Welcome to the MDS 2023 Congress Calendar

Abstract Body

Despite a detailed understanding of the genes mutated in myelodysplastic syndromes (MDS), diagnostic and treatment decisions for patients with MDS rely primarily on clinical and cytogenetic variables as considered by the Revised International Prognostic Scoring System (IPSS-R). Here we describe the recently developed Molecular IPSS (IPSS-M), a clinico-genomic risk stratification system that considers clinical, cytogenetic and genetic parameters; the implementation of a web portal to facilitate its adoption, a strategy to handle missing variables, and the worldwide utilization of the web calculator as a clinical support tool.

Abstract Body

Myelodysplastic syndromes (MDS) are a heterogeneous group of chronic hematological malignancies characterized by dysplasia, ineffective hematopoiesis and a risk of progression to acute myeloid leukemia (AML). During the last decades, the great diversity in MDS characteristics has been shown to be supported by a variety of underlying cytogenetic abnormalities and an increasing number of mutations in genes involved in RNA-splicing (SF3B1, U2AF1, SRSF2, ZRSR2), chromatin modification (ASXL1, EZH2, BCOR, STAG2), DNA methylation (DNMT3A, TET2, IDH1/2), transcriptional regulation (RUNX1, GATA2), tumor suppressing (TP53) and signal transduction (CBL, JAK2, N/KRAS). Those alterations have been shown to be interconnected and dynamic, with some of them defining critical steps in disease progression and thus identified as potential prognostic markers and targets for new therapies.

This great diversity of alterations has made challenging the establishment of recommendations for appropriate MRD markers and/or time points in patients with MDS. Quantitative PCR (qPCR) and droplet digital PCR (ddPCR) remain the most sensitive, standardized, cost-effective, and time-efficient molecular technologies available in most clinical laboratories. They rely on the design and optimization of lesion-specific primers, limiting their use to frequent recurrent mutations without broad applicability in MDS. However, they are a technology of choice for specific markers, particularly in the context of targeted therapies to measure the eradication of a clone (e.g. IDH1/2 mutational hotspots). Besides gene mutations, monitoring of WT1 mRNA expression, which correlates with the blast population, could represent an alternative in patients with high risk disease treated with intensive therapies.

In this context, high-throughput sequencing (NGS) has emerged as an attractive tool, especially in patients with high-risk diseases receiving chemotherapy or undergoing stem cell transplantation. NGS allows the tracking of all variants at once and its repetition at multiple time points allows the appreciation of clonal hierarchy and dynamics and possibly the emergence of resistance and/or progression subclones with appropriate pipelines. However, its implementation in clinical practice still suffers from significant limitations. It requires standardization and important bioinformatic supports and remains expensive if used as a prospective tool. Sensitivity remains highly variable, depending on bioinformatic pipelines, sequencing technologies, as well as the type and location of variants. NGS achieves a sensibility of about 10^-2 with usual pipelines, although a threshold of 10^-4 is attainable with error-corrected sequencing using unique molecular identifiers (UMI). Additionally, many aspects have to be considered regarding result interpretation. Optimal targets and variant allele frequencies for decision making must be defined. Somatic mutations in hematological malignancy-free people, defining age-related clonal hematopoiesis, are found in at least 10% to 20% of people by age 60 years with current NGS technologies and may be identified in all individuals with more sensitive techniques. The most commonly mutated genes include DNMT3A and TET2 (by far the most common) and, to a latter extend, ASXL1, RNA-splicing genes, CBL or TP53. These mutations may accompany (distinct clone) or precede (part of the clone) the acquisition of MDS-driving mutations and may persist after tumor cells clearance. On the other hand, if the persistence of signaling mutations (RAS, CBL, NF1, FLT3) may be considered as poor prognostic signal when positive, their negativity should not lead to false reassurance, especially when the mutation is therapeutically targeted. Finally, germline predisposition to MDS and AML (DDX41, RUNX1, GATA2) may affect up to 10% of patients with high-risk diseases are, by definition, non-informative for MRD monitoring.

In conclusion, molecular MRD assessment is promising regarding new therapeutic strategies offered to patients with MDS. Considering the heterogeneity of molecular landscape, NGS holds great promise but recommendations and standardization are challenging. Furthermore, prospective trials are necessary to evaluate its efficacy in this regard.

Abstract Body

Current status of flow cytometry analysis of measurable residual disease in myelodysplastic neoplasms

AA van de Loosdrecht, LL Ngai, TM Westers, Cloos J.

Amsterdam University Medical Centers, location VU University MC, Cancer Center Amsterdam, The Netherlands

Accurate diagnosis in myelodysplastic neoplasms (MDS) necessitates additional sensitive and specific assays which can discriminate between MDS and non-MDS cytopenic patients, in those cases without molecular abnormalities, ring sideroblasts or other major morphologic dysplastic features in one or more lineages. The recent updated WHO classification (WHO2022) and ICC2022 on myeloid neoplasm contribute to a more refined classification of MDS and distinguish MDS from AML more accurately (Khoury JD et al. Leukemia 2022; Arber DA et al. Blood 2022). Due to the availability of huge data sets cytogenetic abnormalities as well as mutations are re-defined with diagnostic and prognostic implications that, in part, are translated into the risk models such as IPSS-Molecular (Bernard E et al. NeJMEvidence 2022). Since flow cytometry (FC) can identify aberrancies in myeloid progenitor cells and in maturing granulocytic, monocytic and erythroid lineages that are not recognised by cytology, FC may be instrumental in improving the diagnosis and classification of MDS (Van de Loosdrecht AA et al. Cytometry B Clin Cytom 2023; Wang W, Khoury JD Cytometry B Clin Cytom 2023). FC is even considered as co-criterion when regular diagnostic criteria are not met within the WHO/ICC2022 for CMML. In addition, FC aberrancies showed strong prognostic power beyond IPSS-Revised (Alhan C et al. Leukemia 2016; Oelschlaegel U et al. Cytometry B Clin Cytom 2023). Detecting measurable residual disease (MRD) in AML after second induction cycle of intensive chemotherapy has prognostic value and is currently implemented in the clinical decision for consolidation therapy (Jongen-Lavrencic M et al. N Engl J Med 2018; Lowenberg B et al. Blood Adv 2022). The ELN AML-MRD working party just updated the guidelines with novel experiences in the field of AML-MRD (Heuser M et al. Blood 2021). This is not yet established for MDS. However, increasing data support the role of MRD in MDS by tracking patient-specific mutations and using flow cytometry-definedleukemia-associated immunophenotypes (LAIP) where the LAIP is determined at diagnosis and followed during and after treatment (Ngai LL et al. [manuscript in prep]). In addition, a second method is based on the LAIP based different from normal (DfN) method where all LAIPs are evaluated at any measurement indicating possible clonal evolution. The detection of aberrant leukemic stem cells (IA-HSC) next to LAIP may further increase the accuracy of prognostication and hence clinical decision making in AML (Zeijlemaker W et al. Leukemia 2018; Ngai LL et al [submitted]). It is demonstrated that presence of IA-HSCs within MDS predicts leukemic progression indicating the clinical potential of IA-HSC as a prognostic biomarker (Van Spronsen M et al. Leukemia 2023 [epub ahead of print], Bachas C et al. Cytometry B Clin Cytom 2022). In addition, the presence of IA-HSCs correlated with perturbed hematopoiesis which is defined as disproportionally expanded CD34⁺ subsets. MRD assessments by the combination of molecular and FC-defined standardized platforms should be incorporated in emerging new clinical trials in MDS.

Background And Aims

Α shared pathobiology between MDS and CVD is postulated, but current evidence emanates only from retrospective cohorts suffering from crucial shortcomings. We conducted a prospective observational cohort study in lower-risk MDS patients.

Methods

Cerebrovascular, peripheral, and coronary affected vascular beds (AVB) were evaluated every 6 months by ultrasound and coronary artery calcium (CAC), respectively. Multi-Ethnic Study of Atherosclerosis (MESA), Framingham (FRS) risk scores and serum markers of myocardial injury (hsTnT), stress (NT-proBNP), and inflammation (hsCRP) were assessed. Twenty-nine patients are included so far. MDS subtypes were not correlated with CAC, MESA and FRS. In sharp contrast to the general population, FRS was not associated with CAC score (p=0.7, disproportionately increased in low and very low FRS risk patients) and the number of AVB (p=0.51, Figures 1A,B).

Results

In 13/29 patients the annual evaluation revealed a significant increase in CAC score from baseline (average 70.3%, p=0.003) which was markedly higher compared to the age-matched general population and was not associated with baseline MESA score, FRS, preexisting CVD, and sex. Baseline NT-proBNP levels correlated strongly with CAC score (p=0.001, Figure 1C) consistent with numerous evidence supporting an association of NT-proBNP with increased CVD risk.

Conclusions

To our knowledge this is the first study in LR-MDS patients assessing longitudinally and objectively all critical parameters of subclinical atherosclerosis in all vascular beds. Our preliminary results indicate that the established CVD prediction models are dysfunctional in MDS and along with the pronounced annual increase in CAC score argue further for an intrinsic tendency of MDS towards CVD development.

Background And Aims

The IPSS-M score improves outcome prediction at diagnosis in patients with myelodysplastic syndromes (MDS) but is not validated for assessment during follow-up. Here we investigated whether transfusion patterns over time refine prognostic prediction

Methods

We gathered a population-based cohort of 677 Swedish adult patients with MDS. IPSS-M features, transfusion and treatment data were studied for associations with overall survival (OS) and AML-transformation (AML-t). The impact of early changes in transfusion patterns was assessed by a Markov multistate model.

Results

A total of 26,489 units of red blood cells were administrated during follow-up. At diagnosis, 37% of the patients were erythrocyte transfusion-dependent (E-TD) (Figure 1). TET2-mutation, “good” cytogenetics, and SF3B1^alpha were associated with freedom from E-transfusions (E-TF) at diagnosis while E-TD correlated with TP53-multihit (TP53^mh) and poorer cytogenetics (P<0.01). In transfusion-naïve patients at diagnosis, SF3B1^alpha and higher hemoglobin predicted a longer time to first E-TD, while poorer cytogenetics, TP53^mh, and higher bone marrow blast were associated with a shorter time to first E-TD. Transfusion status at 8 and 12 months after diagnosis significantly predicted outcomes across IPSS-M groups as well as within the specific subsets SF3B1^alpha and TP53^mh. In a multivariable model, E-TD at the landmark 8 months was associated with worse outcomes, independently of IPSS-M (Figure 2). Finally, transfusion patterns prior to landmark impacted OS, AML-t and future transfusion requirement.

Conclusions

Prior transfusion patterns and transfusion state at landmark predict outcome in MDS. It is a simple but robust parameter that can refine outcomes prediction in a dynamic setting in addition to IPSS-M.