Pb2001: prognostic factors in myelodysplastic syndromes- a swedish retrospective real-world study

Topic: 10. Myelodysplastic syndromes - Clinical Background: Myelodysplastic Syndromes (MDS) is a heterogenous disease from a pathogenetic, clinical and prognostic viewpoint. Some patients with MDS may progress to Acute Myeloid Leukemia (AML). Erythropoiesis stimulating agents (ESAs) are recommended for treatment of anemia and potentially prevention of disease progression. Patients unresponsive to ESA therapy have limited treatment options, warranting further research. Aims: To identify important predictors of all-cause mortality and excess mortality in patients diagnosed with MDS [MDS cohort] and in a sub-population that received ESA treatment [ESA cohort]. Methods: This population based, retrospective cohort study in Sweden included patients ≥18 years of age with a record of a MDS diagnosis from January 1, 2006 to December 31, 2016 in the Swedish National Cancer Register. Data from the selected cohorts were linked to other national population registers. End of follow-up was December 31, 2016. Patients diagnosed with AML before MDS or at the same date as MDS (n=16) were excluded. Overall survival (OS) for different MDS subtypes was estimated using Kaplan-Meier curves and differences were tested with a log-rank test. A multiple Cox regression model, and an additive model employing an expectation maximization algorithm were implemented, including age, calendar period of diagnosis, sex and time-varying variable for AML status. These models estimated hazard ratios (HRs) for all-cause mortality, and excess mortality rate ratios (EMRRs) for cause-specific mortality, respectively. In the EMRR analyses, the mortality rates were compared with a matched cohort of the general population. Matching was based on age, sex, and calendar year. Results: 2,970 patients diagnosed with MDS 2006-2009 (39%) and 2010-2016 (61%), respectively, were included. Among these patients, 1,061 (36%) received ESA therapy, and of these, 696 (65%) discontinued ESA before December 31, 2015. At diagnosis, median (range) age was 76 (18-97) years and mean (standard deviation) age was 73.9 (+/-11.2) years. Overall, 16.7% (n=490) of patients progressed to AML. OS was significantly different between the MDS-subgroups (p<0.0001, Figure 1). Patients with refractory anemia with ring sideroblast type MDS had the best median OS (4.53 years; 95% confidence interval (CI) 4.26-5.60) while patients with refractory anemia with excess of blasts had the worst (1.18 years; 95% CI 1.05-1.29). Death occurred in 1,922 (65%) patients (due to AML: 14%). Patients undergoing allogeneic stem cell transplantation (SCT) had significantly lower, and patients with AML, significantly higher all-cause and excess mortality in both cohorts (Table 1). Younger age and female sex predicted a better outcome, while calendar period of diagnosis had no impact on all-cause mortality. Predictors of outcome were essentially the same in the ESA cohort. Summary/Conclusion: MDS subgrouping predicted OS. AML transformation carries a dismal prognosis. Older patients and male patients survived slightly shorter than young patients and female patients, respectively. Patients selected to undergo allogeneic SCT showed improved survival. This was also true for patients in the ESA cohort. This retrospective real-world analysis reinforces the need for new disease modifying therapies to improve outcome in MDS and especially to prevent progression to AML.Keywords: Myelodysplastic syndrome