Survival In Myelodysplastic Syndromes Patients Research Articles

Granulocyte Colony-Stimulating Factor and Decitabine-Containing Conditioning Improves Survival in Myelodysplastic Syndrome Patients with Iron Overload Undergoing Allogeneic Hematopoietic Stem Cell Transplantation

Background: Iron overload is considered as an unfavorable prognosis in myelodysplastic syndrome (MDS) even in those undergoing allogeneic hematopoietic stem cell transplantation (allo-HSCT). Although iron chelation therapy has improved the prognosis of these patients to some extent, the effect has not yet been satisfactory. This study aimed to investigate whether granulocyte colony-stimulating factor and decitabine (G-DAC)-containing conditioning improve the prognosis of iron-overloaded MDS patients undergoing allo-HSCT. Study Design: One hundred and ninety-seven patients were enrolled in this retrospective study. Based on the level of serum ferritin (SF) and conditioning regimen, all patients enrolled were divided into 4 groups: SF﹤1000µg/L with G-DAC conditioning (cohort 1), SF﹤1000µg/L with non-G-DAC conditioning (cohort 2), SF ≥ 1000µg/L with G-DAC conditioning (cohort 3) and SF ≥ 1000µg/L with non-G-DAC conditioning (cohort 4). The clinical features and prognosis between cohorts were analyzed. Results: The 2-year overall survival (OS) was 77.0%, 72.1%, 73.8% and 38.7% ( P = 0.001), and disease-free survival (DFS) was 75.7%, 62.8%, 71.7% and 35.5% ( P = 0.001), and the cumulative incidence of non-relapse mortality (NRM) was 17.3%, 25.0%, 15.2% and 53.1% ( P = 0.001), and the incidence of relapse was 8%, 13.6%, 13.1% and 12.5% ( P = 0.592), respectively, in the four groups. To be specific, cohort4 had worse OS and DFS and higher NRM than the other three groups (all P﹤0.05). Multivariate analysis revealed that SF ≥ 1000µg/L was a risk factor for OS, DFS and NRM ( P = 0.022, P = 0.025, P = 0.036), while G-DAC-containin g conditioning was a protective factor ( P = 0.009, P = 0.003, P = 0.004). Intriguingly, when cohort 1 to cohort4 were included in the multivariate analysis, only cohort4 was a risk factor for OS, DFS and NRM (all P﹤0.05), but cohort3, namely iron-overloaded patients who received G-DAC-containing conditioning had no difference in prognosis compared with patients with SF﹤1000µg/L. Conclusions: The poor prognosis of patients with iron overload could be overcome by G-DAC-containing conditioning partly.

High BM plasma S100A8/A9 is associated with a perturbed microenvironmentand poor prognosis in myelodysplastic syndromes.

S100A8/A9 is a proinflammatory protein and plays an essential role in the pathogenesis of myelodysplastic syndromes (MDS) via the S100A8/A9-Toll-like receptors axis. While S100A8/A9 levels have been used as biomarkers in many inflammatory diseases, their clinical relevance has not been conclusively resolved in MDS. To address this, we used an enzyme-linked immunosorbent assay to quantify S100A8/A9 heterodimers in bone marrow (BM) plasma from 215 MDS patients and compared S100A8/A9 levels across patients with various disease risks and genotypes. S100A8/A9 levels correlated with ASXL1 variant allele frequencies significantly. Moreover, mutant ASXL1 with concurrent RUNX1, STAG2, ZRSR2, or EZH2 mutations was associated with higher S100A8/A9 levels. We further showed that higher S100A8/A9 independently predicted inferior leukemia-free survival and overall survival in MDS patients, irrespective of age, Revised International Prognostic Scoring System subgroups, and known detrimental mutations. Lastly, through deep-sequenced transcriptomic analysis, we demonstrated that higher S100A8/A9 in the BM intimated a perturbed microenvironment with enhanced myeloid-derived suppressor cell-mediated tumor immune escape signal, altered metabolism, and impairment in the functions and quantities of CD8+ T cells and NK cells. S100A8/A9 in the BM microenvironment may be a potential biomarker in the prognostication of MDS and target for novel therapy.

Association of systemic inflammatory and autoimmune manifestations with myelodysplastic syndromes: A systematic review and meta-analysis.

Systemic inflammatory and autoimmune manifestations (SIAMs) are frequently reported in Myelodysplastic syndromes (MDS). Studies focused on the impact of SIMAs on survival outcomes of MDS remains controversial. We performed this systematic review and meta-analysis to determine the association of SIAMs with overall survival, median survival, rate of acute myeloid leukemia transformation and mortality of MDS. An electronic search was conducted in 4 databases without any language restrictions, including PubMed, EMBASE, Medicine and Cochrane library up to April 30, 2021. The 18 studies included a total of 4603 MDS patients, of which 1175 (25.5%) patients had SIAMs. MDS patients with SIAMs had a statistically shorter overall survival compared with patient without SIAMs (Hazard ratio, 2.43; 95% confidence interval [CI], 1.34-4.41; P < .01). Our results were most compatible with no effect of SIAMs on median survival, rate of acute myeloid leukemia transformation and mortality (Median survival ratio, 1.16; 95% CI, 0.91-1.47; Odds ratio, 0.96; 95% CI, 0.63-1.45 and 1.2; 95% CI, 0.84-1.7, respectively). In this systematic review and meta-analysis, SIAMs appeared to have an adverse effect on overall survival of MDS patients. This finding suggested that SIAMs may be a potential independent prognostic factor for MDS.

HigherRUNX1 expression levels are associated with worse overall and leukaemia-free survival in myelodysplastic syndrome patients.

RUNX1 mutations are frequently detected in various myeloid neoplasms and implicate unfavourable clinical outcomes in patients with myelodysplastic syndrome (MDS) and acute myeloid leukaemia (AML). On the other hand, high expression of RUNX1 is also correlated with poor prognosis in AML patients. However, the clinical relevancy of RUNX1 expression in MDS patients remains elusive. This study aimed to investigate the prognostic and biologic impacts of RUNX1 expression in MDS patients. We recruited 341 MDS patients who had sufficient bone marrow samples for next-generation sequencing. Higher RUNX1 expression occurred more frequently in the patients with Revised International Prognostic Scoring System (IPSS-R) higher-risk MDS than the lower-risk group. It was closely associated with poor-risk cytogenetics and mutations in ASXL1, NPM1, RUNX1, SRSF2, STAG2, TET2 and TP53. Furthermore, patients with higher RUNX1 expression had significantly shorter leukaemia-free survival (LFS) and overall survival (OS) than those with lower expression. Subgroups analysis revealed that higher-RUNX1 group consistently had shorter LFS and OS than the lower-RUNX1 group, no matter RUNX1 was mutated or not. The same findings were observed in IPSS-R subgroups. In multivariable analysis, higher RUNX1 expression appeared as an independent adverse risk factor for survival. The prognostic significance of RUNX1 expression was validated in two external public cohorts, GSE 114922 and GSE15061. In summary, we present the characteristics and prognosis of MDS patients with various RUNX1 expressions and propose that RUNX1 expression complement RUNX1 mutation in MDS prognostication, wherein patients with wild RUNX1 but high expression may need more proactive treatment.

Platelet Doubling After First Decitabine Cycle Predicts Response and Survival of Myelodysplastic Syndrome Patients.

Although the effect of decitabine on myelodysplastic syndrome (MDS) has been demonstrated, merely a proportion of patients respond to therapy, and no well-recognized predictors have been identified. This study was conducted to investigate the effectiveness of decitabine in real-world clinical practice, and determine the predictive factors of response and overall survival (OS) in MDS patients. Clinical and pathological data were collected from 94 patients and analyzed. These patients were reclassified according to the 2016 World Health Organization classification criteria, and restratified by International Prognostic Scoring System prognostic scores. The response evaluation was performed according to the 2006 modified International Working Group response criteria. In this study, 62% of patients responded to decitabine. Among these patients, 15 patients (16%) obtained complete remission (CR), 15 patients (16%) obtained marrow CR with hematologic improvement (HI), 20 patients (21%) obtained marrow CR without HI, and 8 patients (9%) only obtained HI, and no patient botained partial remission. The OS of the responders was significantly longer than that of non-responders (67 months vs. 7 months, P<0.001). The OS in patients with and without platelet doubling was significantly different in both the low/intermediate and high/very high risk groups (P=0.0398 and P=0.0330). The multivariate analysis revealed that platelet doubling after the first decitabine cycle is an independent predictor of response and OS in MDS patients (P=0.002 and P=0.008). Decitabine is effective for treating MDS patients in real-world clinical practice. Furthermore, platelet doubling after the first decitabine cycle can be used as a predictor of response and survival in MDS patients.

Independent Prognostic Value of Flow Cytometry (FCM) in Myelodysplastic Syndromes (MDS) - Composition of a Prognostic FCM-Score for Overall Survival

Independent Prognostic Value of Flow Cytometry (FCM) in Myelodysplastic Syndromes (MDS) - Composition of a Prognostic FCM-Score for Overall Survival

MDS-439: A Simplified Three-Marker Panel for Myelodysplastic Syndrome Prognostic Outperforms the Well-Established Revised International Prognostic Scoring Systems (IPSS-R)

Context: Myelodysplastic syndromes (MDS) are a group of hematopoietic stem-cell disorders with heterogenous survival rates. In order to make better clinical treatment decisions, there is a constant need to classify the patients using baseline characteristics and improve existing prognostic scoring systems. Objective: The most popular prognostic scoring system for MDS patients is the Revised International Prognostic Scoring System (IPSS-R) (Greenberg et al., Blood, 2012). In this study, we aimed to develop a new prognostic scoring panel using a Cox proportional hazard model, which outperforms the IPSS-R model. Patients: A total of 982 MDS patients prior to hypomethylating agent (HMA) treatment were included in this study. Among them, 45% of patients were MDS-EB, 28% MDS-MLD, 12% MDS-RS, 6% MDS-SLD, 6% MDS-U, 2% 5q-, and the rest MDS-MLD-RS. The median time to follow-up was 12 months (range: 0–193 months) with a 36% incidence of death. Patients were randomly split into 75% training and 25% testing cohorts. Main Outcome Measures: The primary study outcome was overall survival, the time between first diagnosis and death or last follow-up, and probability of death for MDS patients treated at MD Anderson Cancer Center between 2009 and 2019. Results: We performed feature selections using the training set. The methods applied included Lasso, tree regression, univariate, and multivariate analyses. The features that are common across these four approaches are included in the study. Hemoglobin (as a continuous variable), cytogenetic level, and MDS subtype were the features that were selected consistently across multiple approaches. Our proposed model trained using the training set from the three aforementioned features showed AUC performances of 0.85, 0.86, and 0.93 in the testing cohort at 70, 60, and 50 months, respectively. These numbers correspond to an average of 4% AUC improvement compared to IPSS-R. Conclusions: We proposed a new panel comprised of only three markers, hemoglobin, cytogenetic level, and MDS subtype, with high AUC performance for prognostic applications. Although the new model utilizes fewer variables compared to the IPSS-R panel, it outperforms IPSS-R for predicting long-term survival in MDS patients, which might be the result of including disease subtypes and considering hemoglobin as a continuous variable.

The impact of CD33/CD34 ratio in the prognosis of myelodysplastic syndrome treated with azacitidine for first-line therapy

PurposeAzacitidine (AZA), a hypomethylating agent, is the standard treatment for patients with myelodysplastic syndromes (MDS). However, responses are generally transient in approximately 50% of patients. Cytomorphology and immunophenotyping are widely used in the treatment of MDS. Therefore, the present study investigated the relationship between immunophenotyping and the outcomes of MDS patients administered AZA as first-line therapy.MethodsWe retrospectively examined the relationship between immunophenotyping and the outcomes of 118 MDS patients administered AZA as first-line therapy.ResultsThe response to AZA was poorer in patients with a CD33/CD34 ratio > 1 than in those with a CD33/CD34 ratio < 1. Univariate and multivariate analyses identified higher serum LDH levels, higher risk IPSS-R, and a CD33/CD34 ratio > 1 as independent adverse prognostic factors for overall survival. A Kaplan–Meier analysis showed that a CD33/CD34 ratio > 1 was associated with poor survival.ConclusionIn the present study, a CD33/CD34 ratio > 1 in blast cells was associated with poor survival in MDS patients administered AZA as first-line therapy.

Increased serum C-reactive protein is an adverse prognostic factor in low-risk myelodysplastic syndromes.

Inflammatory cytokines play a role in hematopoiesis and development of myelodysplastic syndromes (MDS). Although increased serum levels of inflammatory cytokines are associated with poor survival in MDS patients, clinical management does not include assessment of inflammation. We investigated the significance of inflammation in MDS using serum C-reactive protein (CRP) levels, an indicator of the degree of systemic inflammation that can be used in routine practice. We hypothesized that serum CRP levels can be used to further classify low-risk MDS. We conducted a retrospective analysis of 90 patients with low-risk MDS, defined by the international prognostic scoring system (IPSS). We examined the prognostic relevance of CRP and known prognostic factors at diagnosis. Increased serum CRP (≥ 0.58mg/dL) was associated with poor survival(hazard ratio [HR]: 17.63, 95% confidence interval [CI] 5.83-53.28, P < 0.001) both overall and among the 73 patients with low-risk MDS as defined by the revised IPSS (HR: 28.05, 95% CI 6.15-128.04, P < 0.001). Increased CRP might predict poor prognosis and serum CRP levels can indicate clonal hematopoiesis and non-hematological comorbidity in patients with low-risk MDS.

Genomic Biomarkers to Predict Response to Hypomethylating Agents in Patients with Myelodysplastic Syndromes (MDS)

Genomic Biomarkers to Predict Response to Hypomethylating Agents in Patients with Myelodysplastic Syndromes (MDS)

Overall survival of myelodysplastic syndrome patients after azacitidine discontinuation and applicability of the North American MDS Consortium scoring system in clinical practice.

Azacitidine (AZA) is the standard treatment for myelodysplastic syndromes (MDS); however, many patients prematurely stop therapy and have a dismal outcome. The authors analyzed outcomes after AZA treatment for 402 MDS patients consecutively enrolled in the Italian MDS Registry of the Fondazione Italiana Sindromi Mielodisplastiche, and they evaluated the North American MDS Consortium scoring system in a clinical practice setting. At treatment discontinuation, 20.3% of the patients were still responding to AZA, 35.4% of the cases had primary resistance, and 44.3% developed adaptive resistance. Overall survival (OS) was better for patients who discontinued treatment while in response because of planned allogeneic hematopoietic stem cell transplantation (HSCT; median OS, not reached) in comparison with patients with primary resistance (median OS, 4 months) or adaptive resistance (median OS, 5 months) or patients responsive but noncompliant/intolerant to AZA (median OS, 4 months; P = .004). After AZA discontinuation, 309 patients (77%) received best supportive care (BSC), 60 (15%) received active treatments, and 33 (8%) received HSCT. HSCT was associated with a significant survival advantage, regardless of the response to AZA. The North American MDS Consortium scoring system was evaluable in 278 of the 402 cases: patients at high risk had worse OS than patients at low risk (3 and 7 months, respectively; P < .001). The score was predictive of survival both in patients receiving BSC (median OS, 2 months for high-risk patients vs 5 months for low-risk patients) and in patients being actively treated (median OS, 8 months for high-risk patients vs 16 months for low-risk patients; P < .001), including transplant patients. Real-life data confirm that this prognostic scoring system for MDS patients failing a hypomethylating agent seems to be a useful tool for optimal prognostic stratification and for choosing a second-line treatment after AZA discontinuation.

Myelodysplastic syndrome with t(6;9)(p22;q34.1)/DEK-NUP214 better classified as acute myeloid leukemia? A multicenter study of 107 cases.

t(6;9)(p22;q34.1)/DEK-NUP214 is a recurrent genetic abnormality that occurs in 1-2% of patients with acute myeloid leukemia (AML), and rarely in myelodysplastic syndrome (MDS). It has been suggested by others that all myeloid neoplasms with t(6;9)/DEK-NUP214 may be considered as AML, even when blast count is <20%. In this study, we compared the clinicopathologic features of 107 patients with myeloid neoplasms harboring t(6;9)/DEK-NUP214: 33 MDS and 74 AML. Compared with patients with AML, patients with MDS were older (p = 0.10), had a lower white blood cell count (p = 0.0017), a lower blast count in the peripheral blood (p < 0.0001) and bone marrow (p < 0.0001), a higher platelet count (p = 0.022), and a lower frequency of FLT3-ITD mutation (p = 0.01). In addition, basophilia was not a common feature in the patients of this cohort. Although there was no difference in overall survival between MDS and AML patients (p = 0.18) in the entire cohort, the survival curves did show a trend toward favorable survival in MDS patients. Multivariate analyses showed that initial diagnosis of MDS vs. AML and allogeneic hematopoietic stem cell transplantation were prognostic factors for survival of patients with t(6;9)/DEK-NUP214 (p = 0.008 and p < 0.0001, respectively). Our data suggest that MDS with t(6;9)/DEK-NUP214 is prognostically not equivalent to AML with t(6;9)/DEK-NUP214. These data also show that stem cell transplantation greatly improves the survival of MDS and AML patients with myeloid neoplasms associated with t(6;9)/DEK-NUP214.

LOC101928834, a novel lncRNA in Wnt/β-catenin signaling pathway, promotes cell proliferation and predicts poor clinical outcome in myelodysplastic syndromes.

Long non-coding RNAs (lncRNAs) play important roles in hematological malignancies. We have previously identified several differentially expressed lncRNAs in myelodysplastic syndromes (MDS) by microarray analysis. In the present study, we explored the regulatory circuitry, potential functions, clinical and prognostic relevance of these lncRNAs in MDS by developing a lncRNA regulation network. We identified a novel lncRNA, LOC101928834, which was significantly up-regulated in the bone marrow of patients with MDS and acute myeloid leukemia (AML). We further evaluated the clinical relevance of LOC101928834 in 89 MDS and 110 AML patients and found that higher level of LOC101928834 expression was associated with higher white blood cell count, higher blast percentage, the subtype of refractory cytopenia with excess blasts (RAEB) and shorter overall survival in MDS patients. Receiver operating characteristic (ROC) curve analysis showed that LOC101928834 expression could discriminate MDS-RAEB patients from control with an area under the receiver-operating curve (AUC) of 0.9048. Moreover, functional analysis showed that LOC101928834 promoted cell proliferation and cell cycle progression, and activated Wnt/β-catenin signaling pathway in vitro. In conclusion, LOC101928834 expression is correlated with clinical and biological features of MDS and may serve as a novel diagnostic and prognostic biomarker.

The impact of transfusion burden and comorbidities on the prognosis of patients with myelodysplastic syndromes

PurposeEarly comorbidity detection has been reported to be associated with treatment-related outcomes in several diseases. Two main goals of the present study were to investigate both the impact of comorbidities and transfusion frequencies on the survival and quality of life of patients with myelodysplastic syndromes (MDS). MethodsOne hundred and four MDS patients with a median International Prognostic Scoring System (IPSS) score of 0.5 (range: 0–3) were included in the study. Almost half of the patients had more than one comorbidity. ResultsMedian short form health surveys (SF)-36 mental and physical scores were 42.1 (range: 20.6–66.1) and 38.7 (range: 18–59.7), respectively. Mean scores of the Eastern Cooperative Oncology Group (ECOG) performance scales at diagnosis and during recruitment were 1.0 (1.4 ± 1.0) and 2.0 (1.8 ± 1.1), respectively. The mean Charlson Comorbidity Index (CCI) score was 1.0 (1.4 ± 1.5). In the model that was constructed using variables with a p value < 0.100 in the univariate analysis, factors that predicted death were refractory anemia with excess blasts (RAEB) and ECOG scores at recruitment. When ECOG was removed from the model, RAEB and CCI at diagnosis moved to the forefront as mortality predictors. ConclusionThis study demonstrated that both CCI and ECOG performance status had an impact on survival in MDS patients who had low IPSS scores. ECOG stood out as a better and more practical predictor of survival than CCI, especially after considering its (ECOG) ease of use.

Inhibition of LSD1 in MDS progenitors restores differentiation of CD141Hi conventional dendritic cells.

The use of immunotherapy to treat patients with myelodysplastic syndromes (MDS) shows promise but is limited by our incomplete understanding of the immunologic milieu. In solid tumors, CD141Hi conventional dendritic cells (CD141Hi cDCs) are necessary for anti-tumor immunosurveillance and the response to immunotherapy. Here, we found that CD141Hi cDCs are reduced in MDS bone marrow and based on the premise established in solid tumors, we hypothesized that reduced numbers of CD141Hi cDCs are associated with inferior overall survival in MDS patients. We found that MDS patients with reduced numbers of CD141Hi cDCs, but not other DC populations, showed reduced overall survival. To examine the basis for reduction in CD141Hi cDCs, we found fewer numbers of progenitors committed to DC differentiation in the MDS bone marrow and these progenitors expressed lower levels of interferon regulatory factor-8 (IRF8), a master regulator of CD141Hi cDC differentiation. To rescue impaired CD141Hi cDC differentiation, we used pharmacologic inhibition of lysine-specific demethylase 1A (LSD1) to promote CD141Hi cDC differentiation by MDS progenitors. These data reveal a previously unrecognized element of the MDS immunologic milieu. Epigenetic regulation of CD141Hi cDC differentiation offers an intriguing opportunity for intervention and a potential adjunct to immunotherapy for patients with MDS.

Prognostic significance of U2AF1 mutations in myelodysplastic syndromes: a meta-analysis.

IntroductionAlthough the effects of U2 small nuclear RNA auxiliary factor 1 gene (U2AF1) mutations on the outcomes of patients with myelodysplastic syndromes (MDS) have previously been investigated, their prognostic significance remains controversial. We performed a meta-analysis to investigate the impact of U2AF1 mutations on MDS progression.MethodsTwo reviewers independently extracted information such as hazard ratios (HRs) and 95% confidential intervals (CIs) for overall survival (OS) and leukemia-free survival (LFS) as well as the number of surviving patients each year after diagnosis from the included studies.ResultsThirteen studies with a total of 3038 patients were included. The summary odds ratio (OR) for U2AF1 mutations with an OS of 5 years was 0.37, the summary HR for U2AF1 mutations in OS was 1.60, and the summary OR for an OS of 5 years in patients with U2AF1S34 and U2AF1Q157 was 3.68. There were no significant differences in leukemia-free survival or hypomethylating therapy response between patients with and without U2AF1 mutations.ConclusionU2AF1 mutations were associated with poor survival in MDS patients, and patients with U2AF1Q157 had a worse OS than those with U2AF1S34. Our findings suggest that MDS patients with U2AF1 mutations could benefit more from hypomethylation therapy.

Screening and identification of key candidate genes and pathways in myelodysplastic syndrome by bioinformatic analysis.

Myelodysplastic syndrome (MDS) is a heterogeneous hematologic malignancy derived from hematopoietic stem cells and the molecular mechanism of MDS remains unclear. This study aimed to elucidate potential markers of diagnosis and prognosis of MDS. The gene expression profiles GSE19429 and GSE58831 were obtained and downloaded from the Gene Expression Omnibus (GEO) database. The differentially expressed genes (DEGs) in MDS were screened using GEO2R and overlapped DEGs were obtained with Venn Diagrams. Then, Gene Ontology and Kyoto Encyclopedia of Genes and Genomes pathway functional enrichment analyses, protein–protein interaction network establishment and survival analyses were performed. Functional enrichment analysis indicated that these DEGs were significantly enriched in the interferon signaling pathway, immune response, hematopoietic cell lineage and the FOXO signaling pathway. Four hub genes and four significant modules including 25 module genes were obtained via Cytoscape MCODE. Survival analysis showed that the overall survival of MDS patients having BLNK, IRF4, IFITM1, IFIT1, ISG20, IFI44L alterations were worse than that without alterations. In conclusion, the identification of these genes and pathways helps understand the underlying molecular mechanisms of MDS and provides candidate targets for the diagnosis and prognosis of MDS.

Prognostic significance of SF3B1 mutations in patients with myelodysplastic syndromes: A meta-analysis.

Splicing factor 3B subunit 1 (SF3B1) is a complex takes part in intron splicing of pre-mRNA and mutations within it have been reported frequently in myeloid malignancies including myelodysplastic syndromes (MDS). However, its prognostic value has been controversial. Hence, we aimed this meta-analysis to investigate the prognostic effect of SF3B1 mutations in patients with MDS. Several electronic databases were searched in of EMBASE, PubMed, the Cochrane Library and Web of Science (published up to November 2017) to obtain eligible studies. The pooled Hazard Ratio (HRs) and 95% confidence interval (CI) for overall survival (OS) and leukemia-free survival (LFS) as the primary and secondary endpoint, respectively, were chosen and extracted to determine the prognostic impact of SF3B1 mutations and to compare SF3B1 mutations to those with wild-type. Nine cohort studies with a total of 2259 patients were obtained, and the pooled HRs for OS was 0.93 (95% CI: 0.56-1.52, p-value = 0.78) and revealed no significant effect on overall survival of MDS patients by random effect models. Our meta-analysis suggested that SF3B1 has no impact on OS of patients with MDS, however, an adequately designed prospective study with a large number of patients with different type of SF3B1 mutations is needed to confirm these results. Additionally, Begg's and Egger's tests did not show any publication bias.

Predicting Response to Hypomethylating Agents in Patients with Myelodysplastic Syndromes (MDS) Using Artificial Intelligence (AI)

Introduction While the hypomethylating agents (HMAs) azacitidine (AZA) and decitabine (DAC) improve cytopenias and prolong survival in MDS patients (pts), response is not guaranteed. Timely identification of non-responders could prevent prolonged exposure to ineffective therapy, thereby reducing toxicities and costs. Currently no widely accepted clinical or genomic models exist to predict response or resistance to HMAs. We developed a clinical model to predict response or resistance to HMA after 90 days of initiating therapy based on changes in blood counts using time series analysis technology similar to the kind used in Apple's Siri or Google Assistant. In the setting of voice recognition, the sequence and context of words determines the meaning of a sentence; similarly, we hypothesized that the pattern of changes in MDS pts' blood counts would predict response or resistance early during treatment. Methods We screened a cohort of 107 pts with MDS (per 2016 WHO criteria) who received HMAs at our institution between February 2005 and July 2013 and had regular CBCs drawn during treatment. Mutations from a panel of 60 genes commonly mutated in myeloid malignancy were included. Responses were assessed after 6 months of therapy per International Working Group (IWG) 2006 criteria. Pts were divided randomly into training (80%) and validation (20%) cohorts. To address the potential for bias due to a small sample size, an oversampling algorithm was used to cluster similar pts based on their CBC data, Revised International Prognostic Scoring System (IPSS-R) score, and % bone marrow blasts at the time of diagnosis. CBC data from the first 90 days of treatment were fed into deep neural network (recurrent neural network) and decision tree algorithms, which were trained to predict whether pts would achieve a response (defined as complete remission (CR), partial remission (PR), or hematologic Improvement (HI)). Area under the curve (AUC) was used to assess model performance. Important features that impact the algorithm's predictions were extracted and plotted. Results 20747 unique data points were used, including CBC, clinical and genomic data. Among 107 pts, 61 (57.0%) received AZA only, 19 (17.8%) DAC only, 4 (3.7%) received both DAC and AZA, and 23 (21.5%) received HMA with an additional agent. Median age was 69 years (range: 37-100 years), and 27 (26.4%) were female. Forty pts (37.4%) were very low/low risk, 32 (29.9%) intermediate, 19 (17.8%) high, and 16 (14.9%) very high risk per IPSS-R. Responses included 23 (22.5%) CR, 2 (1.9%) marrow CR, 4 (3.9%) PR, and 20 (19.6%) HI. The most commonly mutated genes were ASXL1 (17.6%), TET2 (16.7%), SRSF2 (15.7%), SF3B1 (11.8%), RUNX1 (10.8%), STAG2(10.8%), and DNMT3A (10.8%). The median number of mutations per sample was 1 (range, 0-11), and 40 pts (39.2%) had &gt; 3 mutations per sample. When trained using absolute values and changes in CBC values, the model's AUC was 0.95 in the training cohort and 0.83 in the validation cohort. When the cohort was oversampled to 1000 pts, the validation cohort AUC increased to 0.89. Feature extraction algorithms identified increases in MCV and RDW during weeks 2-8 of treatment, increased proportion of lymphocytes, decreased proportion of monocytes, and increased platelet counts during weeks 6-8 as factors favoring response to HMA. The model provides personalized, patient-specific predictions that correlate with blood counts (Figure 1). Conclusions We describe a machine learning model that monitors changes in blood counts during therapy with HMA to predict response or resistance to HMA in MDS pts. Such a model can be used to develop novel trial designs wherein pts predicted to not respond after 90 days of HMA treatment could be assigned to an investigational agent. Conversely, it would help inform the decision to continue HMA therapy in pts predicted to respond. Increasing sample size with oversampling dramatically increased model accuracy; a larger cohort of pts treated at different institutions is currently under development. Disclosures Sekeres: Millenium: Membership on an entity's Board of Directors or advisory committees; Syros: Membership on an entity's Board of Directors or advisory committees; Celgene: Membership on an entity's Board of Directors or advisory committees. Mukherjee:Partnership for Health Analytic Research, LLC (PHAR, LLC): Consultancy; Takeda: Membership on an entity's Board of Directors or advisory committees; Celgene Corporation: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Projects in Knowledge: Honoraria; Novartis: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Pfizer: Honoraria; McGraw Hill Hematology Oncology Board Review: Other: Editor; Bristol-Myers Squibb: Speakers Bureau. Advani:Glycomimetics: Consultancy, Research Funding; Kite Pharmaceuticals: Consultancy; Amgen: Research Funding; Pfizer: Honoraria, Research Funding; Macrogenics: Research Funding; Abbvie: Research Funding. Maciejewski:Alexion: Consultancy; Novartis: Consultancy. Nazha:Novartis: Speakers Bureau; Tolero, Karyopharma: Honoraria; Abbvie: Consultancy; Jazz Pharmacutical: Research Funding; Incyte: Speakers Bureau; Daiichi Sankyo: Consultancy; MEI: Other: Data monitoring Committee.

Inhibition of LSD1 in Myelodysplastic Syndrome Progenitors Restores Differentiation of CD141Hi Conventional Dendritic Cells

Background: Immunotherapeutic approaches for myelodysplastic syndrome (MDS) show promise, but progress is limited by our incomplete understanding of the immunologic milieu. In a recent Phase I trial, we found that MDS patients with higher numbers of CD141Hi conventional dendritic cells (cDCs) were more likely to respond to NY-ESO-1 vaccination. In solid tumor models, the CD141Hi cDC is critical for initiating anti-tumor immune responses but its impact in myeloid malignancies is unknown. In studies of primary human specimens and mouse models, we tested the hypothesis that MDS patients exhibit decreased quantity and quality of CD141Hi cDCs due to impaired myeloid differentiation. Methods: Bone marrow (BM) cells were collected from MDS patients (pre-treatment) and age matched healthy donors (HD; defined as absence of hematologic malignancy). We quantified DC populations, stem, progenitor cells and interferon regulatory factor-8 (IRF-8) expression using flow cytometry and RT-qPCR. Histone modifications were assessed by chromatin immunoprecipitation. To assess DC differentiation of progenitors, human CD34+ and mouse c-kit+ cells were expanded and differentiated in vitro. Results: We found fewer CD141Hi cDCs (p&lt;0.0001), CD1c+ cDCs (p&lt;0.005) and plasmacytoid DCs (CD123+ pDCs; p&lt;0.005) overall in BM samples from MDS patients (n=71) compared to HD (n=17). We stratified MDS patients based on the relative number of DCs and found that only those patients with highest number of CD141Hi cDCs had superior survival (p &lt; 0.05). No differences in survival were seen in patients stratified by the CD1c+ cDCs (p = 0.96) and CD123+ pDCs (p = 0.32) populations. We hypothesized that decreased numbers of CD141Hi cDCs and adverse survival in MDS patients resulted from impaired differentiation of DC progenitors. We showed that MDS patients (n=19) have fewer monocyte-DC progenitors (MDP) and common DC progenitors (CDP) compared to HD (n=11; p&lt;0.01). We then hypothesized that MDS progenitors express lower levels of IRF8, a master regulator of CD141Hi cDC differentiation. IRF8 expression was significantly lower in CDPs from MDS patients compared to HD (p&lt;0.05). Furthermore, MDS patients with lower levels of IRF8 (n=8) in their MDPs showed a trend towards production of fewer CDPs and significantly fewer CD141Hi cDCs compared to those with higher levels of IRF8 (n=10; p&lt;0.005). These results suggest that approaches to increase IRF8 expression could enhance CD141Hi cDC differentiation. We hypothesized that inhibition of lysine-specific histone demethylase 1A (LSD1), which increases IRF8 expression in myeloid leukemia cells, would induce CD141Hi cDC differentiation. Pharmacologic inhibition of LSD1 increased IRF8 expression (both mRNA and protein) in KG-1 cells, a model of human CD34+ cells, and in HD and MDS CD34+ progenitors (p&lt;0.05). LSD1 inhibition in KG-1 cells resulted in increased H3K27 acetylation (3861-fold change) and H3K4 dimethylation (922.4-fold change) compared to PBS (p&lt;0.05) at a region -70 kb to the IRF8 transcriptional start site, a putative regulatory element that demonstrated the highest level of LSD1 binding. These data indicate that LSD1 inhibition alters histone modifications at the IRF8 locus, resulting in increased expression of IRF8. Pharmacologic inhibition of LSD1 in HD CD34+ cells increased the number of mature CD141Hi cDCs in 92% of specimens (n = 12; 3.4 fold-change) compared to PBS. Similarly, LSD1 inhibition in MDS CD34+ cells increased the number of CD141Hi cDCs (n = 12) in 75% of patient specimens (16.8 fold-change) compared to PBS. IRF8 function is conserved between mice and humans. To test whether the effect of LSD1 inhibition on cDC differentiation was dependent on IRF8, we compared the effect of LSD1 inhibition on BM c-kit+ cells from Irf8 knock-out mice (Irf8-KO) and littermate controls (WT). LSD1 inhibition in WT c-kit+ cells resulted in increased numbers of CD141Hi cDCs in vitro (p&lt;0.05). By contrast, LSD1 inhibition of Irf8-KO c-kit+ cells did not result in differentiation of CD141Hi cDCs. These data suggest that LSD1 inhibition drives CD141Hi cDCs differentiation through IRF8. Conclusion: These data reveal a previously unrecognized determinant of the immune microenvironment in MDS. The opportunity for epigenetic regulation of CD141Hi cDC differentiation in MDS offers an opportunity for intervention and a potential adjunct to immunotherapy for patients. Disclosures Sait: Celgene: Consultancy. Griffiths:Boston Scientific: Consultancy; Boston Scientific: Consultancy; Genentech, Inc.: Research Funding; Persimmune: Consultancy; Persimmune: Consultancy; Abbvie, Inc.: Consultancy; Genentech, Inc.: Research Funding; Novartis Inc.: Consultancy; Celgene, Inc: Consultancy, Research Funding; New Link Genetics: Consultancy; New Link Genetics: Consultancy; Astex Phramaceuticals/Otsuka Pharmaceuticals: Consultancy, Research Funding; Astex Phramaceuticals/Otsuka Pharmaceuticals: Consultancy, Research Funding; Novartis Inc.: Consultancy; Partner Therapeutics: Consultancy; Partner Therapeutics: Consultancy; Appelis Pharmaceuticals: Other: PI on a clinical trial; Onconova Therapeutics: Other: PI on a clinical trial; Appelis Pharmaceuticals: Other: PI on a clinical trial; Onconova Therapeutics: Other: PI on a clinical trial; Celgene, Inc: Consultancy, Research Funding; Abbvie, Inc.: Consultancy, PI on a clinical trial.