Study On The Relationship Between Biological Characteristics And Prognosis Of Myelodysplastic Syndromes

Background and purposeMyelodysplastic syndrome(MDS)is a heterogeneous hematopoietic disease characterized by bone marrow failure,morbid hematopoiesis,and a predisposition to acute myeloid leukemia(AML).Due to its high heterogeneity,patients’ survival can vary from months to years.Identification of individualized prognostic risk factors is important for guiding optimal treatment to improve outcomes.Although the revised International Prognostic Scoring System(IPSS-R),which includes the proportion of blast cells,the severity of cytopenia and cytogenetics as parameters,is widely used in clinical practice,there are still some patients whose clinical outcomes do not match the risk assessment.This suggests that we need to explore some new biological characteristics related to the prognosis of MDS patients to improve the prognositic scoring system.Therefore,this study focused on three aspects:gene mutation,immune dysregulation and myelofibrosis,and explored their relationship with the prognosis of MDS patients.Methods1.In the first part,this retrospective study examined all hospitalized MDS patients from October 2012 to June 2019 at our institution.Patients were enrolled in this study if they were:(1)aged≥14 years;(2)the results of gene sequencing at diagnosis were available;(3)complete parameters for IPSS-R scoring.Patients were excluded if they were diagnosed with AML(MDS transformation)at the first bone marrow aspiration.Cytogenetic analysis was performed by conventional chromosome banding technique and fluorescence in situ hybridization.The detection method of gene sequencing is next generation sequencing.By analyzing the mutation landscape,the relationship between gene mutation and chromosome karyotype and the impact of gene mutation on treatment evaluation,leukemia transformation and overall survival(OS),we explored the significance of gene mutation in the prognosis of MDS patients.Patients were divided into normal and aberrant cytogenetic groups for subgroup analysis to find out whether there were differences in the impact of the gene mutation between the two groups.2.In the second part,CD34+cells were separated and their RNA was extracted from bone marrow of MDS patients at diagnosis and those of healthy donors.These bone marrow specimens were frozen in the specimen bank of our hospital from April 2017 to November 2020 before the study began.In total,45 samples met the sequencing requirements,including 42 MDS patients and 3 healthy donors.Transcriptome sequencing was performed in these samples by the next generation sequencing.DESeq2 was used to analyze the differential expression of mRNA between MDS patients and healthy control group,relative-low risk and relative-high risk MDS patients respectively.Gene Ontology(GO),Kyoto Encyclopedia of Genes(KEGG)and Disease ontology(DO)were used for functional enrichment analysis.By Cox and LASSO-Cox regression,immune-related genes associated with prognosis were selected and a linear prognostic model was established.Time-dependent receiver operating characteristic(ROC)curves were made.To evaluate the predictive performance of the new model and IPSS-R model,area under the curve(AUC)and consistency index(C-index)were calculated using survival ROC software packages.3.In the third part,this retrospective study examined all hospitalized MDS patients from October 2012 to June 2019 at our institution.Patients were eligible if they were as follows:(1)ag≥14 years;(2)bone marrow biopsy specimens at diagnosis could be used to assess fibrosis;(3)complete parameters for IPSS-R scoring.Patients were excluded if they were diagnosed with primary myelofibrosis(PMF)or overlap syndromes.Based on the degree of MF,we divided the patients into 2 groups:grade 0-1(MF-0/1)and grade 2-3(MF-2/3)groups.To explore the significance of MF in prognosis of MDS patients,we compared the clinical characteristics,treatment evaluation and OS between the two groups.4.Statistical analysis:Chi-square test and Mann-Whitney U test were used for categorical variables and continuous variables,respectively.OS and disease-free survival(DFS)were estimated by Kaplan-Meier(KM)method and compared by logrank test.Cox proportional risk model was used for risk factor analysis of timeevent variables.The cumulative incidence of relapse,non-relapse mortality(NRM),engraftment and graft versus host disease(GVHD)was calculated and compared using a Fine-gray competitive risk model.P values were two-sided and considered significant if less than 0.05.SPSS 21.0(SPSS,Inc,Chicago,Illinois)and R version 4.0.3(R Development Core Team,Vienna,Austria)were used for data analysis.Results1.Impact of gene mutation on the progosis in MDS.1.1 In total,304 patients were enrolled in this study.Two hundred and eightyone patients(92.4%)harbored at least one mutation.Genes with mutation frequency≥10%were TET2(27%),TP53(17%),TET1(17%),ASXL1(17%),EZH2(16%),FAT1(14%),EP300(13%),DDX18(12%),CD101(11%),SF3B1(10%),RUNX1(10%),U2AF1(10%).1.2 There were 164 cases in normal karyotype group and 140 cases in aberrant karyotype group.The number of mutations in the normal was greater than that in aberrant groups,with the medians of 4(0-16)and 3(0-10),respectively(P=0.029).The correlation analysis showed that ASXL1,CD101,KDM6A,SH2B3,and IL-3RA mutations were more common in the normal group,while TET2 and TP53 were more common in the aberrant group(all P<0.05).And positive correlation was found between complex karyotypes and TP53 mutations(P<0.05).1.3 Compared to wild-types,the overall response rate(ORR)was lower in patients with TP53 mutations,while it was higher with EP300 mutations(P=0.003,P=0.004,respectively).Allo-HSCT can improve OS of patients with TP53 mutation,no matter in normal or aberrant karyotype group(all P<0.05).However,allo-HSCT may not improve OS of patients with DNMT3A,FAT1 or IL-7R mutations in normal karyotype group(all P>0.05).1.4 In multivariate Cox analysis,IPSS-R and DDX18 mutation were risk factors for leukemia transformation in the whole cohort,while TET2 mutation in the normal karyotype group and RUNX1,U2AF1 and DDX18 mutations in the aberrant karyotype group(all P<0.05).1.5 In multivariate Cox analysis,age,IPSS-R,TP53 and DNMT3A were risk factors while allo-HSCT was a protective factor for OS in the whole cohort(all P<0.05).In the normal karyotype group,the results showed that age,IPSS-R,TP53,DNMT3A,FAT1 and IL-7R were risk factors while allo-HSCT was a protective factor for OS(all P<0.05).In multivariate Cox analysis,age,IPSS-R and TP53 were risk factors while allo-HSCT was a protective factor for OS in the aberrant karyotype group(all P<0.05).2.Functional enrichment of differentially expressed genes and establishment of a prognostic model based on immune-related genes.2.1 There were 249 differentially expressed genes between MDS patients and healthy control group,including 162 up-regulated genes and 87 down-regulated genes.GO gene enrichment analysis of differentially expressed genes showed that the top 5 clusters in biological process classification were positive regulation of leukocycle activation,positive regulation of cell activation,leukocycle mediatied immunity,positive regulation of lymphocyte activation and adaptive immune response based on somatic recombination of immune receptors built from immunoglobulin superfamily domains respectively.There were 39 differentially expressed immune-related genes in MDS patients in the relative-low risk group and the relative-high risk group,including 32 up-regulated genes and 7 down-regulated genes.GO gene enrichment analysis of differentially expressed immune-related genes showed that the top 5 clusters in biological process classification were cytokine-mediated signaling pathway,response to lipopolysaccharide,response to molecule of bacterial origin,positive regulation of cytokine production and T cell activation respectively.2.2 Univariate Cox analysis showed that the mRNA expression levels of 42 immune-related genes were correlated with poor prognosis,and 12 genes were correlated with good prognosis.Eight genes with the greatest influence on prognosis were screened out by LASSO-Cox regression.A new prognostic model was established based on these 8 immune-related genes.Risk score=ADIPOR2×1.9+MAPK1×1.7+MAPK14×1.9+INHA×1.9+ILlβ×l.6+PTH2R×2.1+IL17D×1.5+HLA-DQB1×0.4.The AUC values of 2-year OS in the new model and IPSS-R model were 0.903 and 0.717 respectively,and C-index values were 0.817 and 0.763,respectively.3.Impact of MF on the progosis in MDS.3.1 A total of 316 patients with MDS were enrolled in this study.There were 43 cases in MF-2/3 group and 273 cases in MF-0/1 group.Complex karyotypes were more common in the MF-2/3 group when compared with MF-0/1 group(P=0.002).3.2 A total of 102 patients received cytoreduction,including 90 cases in MF-0/1 group and 12 in MF-2/3 group.The overall response rate(ORR)of cytoreduction was 49.0%,along with 53.3%in the MF-0/1 and 16.7%in MF-2/3 groups(P=0.017).3.3 In total,141 patients underwent allo-HSCT,including 121 in the MF-0/1 and 20 in MF-2/3 groups.Among patients with allo-HSCT,the 2-year overall survival(OS)was 68.5%(95%CI:60.1-76.9%)and 68.4%(95%CI:47.4-89.4%)in the MF0/1 and MF-2/3 groups,respectively,(P=0.636).Among patients without allo-HSCT,the 2-year OS was 49.9%(95%CI:40.7-59.1%)and 19.2%(95%CI:0-39.6%)in the MF-0/1 and MF-2/3 groups,respectively,(P=0.002).3.4 In the whole cohort,multivariate Cox analysis suggested that age(HR=1.02,P=0.003),IPSS-R(HR=1.83,P<0.0001)and MF-2/3(HR=1.75,P=0.039)were adverse factors for OS in MDS patients.Conclusions1.In cytoruduction,the ORR was lower in patients with TP53 mutations,while it was higher with EP300 mutations when compared to wild-types.Allo-HSCT can improve the OS of patients with TP53 mutation,but patients with DNMT3A,FAT1,or IL-7R in normal karyotype group may not benefit from allo-HSCT.This suggests that cytoruduction may be attempted for patients with EP300 mutation,but alloHSCT is recommended for patients with TP53 mutation as soon as possible.2.The mutations associated with rapid leukemia transformation were DDX18 mutations in the whole cohort,TET2 and DDX18 mutations in the normal karyotype group,and DDX18,RUNX1,and U2AF1 mutations in the aberrant karyotype group.3.The mutations associated with poor OS were TP53 and DNMT3A mutations in the whole cohort,TP53,DNMT3A,FAT1 and IL-7R mutations in the normal karyotype group,and TP53 mutations in the aberrant karyotype group.4.Genes differentially expressed between MDS patients and healthy control group mainly focus on immune-related pathways.MDS patients with different risk stratification have different immune status.The expression of immune-related genes in mRNA level is closely related to the prognosis of MDS patients.The prognostic model based on immune-related genes performed well and can be used as a supplement to the IPSS-R.5.MF-2/3 is a poor prognostic factor for MDS.And the adverse impact of MF on prognosis may be overcome by allo-HSCT.