Validation of a prognostic model and the impact of mutations in patients with lower-risk myelodysplastic syndromes

Abstract

Purpose: A subset of patients with myelodysplastic syndromes (MDS) who are predicted to have lower-risk disease as defined by the International Prognostic Scoring System (IPSS) demonstrate more aggressive disease and shorter overall survival than expected. The identification of patients with greater-than-predicted prognostic risk could influence the selection of therapy and improve the care of patients with lower-risk MDS.

Patients and methods: We performed an independent validation of the MD Anderson Lower-Risk Prognostic Scoring System (LR-PSS) in a cohort of 288 patients with low- or intermediate-1 IPSS risk MDS and examined bone marrow samples from these patients for mutations in 22 genes, including SF3B1, SRSF2, U2AF1, and DNMT3A.

Results: The LR-PSS successfully stratified patients with lower-risk MDS into three risk categories with significant differences in overall survival (20% in category 1 with median of 5.19 years [95% CI, 3.01 to 10.34 years], 56% in category 2 with median of 2.65 years [95% CI, 2.18 to 3.30 years], and 25% in category 3 with median of 1.11 years [95% CI, 0.82 to 1.51 years]), thus validating this prognostic model. Mutations were identified in 71% of all samples, and mutations associated with a poor prognosis were enriched in the highest-risk LR-PSS category. Mutations of EZH2, RUNX1, TP53, and ASXL1 were associated with shorter overall survival independent of the LR-PSS. Only EZH2 mutations retained prognostic significance in a multivariable model that included LR-PSS and other mutations (hazard ratio, 2.90; 95% CI, 1.85 to 4.52).

Conclusion: Combining the LR-PSS and EZH2 mutation status identifies 29% of patients with lower-risk MDS with a worse-than-expected prognosis. These patients may benefit from earlier initiation of disease-modifying therapy.

Fig 1.

(A) Kaplan-Meier survival curves for 288 patients with low and intermediate-1 International Prognostic Scoring System risk. (B) Kaplan-Meier survival curves for the same patients assigned to categories 1 to 3 by the MD Anderson Lower-Risk Prognostic Scoring System. Overall survival was calculated from the time of sample collection to the time of death from any cause.

Fig 2.

Distribution of mutations in 204 of 288 samples from patients with lower-risk myelodysplastic syndromes with one or more mutations. Each column represents an individual sample. Colored cells indicate a mutation in the gene(s) described in that row on the left. Darker bars indicate two or more distinct mutations. Tyrosine kinase (TK) pathway genes include NRAS, KRAS, BRAF, CBL, and JAK2.

Fig 3.

Kaplan-Meier overall survival curves for patients with myelodysplastic syndromes in each Lower-Risk Prognostic Scoring System (LR-PSS) risk category stratified by EZH2 mutation status. (A) Category 1 patients; (B) category 2 patients; (C) category 3 patients.