MDS subclassification-do we still have to count blasts?

Using Artificial Intelligence to Enhance Myelodysplastic Syndrome Diagnosis, Prognosis, and Treatment

Myelodysplastic syndromes represent a group of hematological neoplastic diseases caused by defective stem cells causing cytopenia and abnormal hematopoiesis. More than 30% of myelodysplastic syndrome cases develop into acute myeloid leukemia. An analysis of bone marrow samples, peripheral blood smears, multiparametric flow cytometry data, and clinical patient information is part of the current, time-consuming, and labor-intensive work up for myelodysplastic syndromes. Nowadays, clinical biomedical research has been transformed by the advent of artificial intelligence, specifically machine learning. Artificial intelligence (AI) can improve risk assessment and diagnosis, as well as boost the precision of clinical outcome prediction and illness classification. Algorithms based on artificial intelligence may be potentially helpful in discovering new needs for myelodysplastic syndrome-affected patients, choosing treatment and assessing minimal residual disease. In this review, we seek to identify the primary mechanisms and uses of artificial intelligence in myelodysplastic syndrome, pointing out its advantages and disadvantages while discussing the possible benefits of using AI pipelines in a therapeutic setting.

MicroRΝΑ analysis in patients with myelodysplastic neoplasms. Possible implications in risk stratification

MiRNAs have been identified as participants in leukemogenesis by controlling several cellular functions, such as differentiation, proliferation, and apoptosis. Their role in myelodysplastic neoplasms (MDS) pathogenesis is researched due to implementations in early identification, classification, and therapeutical options. IPSS-R, being the most widely used MDS classification, underestimates early biological events that can alter the disease's prognosis. The purpose of this study is to determine whether miRNA levels are aligned to MDS risk stratification groups and can therefore be used as diagnostic biomarkers. To evaluate miRNAs as possible biomarkers, we measured the levels of miR-181a-2-3p, miR-124-3p, miR-550a-3p, miR-155-5p, miR-151a-3p, and miR-125b-5p by a quantitative real-time PCR in bone marrow samples of 41 MDS patients. In conclusion, in myeloid malignancies, genomic characteristics may provide a wider apprehension of its clinical course and prognosis. MiRNAs constitute a possible diagnostic biomarker and therapeutic target, allowing intermediate-risk patients that express high levels of specific miRNAs to be re-classified and receive more advanced therapeutic agents. In our study, an association between high levels of miRNAs and worsening outcomes is established, supporting the need for further incorporation of molecular data into currently used classification systems.

Can molecular patterns help to classify overlapping entities in myeloid neoplasms?

Myeloid neoplasms include myeloproliferative and myelodysplastic neoplasms and acute myeloid leukaemia. Historically, these diseases have been diagnosed based on clinicopathological features with sometimes arbitrary thresholds that have persisted even as molecular features were gradually incorporated into their classification. As such, although current diagnostic approaches can classify the majority of myeloid neoplasms accurately using a combination of molecular and clinicopathological features, some areas of overlap persist and occasionally pose diagnostic challenges. These include overlap across BCR::ABL1-negative myeloproliferative neoplasms; between clonal cytopenia of undetermined significance and myelodysplastic neoplasms; myelodysplastic/myeloproliferative neoplasms; and, detection of KIT mutations in myeloid neoplasms other than mastocytosis, raising the prospect of systemic mastocytosis. Molecular testing has become state of the art in the diagnostic work-up of myeloid neoplasms, and molecular patterns can inherently help to classify overlapping entities if considered within a framework of haematological presentations. For future development, molecular testing will likely include whole genome and transcriptome sequencing, and primarily molecular classifications of myeloid neoplasms have already been suggested. As such, genetically defined groups should still constitute the basis for our understanding of disease development from early onset to progression, while clinicopathological features could then be used to describe the stage of the disease rather than the specific type of myeloid neoplasm.

Allogeneic hematopoietic cell transplantation in elderly patients with myelodysplastic syndromes: Considerations and challenges

Myelodysplastic syndromes/neoplasms (MDS) and related diseases are highly heterogeneous myeloid stem cell cancers that predominantly affect the elderly. The only curative treatment is allogeneic hematopoietic cell transplantation (HCT). Given the prevalence of age-related comorbidities, HCT in patients aged 65 years or older requires a highly personalized approach. This review summarizes disease risk stratification, treatment modalities, and outcomes for patients with MDS and related disorders, and discusses specific considerations and challenges for elderly patients undergoing HCT, including geriatric assessment, timing, conditioning treatment, donor and graft selection, and graft-versus-host disease prophylaxis.

Incorporation mutational profile might reduce the importance of blast count in prognostication of low-risk myelodysplastic syndromes

Addition of molecular data to prognostic models has improved risk stratification of myelodysplastic neoplasms (MDS). However, the role of molecular lesions, particularly in the group of low-risk disease (LR-MDS), is uncertain. We evaluated a set of 227 patients with LR-MDS. Overall survival (OS) and probability of leukaemic progression were the main endpoints. RUNX1 was associated with lower OS and SF3B1 with a reduced risk of death (HR: 1.7, 95% CI, 1.1-2.9; p = 0.05; and HR: 0.23, 95% CI 0.1-0.5; p < 0.001; respectively). TP53 and RUNX1 mutations were predictive covariates for the probability of leukaemic progression (p < 0.001). Blast percentage, neither analysed as categorical (<5% vs. 5%-9%; HR: 1.3, 95% CI, 0.7-2.9; p = 0.2) nor as a continuous variable (HR: 1.07, 95% CI, 0.9-1.1; p = 0.07), had impact on survival or probability of progression (sHR: 1.05, 95% CI, 0.9-1.1; p = 0.2). These results retained statistical significance when analysis was restricted to the definition of LR-MDS according to the WHO 2022 and ICC classifications (<5% blasts). Thus, with the incorporation of molecular data, blast percentage happens to lose clinical significance both for survival and probability of progression in the group of patients with LR-MDS.

Molecular taxonomy of myelodysplastic syndromes and its clinical implications

ABSTRACT

Myelodysplastic syndromes (MDS) are clonal hematologic disorders characterized by morphologic abnormalities of myeloid cells and peripheral cytopenias. Although genetic abnormalities underlie the pathogenesis of these disorders and their heterogeneity, current classifications of MDS rely predominantly on morphology. We performed genomic profiling of 3233 patients with MDS or related disorders to delineate molecular subtypes and define their clinical implications. Gene mutations, copy-number alterations, and copy-neutral loss of heterozygosity were derived from targeted sequencing of a 152-gene panel, with abnormalities identified in 91%, 43%, and 11% of patients, respectively. We characterized 16 molecular groups, encompassing 86% of patients, using information from 21 genes, 6 cytogenetic events, and loss of heterozygosity at the TP53 and TET2 loci. Two residual groups defined by negative findings (molecularly not otherwise specified, absence of recurrent drivers) comprised 14% of patients. The groups varied in size from 0.5% to 14% of patients and were associated with distinct clinical phenotypes and outcomes. The median bone marrow (BM) blast percentage across groups ranged from 1.5% to 10%, and the median overall survival ranged from 0.9 to 8.2 years. We validated 5 well-characterized entities, added further evidence to support 3 previously reported subsets, and described 8 novel groups. The prognostic influence of BM blasts depended on the genetic subtypes. Within genetic subgroups, therapy-related MDS and myelodysplastic/myeloproliferative neoplasms had comparable clinical and outcome profiles to primary MDS. In conclusion, genetically-derived subgroups of MDS are clinically relevant and might inform future classification schemas and translational therapeutic research.

SOHO State of the Art Updates and Next Questions: An Update on Higher Risk Myelodysplastic Syndromes

Higher-risk myelodysplastic syndromes (HR-MDS) are clonal myeloid neoplasms that cause life-limiting complications from severe cytopenias and leukemic transformation. Efforts to better classify, prognosticate, and assess therapeutic responses in HR-MDS have resulted in publication of new clinical tools in the last several years. Given limited current treatment options and suboptimal outcomes, HR-MDS stands to benefit from the study of investigational agents.Higher-risk myelodysplastic syndromes (HR-MDS) are a heterogenous group of clonal myeloid-lineage malignancies often characterized by high-risk genetic lesions, increased blood transfusion needs, constitutional symptoms, elevated risk of progression to acute myeloid leukemia (AML), and therapeutic need for allogeneic bone marrow transplantation. Use of blast percentage and other morphologic features to define myelodysplastic neoplasm subtypes is rapidly shifting to incorporate genetics, resulting in a subset of former HR-MDS patients now being considered as AML in presence of leukemia-defining genetic alterations. A proliferation of prognostic tools has further focused use of genetic features to drive decision making in clinical management. Recently, criteria to assess response of HR-MDS to therapy were revised to incorporate more clinically meaningful endpoints and better match AML response criteria. Basic science investigations have resulted in improved understanding of the relationship between MDS genetic lesions, bone marrow stromal changes, germline predispositions, and disease phenotype. However, therapeutic advances have been more limited. There has been import of the IDH1 inhibitor ivosidenib, initially approved for AML; the Bcl-2 inhibitor venetoclax and liposomal daunorubicin/cytarabine (CPX-351) are under active investigation as well. Unfortunately, effective treatment of TP53-mutated disease remains elusive, though preliminary evidence suggests improved outcomes with oral decitabine/cedazuridine over parenteral hypomethylating agent monotherapy. Investigational agents with novel mechanisms of action may help expand the repertoire of treatment options for HR-MDS and trials continue to offer a hopeful therapeutic avenue for suitable patients.

WHO/ICC Classification for Myelodysplastic Neoplasms/Syndromes Performs Better for Subtype Cytomorphological Diagnosis?

The International Consensus Classification of Myeloid Neoplasms and Acute Leukemias (ICC) and the 5th edition of the WHO classification (WHO 2022) have refined the diagnosis of myelodysplastic syndromes (MDS). Both classifications segregate MDS subtypes based on molecular or cytogenetic findings but rely on the subjective assessment of blast cell percentage and dysplasia in hematopoietic cell lineages. This study aimed to evaluate interobserver concordance among 13 cytomorphologists from eight hospitals in assessing blast percentages and dysplastic features in 44 MDS patients. The study found fair interobserver agreement for the PB blast percentage and moderate agreement for the BM blast percentage, with the best concordance in cases with <5% BM blasts and >10% BM blasts. Monocyte count agreement was fair, and dysplasia assessment showed moderate concordance for megakaryocytic lineage but lower concordance for erythroid and granulocytic lineages. Overall, interobserver concordance for MDS subtypes was moderate across all classifications, with slightly better results for WHO 2022. These findings highlight the ongoing need for morphological evaluation in MDS diagnosis despite advances in genetic and molecular techniques. The study supports the blast percentage ranges established by the ICC but suggests refining BM blast cutoffs. Given the moderate interobserver concordance, a unified classification approach for MDS is recommended.

Diagnosis and classification of myelodysplastic syndromes

ABSTRACT

Myelodysplastic syndromes (MDSs) are neoplastic myeloid proliferations characterized by ineffective hematopoiesis resulting in peripheral blood cytopenias. MDS is distinguished from nonneoplastic clonal myeloid proliferations by the presence of morphologic dysplasia and from acute myeloid leukemia by a blast threshold of 20%. The diagnosis of MDS can be challenging because of the myriad other causes of cytopenias: accurate diagnosis requires the integration of clinical features with bone marrow and peripheral blood morphology, immunophenotyping, and genetic testing. MDS has historically been subdivided into several subtypes by classification schemes, the most recent of which are the International Consensus Classification and World Health Organization Classification (fifth edition), both published in 2022. The aim of MDS classification is to identify entities with shared genetic underpinnings and molecular pathogenesis, and the specific subtype can inform clinical decision-making alongside prognostic risk categorization. The current MDS classification schemes incorporate morphologic features (bone marrow and blood blast percentage, degree of dysplasia, ring sideroblasts, bone marrow fibrosis, and bone marrow hypocellularity) and also recognize 3 entities defined by genetics: isolated del(5q) cytogenetic abnormality, SF3B1 mutation, and TP53 mutation. It is anticipated that with advancing understanding of the genetic basis of MDS pathogenesis, future MDS classification will be based increasingly on genetic classes. Nevertheless, morphologic features in MDS reflect the phenotypic expression of the underlying abnormal genetic pathways and will undoubtedly retain importance to inform prognosis and guide treatment.

Targeted NGS on sequential bone marrow biopsies aids in the evaluation of cytopenias and monocytosis and documents clonal evolution-a proof of principle study

Differential diagnosis of clonal versus reactive cytopenia and monocytosis, respectively, frequently presents a diagnostic challenge. With the two recent classifications of myeloid disorders, mutational analysis has gained importance as a diagnostic tool. However, reports on its utility on trephine bone marrow biopsies (BMB) are sparse. The aim of our proof of principle study was to determine the suitability of targeted sequencing for the longitudinal evaluation of cytopenia and monocytosis and demonstration of clonal evolution on sequential BMB. Seventy-seven EDTA-decalcified BMB of 33 patients with peripheral cytopenia and/or monocytosis, including at least one follow-up biopsy/patient, were included. Initial morphological diagnoses were idiopathic cytopenia of undetermined significance (ICUS, 8 cases), MDS (without blast increase, 7 cases), MDS with increased blasts/excess blasts (MDS-IB/EB) (11 cases), and CMML (7 cases). Thirty-one genes relevant for myeloid disorders were examined using two custom AmpliSeq NGS panels. Mutations were found in the initial BMB of 5/8 cases of ICUS, thus changing the diagnosis to clonal cytopenia of unknown significance (CCUS), 5/7 MDS, 10/11 MDS-IB/EB, and 7/7 CMML. Clonal evolution was observed in 14/33 (42%) cases, mostly associated with disease progression. None of the wild-type patients acquired mutations during follow-up. NGS-based mutation profiling is a robust diagnostic tool for BMB and provides valuable additional information, especially for cases with no/minimal dysplasia, and for better risk stratification of MDS. Tracking variant allele frequency and appearance of mutations over time allows for observing clonal evolution or relapse.