monitor the response to treatment would be highly beneficial. Long noncoding RNAs (lncRNAs) are protein-noncoding RNAs that play multiple roles in hematopoietic differentiation. They contribute to the pathogenesis of hematologic malignancies, representing a new class of potential biomarkers and therapeutic targets. AIMS: In this study, we aimed to identify lncRNAs with deregulated expression in MDS that could serve as novel potential molecular markers of poor outcome. METHODS: The discovery cohort included 54 MDS patients, 14 AML patients with myelodysplasia-related changes (AML-MRC), and 9 controls. Agilent microarrays were used to profile lncRNAs expression in CD34+ bone marrow (BM) cells. Data analyses were focused on the identification of lncRNAs with altered levels between MDS patients (i) with (very) low vs. (very) high risk IPSS-R, and (ii) with progression-free survival (PFS) shorter vs. longer than 2 years. RT-qPCR was applied to validate expression levels of selected deregulated lncRNAs (H19, TCL6, LEF1-AS1, and WT1-AS) and levels of their protein-coding counterparts (LEF1 and WT1) in an independent validation cohort (79 MDS and 14 AML-MRC patients, and 13 controls). Further, we examined whether the expression of H19, TCL6, LEF1, and WT1 changed in patients treated with azacitidine (AZA). For this purpose, we retrieved data from our previous microarray analysis (GEO GSE77750) performed on a cohort of 19 MDS, 3 chronic myelomonocytic leukemia (CMML), 10 AML-MRC patients and 10 controls. RESULTS: In the MDS patients with higher risk IPSS-R, 23 lncRNAs were downregulated (e.g., CXADRP3, LEF1-AS1, PCED1B-AS1, SLC8A-AS1, and TCL6) and 35 lncRNAs were upregulated (e.g., BCAR4, FAM225A, FARP1-AS1, H19, and RBPMS-AS1) vs. lower risk IPSS-R (|logFC| ≥ 1, p ≤ 0.01). Furthermore, we identified significant upregulation of e.g. H19, SATB2-AS1, and WT1-AS in patients with PFS < 2 years vs. those with longer survival. For subsequent study, we selected H19, LEF1-AS1, TCL6, and WT1-AS lncRNAs based on their known association with hematopoiesis. In the validation cohort, we proved that downregulation of LEF1-AS1 and TCL6 and upregulation of H19 and WT1-AS were associated with poor PFS and OS (p < 0.01, Figure 1). Further, we correlated lncRNA expression levels with multiple clinical parameters (sex, age, BM blasts, hemoglobin, neutrophils, platelets, and IPSS-R-based cytogenetics) and found out that only level of H19 was independent from all these variables whereas expression of other lncRNAs was associated with BM blast count. Multivariate analysis confirmed that H19 level, BM blasts, and platelets remained independent factors for OS and PFS in MDS (p < 0.05). Importantly, WT1-AS level, platelets, and hemoglobin were identified as independent variables for survival of MDS patients with BM blast counts < 5 % (p < 0.05). Additionally, we measured levels of appropriate protein-coding counterparts of LEF1-AS1 (i.e. LEF1) and WT1-AS (i.e. WT1) and found strong positive correlation between these pairs (p < 0.001), indicating their co-transcriptional regulation. Because expression of the studied lncRNAs significantly varied between lower and higher risk MDS patients, we further investigated impact of AZA treatment on the expression of H19, TCL6, LEF1, and WT1 in higher risk disease. However, we detected neither any difference in their expression between responders and nonresponders nor any significant return to normal expression levels caused by AZA therapy. CONLUSIONS: Our data proved that particular lncRNAs are specifically deregulated in higher risk MDS. Altered transcript levels of these lncRNAs may serve as independent markers of poor survival, even in the patients with low BM blast counts, providing additional information about prognosis of MDS patients beyond the currently used scoring system. The work was supported by grants 17-31398A, 16-33617A, 16-33485A, and 00023736 from the Ministry of Health of the Czech Republic. Disclosures No relevant conflicts of interest to declare.

Background Myelodysplastic syndromes (MDS) are clonal disorders of hematopoietic stem cells characterized by ineffective hematopoiesis. High-risk MDS patients are treated by hypomethylating agents, of which they benefit significantly. However, only half of the patients respond positively to the treatment. Aberrant DNA methylation and mRNA expression in MDS were documented in several studies, but their prognostic impact in response to hypomethylating therapy is still unclear. The aim of the project was to find a relationship between methylation and expression status prior to azacytidine (AZA) treatment and the overall survival and clinical response of MDS patients. Methods We performed methylation and expression profiling in CD34+ cells from 30 samples from MDS patients before AZA treatment and after 4-8 treatment cycles. HumanMethylation27 BeadChips and HumanHT-12 v4 Expression BeadChips (Illumina) were used to generate profiles. DNA and RNA were isolated from same CD34+ cells separated from bone marrow by magnetic beads. The β-values represent quantitative measurements of DNA methylation levels of specific CpGs, and range from 0 for completely unmethylated to 1 for completely methylated DNA. The nonparametric Mann-Whitney test was used for comparison of β-values and expression levels between responders and nonresponders. Results To determine whether DNA methylation and expression might predict a response to AZA treatment, we compared methylation and expression status at baseline with clinical responses in 30 MDS patients. Twelve patients of 30 (40%) achieved complete remission or partial remission, 10 had stable disease (33.3%), and 8 showed progression (26.7%). Median survival after initiation of AZA treatment in progression patient group was 8.7 months, stable group 21.2 months, and group with complete or partial remission 24.5 months. We found significant differences in methylation status in 20 genes (p Conclusions Our finding of different DNA methylation levels at baseline between groups of responders and nonresponders as well as detection of decreased methylation after AZA treatment in the group of patients with clinical response may represent useful prediction markers of treatment success. However, the data require detailed examination along with confirmative cohort of patients. Supported by grant (NT/13899, NT/14377, NT/14539, NT/13847) and the project for conceptual development of research organization (00023736) from the Ministry of Health of the Czech Republic. Disclosures: No relevant conflicts of interest to declare.