In a study reported in the Journal of Clinical Oncology, Bersanelli et al reported that researchers in the EuroMDS Project found that patients with myelodysplastic syndromes (MDS) could be classified into eight distinct subtypes based on genomic characteristics. They also developed a new prognostic model incorporating clinical and genomic variables that permits individually tailored survival predictions.
The study included data from an international retrospective cohort (EuroMDS cohort) of 2,043 patients with primary MDS according to 2016 World Health Organization (WHO) criteria. Mutations in 47 genes and cytogenetic abnormalities were analyzed to identify genetic associations and genomic subgroups. Random-effects Cox proportional hazards modeling was used to develop prognostic models. The prognostic model was tested in an independent validation cohort of 318 patients from the Humanitas Research Hospital in Milan (Humanitas cohort).
A total of eight distinct genomic subgroups were identified that did not correlate with morphologic categories defined by current WHO classification and were associated with different clinical phenotypes and outcomes. The genomic groups were characterized by different survival probabilities (overall P < .0001).
Genomic landscape in MDS reveals distinct subgroups associated with specific clinical features and discrete patterns of evolution, providing a proof-of-concept for next-generation disease classification and prognosis…The integration of clinical data with diagnostic genome profiling in MDS may provide prognostic predictions that are personally tailored to individual patients. Such information will empower the clinician and support complex decision-making process in these patients.— Bersanelli et al
Tweet this quote
In five groups, dominant genomic features included splicing mutations in SF3B1, SRSF2, and U2AF1 that occur early in disease history, are associated with specific clinical phenotypes, and drive disease evolution. These groups are associated with differing prognosis, with groups with SF3B1 mutations having better overall survival (eg, vs one group defined by SRSF2 mutations with coexisting mutations in other genes).
Specific comutation patterns accounted for clinical heterogeneity within SF3B1- and SRSF2-related disease. For example, one group defined by isolated SF3B1 mutation and ring sideroblasts was characterized by isolated anemia, normal or high platelet count, single or multilineage dysplasia, and low percentage of marrow blasts. Another group consisting of patients with SF3B1 mutation with coexisting mutations (ASXL1 and RUNX1) was characterized by anemia associated with mild neutropenia and thrombocytopenia, multilineage dysplasia, and higher marrow blast percentage vs the former group.
MDS with complex karyotype and/or TP53 gene abnormalities and MDS with acute leukemia–like mutations were associated with the poorest survival outcomes.
MDS cases with 5q deletion were clustered into two distinct groups based on number of mutated genes and/or presence of TP53 mutations. The group with no or single mutations exhibited better survival vs the group with two or more mutations or TP53 mutations.
A novel prognostic model integrating 63 clinical and genomic variables was developed that generated personally tailored predictions of survival. Internal cross-validation in the EuroMDS cohort showed model concordance of predicted vs observed survival of 0.74 in a training set (67% of patients) and 0.71 in a test set (33% of patients). Concordance in the entire cohort was 0.74. Model performance was better than that observed using conventional scoring with the age-adjusted International Prognostic Scoring System-Revised (0.62 and 0.65 in EuroMDS training and test sets). Model concordance in the Humanitas validation cohort was 0.75.
The investigators have produced a Web portal that can be used to generate outcome predictions based on user-defined constellations of genomic and clinical features.
The investigators concluded, “Genomic landscape in MDS reveals distinct subgroups associated with specific clinical features and discrete patterns of evolution, providing a proof-of-concept for next-generation disease classification and prognosis…The integration of clinical data with diagnostic genome profiling in MDS may provide prognostic predictions that are personally tailored to individual patients. Such information will empower the clinician and support complex decision-making process in these patients.”
Matteo G. Della Porta, MD, Comprehensive Cancer Center, Humanitas Clinical and Research Center-IRCCS and Humanitas University, Milan, is the corresponding author for the Journal of Clinical Oncology article.
Disclosire: The study was supported by the European Union, Associazione Italiana per la Ricerca contro il Cancro Foundation, Italian Ministry of Health, and others. For full disclosures of the study authors, visit ascopubs.org.The content in this post has not been reviewed by the American Society of Clinical Oncology, Inc. (ASCO®) and does not necessarily reflect the ideas and opinions of ASCO®.