Multicolour and lineage-specific interphase chromosome Flow-FISH: method development and clinical validation.

Flow cytometry can be applied in the detection of fluorescence in situ hybridisation (FISH) signals to efficiently analyse chromosomal aberrations. However, such interphase chromosome (IC) Flow-FISH protocols are currently limited to detecting a single colour. Furthermore, combining IC Flow-FISH with conventional multicolour flow cytometry is difficult because the DNA-denaturation step in FISH assay also disrupts cellular integrity and protein structures, precluding subsequent antigen-antibody binding and hindering concurrent labeling of surface antigens and FISH signals. We developed a working protocol for concurrent multicolour flow cytometry detection of nuclear IC FISH signals and cell surface markers. The protocol was validated by assaying sex chromosome content of blood cells, which was indicative of chimerism status in patients who had received sex-mismatched allogeneic haematopoietic stem cell transplants (allo-HSCT). The method was also adapted to detect trisomy 12 in chronic lymphocytic leukaemia (CLL) subjects. We first demonstrated the feasibility of this protocol in detecting multiple colours and concurrent nuclear and surface signals with high agreement. In clinical validation experiments, chimerism status was identified in clinical samples (n=56) using the optimised IC Flow-FISH method; the results tightly corresponded to those of conventional slide-based FISH (R2=0.9649 for XX cells and 0.9786 for XY cells). In samples from patients who received sex-mismatched allo-HSCT, individual chimeric statuses in different lineages could be clearly distinguished with high flexibility in gating strategies. Furthermore, in CLL samples with trisomy 12, this method could demonstrate that enriched trisomy 12 FISH signal was present in B cells rather than in T cells. Finally, by performing combined labelling of chromosome 12, X chromosome, and surface markers, we could detect rare residual recipient CLL cells with trisomy 12 after allo-HSCT. This adaptable protocol for multicolour and lineage-specific IC Flow-FISH advances the technique to allow for its potential application in various clinical contexts where conventional FISH assays are currently being utilised.

Stellae-123 gene expression signature improved risk stratification in taiwanese acute myeloid leukemia patients.

The European Leukemia Net recommendations provide valuable guidance in treatment decisions of patients with acute myeloid leukemia (AML). However, the genetic complexity and heterogeneity of AML are not fully covered, notwithstanding that gene expression analysis is crucial in the risk stratification of AML. The Stellae-123 score, an AI-based model that captures gene expression patterns, has demonstrated robust survival predictions in AML patients across four western-population cohorts. This study aims to evaluate the applicability of Stellae-123 in a Taiwanese cohort. The Stellae-123 model was applied to 304 de novo AML patients diagnosed and treated at the National Taiwan University Hospital. We find that the pretrained (BeatAML-based) model achieved c-indexes of 0.631 and 0.632 for the prediction of overall survival (OS) and relapse-free survival (RFS), respectively. Model retraining within our cohort further improve the cross-validated c-indexes to 0.667 and 0.667 for OS and RFS prediction, respectively. Multivariable analysis identify both pretrained and retrained models as independent prognostic biomarkers. We further show that incorporating age, Stellae-123, and ELN classification remarkably improves risk stratification, revealing c-indices of 0.73 and 0.728 for OS and RFS, respectively. In summary, the Stellae-123 gene expression signature is a valuable prognostic tool for AML patients and model retraining can improve the accuracy and applicability of the model in different populations.

Comparison of the 2022 world health organization classification and international consensus classification in myelodysplastic syndromes/neoplasms.

In 2022, two novel classification systems for myelodysplastic syndromes/neoplasms (MDS) have been proposed: the International Consensus Classification (ICC) and the 2022 World Health Organization (WHO-2022) classification. These two contemporary systems exhibit numerous shared features but also diverge significantly in terminology and the definition of new entities. Thus, we retrospectively validated the ICC and WHO-2022 classification and found that both systems promoted efficient segregation of this heterogeneous disease. After examining the distinction between the two systems, we showed that a peripheral blood blast percentage ≥ 5% indicates adverse survival. Identifying MDS/acute myeloid leukemia with MDS-related gene mutations or cytogenetic abnormalities helps differentiate survival outcomes. In MDS, not otherwise specified patients, those diagnosed with hypoplastic MDS and single lineage dysplasia displayed a trend of superior survival compared to other low-risk MDS patients. Furthermore, the impact of bone marrow fibrosis on survival was less pronounced within the ICC framework. Allogeneic transplantation appears to improve outcomes for patients diagnosed with MDS with excess blasts in the ICC. Therefore, we proposed an integrated system that may lead to the accurate diagnosis and advancement of future research for MDS. Prospective studies are warranted to validate this refined classification.

Kinome expression profiling improves risk stratification and therapeutic targeting in myelodysplastic syndromes.

ABSTRACT: The human kinome, which comprises >500 kinases, plays a critical role in regulating numerous essential cellular functions. Although the dysregulation of kinases has been observed in various human cancers, the characterization and clinical implications of kinase expressions in myelodysplastic syndromes (MDS) have not been systematically investigated. In this study, we evaluated the kinome expression profiles of 341 adult patients with primary MDS and identified 7 kinases (PTK7, KIT, MAST4, NTRK1, PAK6, CAMK1D, and PRKCZ) whose expression levels were highly predictive of compromised patient survival. We then constructed the kinase stratification score (KISS) by combining the weighted expressions of the 7 kinases and validated its prognostic significance in 2 external MDS cohorts. A higher KISS was associated with older age, higher peripheral blood and marrow blast percentages, higher Revised International Prognostic Scoring System (IPSS-R) risks, complex karyotype, and mutations in several adverse-risk genes in MDS, such as ASXL1, EZH2, NPM1, RUNX1, STAG2, and TP53. Multivariate analysis confirmed that a higher KISS was an independent unfavorable risk factor in MDS. Mechanistically, the KISS-high patients were enriched for gene sets associated with hematopoietic and leukemic stem cell signatures. By investigating the Genomics of Drug Sensitivity in Cancer database, we identified axitinib and taselisib as candidate compounds that could potentially target the KISS-high myeloblasts. Altogether, our findings suggest that KISS holds the potential to improve the current prognostic scheme of MDS and inform novel therapeutic opportunities.

Validation of the Artificial Intelligence Prognostic Scoring System for Myelodysplastic Syndromes in chronic myelomonocytic leukaemia: A novel approach for improved risk stratification.

Chronic myelomonocytic leukaemia (CMML) is a rare haematological disorder characterized by monocytosis and dysplastic changes in myeloid cell lineages. Accurate risk stratification is essential for guiding treatment decisions and assessing prognosis. This study aimed to validate the Artificial Intelligence Prognostic Scoring System for Myelodysplastic Syndromes (AIPSS-MDS) in CMML and to assess its performance compared with traditional scores using data from a Spanish registry (n = 1343) and a Taiwanese hospital (n = 75). In the Spanish cohort, the AIPSS-MDS accurately predicted overall survival (OS) and leukaemia-free survival (LFS), outperforming the Revised-IPSS score. Similarly, in the Taiwanese cohort, the AIPSS-MDS demonstrated accurate predictions for OS and LFS, showing superiority over the IPSS score and performing better than the CPSS and molecular CPSS scores in differentiating patient outcomes. The consistent performance of the AIPSS-MDS across both cohorts highlights its generalizability. Its adoption as a valuable tool for personalized treatment decision-making in CMML enables clinicians to identify high-risk patients who may benefit from different therapeutic interventions. Future studies should explore the integration of genetic information into the AIPSS-MDS to further refine risk stratification in CMML and improve patient outcomes.

Dysregulated immune and metabolic pathways are associated with poor survival in adult acute myeloid leukemia with CEBPA bZIP in-frame mutations.

Acute myeloid leukemia (AML) with CEBPA bZIP in-frame mutations (CEBPAbZIP-inf) is classified within the favorable-risk group by the 2022 European LeukemiaNet (ELN-2022). However, heterogeneous clinical outcomes are still observed in these patients. In this study, we aimed to investigate the mutation profiles and transcriptomic patterns associated with poor outcomes in patients with CEBPAbZIP-inf. One hundred and thirteen CEBPAbZIP-inf patients were identified in a cohort of 887 AML patients homogeneously treated with intensive chemotherapy. Concurrent WT1 or DNMT3A mutations significantly predicted worse survival in AML patients with CEBPAbZIP-inf. RNA-sequencing analysis revealed an enrichment of interferon (IFN) signaling and metabolic pathways in those with a shorter event-free survival (EFS). CEBPAbZIP-inf patients with a shorter EFS had higher expression of IFN-stimulated genes (IRF2, IRF5, OAS2, and IFI35). Genes in mitochondrial complexes I (NDUFA12 and NDUFB6) and V (ATP5PB and ATP5IF1) were overexpressed and were associated with poorer survival, and the results were independently validated in the TARGET AML cohort. In conclusion, concurrent WT1 or DNMT3A mutations and a dysregulated immune and metabolic state were correlated with poor survival in patients with CEBPAbZIP-inf, and upfront allogeneic transplantation may be indicated for better long-term disease control.

Validation of the molecular international prognostic scoring system in patients with myelodysplastic syndromes defined by international consensus classification.

Myelodysplastic syndromes (MDS) have varied prognoses and require a risk-adapted treatment strategy for treatment optimization. Recently, a molecular prognostic model (Molecular International Prognostic Scoring System [IPSS-M]) that combines clinical parameters, cytogenetic abnormalities, and mutation topography was proposed. This study validated the IPSS-M in 649 patients with primary MDS (based on the 2022 International Consensus Classification [ICC]) and compared its prognostic power to those of the IPSS and revised IPSS (IPSS-R). Overall, 42.5% of the patients were reclassified and 29.3% were up-staged from the IPSS-R. After the reclassification, 16.9% of the patients may receive different treatment strategies. The IPSS-M had greater discriminative potential than the IPSS-R and IPSS. Patients with high, or very high-risk IPSS-M might benefit from allogeneic hematopoietic stem cell transplantation. IPSS-M, age, ferritin level, and the 2022 ICC categorization predicted outcomes independently. After analyzing demographic and genetic features, complementary genetic analyses, including KMT2A-PTD, were suggested for accurate IPSS-M categorization of patients with ASXL1, TET2, STAG2, RUNX1, SF3B1, SRSF2, DNMT3A, U2AF1, and BCOR mutations and those classified as MDS, not otherwise specified with single lineage dysplasia/multi-lineage dysplasia based on the 2022 ICC. This study confirmed that the IPSS-M can better risk-stratified MDS patients for optimized therapeutic decision-making.

Japanese subgroup analysis in the Asian phase II study of darinaparsin in patients with relapsed or refractory peripheral T-cell lymphoma.

A Japanese subgroup analysis from the Asian phase II study of darinaparsin in patients with relapsed or refractory peripheral T-cell lymphoma (PTCL) was performed to evaluate the efficacy and safety outcomes of the Japanese population. In this Asian phase II study, darinaparsin was administered to 65 patients, including 37 Japanese patients. In the Japanese population, the histopathological type of PTCL was PTCL, not otherwise specified in 26 patients (70.3%), angioimmunoblastic T-cell lymphoma in 9 patients (24.3%) and anaplastic large cell lymphoma, anaplastic lymphoma kinase (ALK) -negative in 2 patients (5.4%), and the median patient age was 70.0 (range: 43-85). 94.6% and 35.1% of the Japanese population had previously received multi-agent and single-agent regimen, respectively. The efficacy and safety were summarized and compared between the overall and Japanese populations. Based on central assessment, the overall response rate was 22.2% (8/36; 90% confidence interval [CI]: 11.6-36.5) in the Japanese population and 19.3% (11/57; 90% CI: 11.2-29.9) in the overall population. There were no essential differences in the safety profile of darinaparsin between the Japanese population and the overall population. The results of this subgroup analysis indicate that the efficacy and safety profiles of the Japanese subpopulation were broadly consistent with that of the overall population, and that darinaparsin is potentially an effective treatment with a manageable safety profile in Japanese patients with relapse or refractory PTCL.

Evaluation of the clinical significance of global mRNA alternative splicing in patients with acute myeloid leukemia.

Aberrant alternative splicing (AS) is involved in leukemogenesis. This study explored the clinical impact of alterations in global AS patterns in 341 patients with acute myeloid leukemia (AML) newly diagnosed at the National Taiwan University Hospital and validated it using The Cancer Genome Atlas (TCGA) cohort. While studying normal cord blood CD34+ /CD38- cells, we found that AML cells exhibited significantly different global splicing patterns. AML with mutated TP53 had a particularly high degree of genome-wide aberrations in the splicing patterns. Aberrance in the global splicing pattern was an independent unfavorable prognostic factor affecting the overall survival of patients with AML receiving standard intensive chemotherapy. The integration of global splicing patterns into the 2022 European LeukemiaNet risk classification could stratify AML patients into four groups with distinct prognoses in both our experimental and TCGA cohorts. We further identified four genes with AS alterations that harbored prognostic significance in both of these cohorts. Moreover, these survival-associated AS events are involved in several important cellular processes that might be associated with poor response to intensive chemotherapy. In summary, our study demonstrated the clinical and biological implications of differential global splicing patterns in AML patients. Further studies with larger prospective cohorts are required to confirm these findings.

Validation of the prognostic significance of the 2022 European LeukemiaNet risk stratification system in intensive chemotherapy treated aged 18 to 65 years patients with de novo acute myeloid leukemia.

The European LeukemiaNet (ELN) recently proposed a revised recommendation for the diagnosis and management of acute myeloid leukemia (AML) in adults, recognized as ELN-2022. However, validation in a large real-world cohort remains lacking. In this study, we aimed to validate the prognostic relevance of the ELN-2022 in a cohort of 809 de novo, non-M3, younger (ages 18-65 years) AML patients receiving standard chemotherapy. The risk categories of 106 (13.1%) patients were reclassified from that determined using ELN-2017 to that determined using ELN-2022. The ELN-2022 effectively helped distinguish patients as favorable, intermediate, and adverse risk groups in terms of remission rates and survival. Among patients who achieved first complete remission (CR1), allogeneic transplantation was beneficial for those in the intermediate risk group, but not for those in the favorable or adverse risk groups. We further refined the ELN-2022 system by re-categorizing AML patients with t(8;21)(q22;q22.1)/RUNX1::RUNX1T1 with KIThigh , JAK2 or FLT3-ITDhigh mutations into the intermediate risk subset, AML patients with t(7;11)(p15;p15)/NUP98::HOXA9 and AML patients with co-mutated DNMT3A and FLT3-ITD into the adverse risk subsets, and AML patients with complex or monosomal karyotypes, inv (3)(q21.3q26.2) or t(3;3)(q21.3;q26.2)/GATA2,MECOM(EVI1) or TP53 mutation into the very adverse risk subset. The refined ELN-2022 system performed effectively to distinguish patients as favorable, intermediate, adverse, and very adverse risk groups. In conclusion, the ELN-2022 helped distinguish younger, intensively treated patients into three groups with distinct outcomes; the proposed refinement of ELN-2022 may further improve risk stratification among AML patients. Prospective validation of the new predictive model is necessary.

A three-gene leukaemic stem cell signature score is robustly prognostic in chronic myelomonocytic leukaemia.

Leukaemic stem cell (LSC) gene expression has recently been linked to prognosis in patients with acute myeloid leukaemia (17-gene LSC score, LSC-17) and myelodysplastic syndromes. Although chronic myelomonocytic leukaemia (CMML) is regarded as a stem cell disorder, the clinical and biological impact of LSCs on CMML patients remains elusive. Making use of multiple independent validation cohorts, we here describe a concise three-gene expression signature (LSC-3, derived from the LSC-17 score) as an independent and robust prognostic factor for leukaemia-free and overall survival in CMML. We propose that LSC-3 could be used to supplement existing risk stratification systems, to improve prognostic performance and guide management decisions.

Poor prognostic implications of myelodysplasia-related mutations in both older and younger patients with de novo AML.

A set of myelodysplasia-related (MDS-R) gene mutations are incorporated into the 2022 European LeukemiaNet risk classification as adverse genetic factors for acute myeloid leukemia (AML) based on their poor prognostic impact on older patients. The impact of these mutations on younger patients (age < 60 years) remains elusive. In the study of 1213 patients with de novo non-M3 AML, we identified MDS-R mutations in 32.7% of the total cohort, 44.9% of older patients and 23.4% of younger patients. The patients with MDS-R mutations had a significantly lower complete remission rate in both younger and older age groups. With a median follow-up of 9.2 years, the MDS-R group experienced shorter overall survival (P = 0.034 for older and 0.035 for younger patients) and event-free survival (P = 0.004 for older and 0.042 for younger patients). Furthermore, patients with MDS-R mutations more frequently harbored measurable residual disease that was detectable using next generation sequencing at morphological CR than those without MDS-R mutations. Allogeneic hematopoietic stem cell transplantation (allo-HSCT) might ameliorate the negative impact of MDS-R mutations. In summary, AML patients with MDS-R mutations have significantly poorer outcomes regardless of age. More intensive treatment, such as allo-HSCT and/or novel therapies, is warranted for AML patients with MDS-R mutations.

Long-term outcomes of combined intravitreal methotrexate and systemic high-dose methotrexate therapy in vitreoretinal lymphoma.

OBJECTIVE: The optimal treatment for vitreoretinal lymphoma (VRL) remains a challenge, as central nervous system (CNS) relapse occurs frequently, leading to the worst impact on survival. We previously proposed combined intravitreal methotrexate and systemic high-dose methotrexate therapy for this disease. This study aimed to report the long-term outcomes of patients with VRL using this combination treatment. METHODS: We conducted a retrospective cohort study on patients with VRL at a tertiary referral center between 2003 and 2018. RESULTS: Thirty-two patients were included, of whom 23 had primary VRL (PVRL) and nine had concurrent intraocular and CNS diseases. The treatment was well tolerated. Twenty-six (81.3%) patients achieved complete response (CR). After a median follow-up time of 103.5 months, the 5-year survival rate was 73.3%, whereas the 5-year progression-free survival (PFS) rate was 29.9%. Twenty-four (75%) patients relapsed, including 12 with isolated intraocular relapses at first relapse and a total of 17 with CNS/systemic relapses. The development of CNS/systemic relapse negatively affected survival, but intraocular relapse did not. The median CNS/systemic PFS was 69.5 months, but the risk of CNS/systemic relapse increased steadily with a cumulative incidence rate at 2, 5, and 10 years being 22.6%, 44.2%, and 65%, respectively. Multivariate analysis identified concurrent CNS disease at diagnosis as the only poor-risk factor for CNS/systemic relapse. CONCLUSIONS: This study confirms good efficacy and acceptable toxicities of the combination approach. However, incorporation of further intensive consolidation strategies into the treatment protocol to effectively prevent subsequent CNS/systemic relapse deserves to be considered.

Clinico-genetic and prognostic analyses of 716 patients with primary myelodysplastic syndrome and myelodysplastic syndrome/acute myeloid leukemia based on the 2022 International Consensus Classification.

The 2022 International Consensus Classification (ICC) recategorized myeloid neoplasms based on recent advances in the understanding of the biology of hematologic malignancies, in which myelodysplastic syndrome (MDS) with blasts of 10%-19% is classified as MDS/acute myeloid leukemia (AML), MDS with mutated SF3B1, irrespective of the number of ring sideroblasts, as MDS-SF3B1, and those with multi-hit TP53 mutations as MDS with mutated TP53. In the analysis of 716 patients with MDS diagnosed according to the 2016 WHO classification, we found that 75.3% of patients remained in the MDS group based on the ICC, while 24.7% of patients were reclassified to the MDS/AML group after the exclusion of 15 patients who were classified to the AML group. Patients with MDS/AML showed a distinct mutational landscape and had poorer outcomes, compared to those with MDS. In the MDS group, patients with MDS-SF3B1 had higher frequencies of DNMT3A and TET2 mutations than those with MDS, not otherwise specified, with single lineage dysplasia or multilineage dysplasia. Patients with mutated TP53 were associated with dismal outcomes, irrespective of the blast percentage. In conclusion, this study showed that the ICC facilitates efficient segregation and risk-stratification of MDS which can help guide the treatment choice of patients with the disease.

Darinaparsin in patients with relapsed or refractory peripheral T-cell lymphoma: results of an Asian phase 2 study.

Darinaparsin is a novel organic arsenical compound of dimethylated arsenic conjugated to glutathione, with antitumor activity and a mechanism of action markedly different from other available agents. This phase 2, nonrandomized, single-arm, open-label study evaluated the efficacy and safety of intravenous darinaparsin (300 mg/m2 over 1 hour, once daily for 5 consecutive days, per 21-day cycle) and its pharmacokinetics at multiple doses in 65 Asian patients with relapsed or refractory peripheral T-cell lymphoma (PTCL). The primary end point was the overall response rate (ORR). The ORR based on central assessment was 19.3% (90% confidence interval, 11.2-29.9), which was significantly higher than the predefined threshold of 10% (P = .024). The ORR was 16.2% in patients with PTCL-not otherwise specified and 29.4% in patients with angioimmunoblastic T-cell lymphoma. Tumor size decreased in 62.3% of patients. Treatment-emergent adverse events (TEAEs) were observed in 98.5% of patients. Grade ≥3 TEAEs with an incidence rate of ≥5% included anemia (15.4%), thrombocytopenia (13.8%), neutropenia (12.3%), leukopenia (9.2%), lymphopenia (9.2%), and hypertension (6.2%). Darinaparsin is effective and well tolerated, with TEAEs that were clinically acceptable and manageable with symptomatic treatment and dose reductions. This trial was registered at www.clinicaltrials.gov as #NCT02653976.

Cerebrospinal fluid soluble programmed death-ligand 1 is a useful prognostic biomarker in primary central nervous system lymphoma.

The increased expression of programmed death-ligands 1 and 2 (PD-L1 and PD-L2, respectively) on tumour cells contributes to immune evasion, suggesting that these proteins are attractive therapeutic targets. This study aimed to evaluate the validity of cerebrospinal fluid (CSF) soluble PD-L1 (sPD-L1) and soluble PD-L2 (sPD-L2) as biomarkers for primary central nervous system lymphoma (PCNSL). We determined the CSF concentrations of sPD-L1 and sPD-L2 in 46 patients with PCNSL using enzyme-linked immunosorbent assays (ELISAs). A control group comprised 153 patients with other brain tumours, inflammatory/infectious status, or neurodegenerative diseases. Only CSF sPD-L1 levels were significantly higher in patients with PCNSL relative to the controls. CSF sPD-L1 also exhibited superior overall discrimination performance compared to CSF sPD-L2 in diagnosing PCNSL. Compared with patients with PCNSL with low CSF sPD-L1 levels, more patients with high levels had high serum lactate dehydrogenase levels, leptomeningeal involvement, and deep-brain involvement. Furthermore, CSF sPD-L1 could predict poor survival in PCNSL but CSF sPD-L2 could not. Intriguingly, CSF sPD-L1 levels were correlated with disease status and their dynamic changes post treatment could predict time to relapse. In conclusion, this study identified CSF sPD-L1 as a promising prognostic biomarker, indicating a therapeutic potential of PD-L1 blockade in PCNSL.

Distinct genetic landscapes and their clinical implications in younger and older patients with myelodysplastic syndromes.

Myelodysplastic syndromes (MDS) are a group of clinically and genetically diverse diseases that impose patients with an increased risk of leukemic transformation. While MDS is a disease of the elderly, the interplay between aging and molecular profiles is not fully understood, especially in the Asian population. Thus, we compared the genetic landscape between younger and older patients in a cohort of 698 patients with primary MDS to advance our understanding of the distinct pathogenesis and different survival impacts of gene mutations in MDS according to age. We found that the average mutation number was higher in the older patients than younger ones. The younger patients had more WT1 and CBL mutations, but less mutated ASXL1, DNMT3A, TET2, SF3B1, SRSF2, STAG2, and TP53 than the older patients. In multivariable survival analysis, RUNX1 mutations with higher variant allele frequency (VAF) and U2AF1 and TP53 mutations were independent poor prognostic indicators in the younger patients, whereas DNMT3A and IDH2 mutations with higher VAF and TP53 mutations conferred inferior outcomes in the older patients. In conclusion, we demonstrated the distinct genetic landscape between younger and older patients with MDS and suggested that mutations impact survival in an age-depended manner.

The clinical and biological characterization of acute myeloid leukemia patients with S100A4 overexpression.

BACKGROUND/PURPOSE: The S100 family proteins are involved in a variety of important biological processes, most notably immune and inflammatory responses. Their dysregulation also plays a role in the pathogenesis of human cancers. S100A4, also known as metastasin, has long been regarded as a biological marker in tumor progression and metastasis in multiple solid cancers, but its clinical significance in acute myeloid leukemia (AML) has not been extensively studied. METHODS: We retrospectively studied the association between S100A4 gene expression and the clinical characteristics, mutational and transcriptomic profiles of 227 AML patients treated with standard intensive chemotherapy. Genetic mutations of myeloid disease associated genes were analyzed by Sanger sequencing. Microarray-based transcriptomic gene expression profiling was performed on archived bone marrow mononuclear cells. Bioinformatic analyses, including differential gene expression and gene set enrichment analysis, were conducted to delineate the underlying pathogenic mechanisms. RESULTS: Higher S100A4 expression was associated with older age, monocytic differentiation of leukemic cells, and adverse clinical outcome. S100A4 high-expressors had inferior overall survival and disease-free survival; this finding could be validated in the TCGA AML cohort (both the microarray and RNA-seq platforms). Multivariate Cox regression analysis supported S100A4 as an independent prognostic factor. Bioinformatic analysis showed that AML with higher S100A4 expression was enriched for the interferon, NLRP3 inflammasome, and epithelial-mesenchymal transition pathways. CONCLUSION: This study provides evidence that S100A4 overexpression serves as a poor prognostic biomarker in AML, holds potential to guide treatment planning in the clinic, and indicates novel therapeutic directions.