
PS1503 ESTIMATING LIFE‐YEAR LOSS OF LOWER‐RISK MYELODYSPLASTIC SYNDROMES IN EUROPE
Author(s) -
Li Z.,
Tang D.,
Tang J.,
Morison J.,
Oliva E.N.
Publication year - 2019
Publication title -
hemasphere
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.677
H-Index - 11
ISSN - 2572-9241
DOI - 10.1097/01.hs9.0000564272.96038.ea
Subject(s) - medicine , international prognostic scoring system , survival analysis , myelodysplastic syndromes , population , cohort , hazard ratio , proportional hazards model , life expectancy , relative survival , demography , epidemiology , confidence interval , environmental health , cancer registry , bone marrow , sociology
Background: Lower‐risk (LR) myelodysplastic syndromes (MDS) are a group of hematologic disorders impacting patient survival and quality of life. Existing published survival data focus on describing duration of survival; however, it is difficult to interpret the seriousness of the disease objectively, since survival data have not been compared with data from the general population. Aims: To quantify the real‐world life‐year loss due to LR‐MDS in comparison with the general population in Germany, France, Italy, Spain, and the United Kingdom (EU5). Methods: A simulation model was developed in Microsoft Excel using parameters sourced from a recently published report of the European LeukaemiaNet MDS (EUMDS) registry. 1 The patient cohort comprised patients diagnosed with LR‐MDS from December 2007 to December 2010, during which time recombinant human erythropoietin was available as treatment in most EU countries. Baseline demographics predictive of survival, including age, sex, and revised International Prognostic Scoring System (IPSS‐R) risk score, were simulated using published patient characteristics. 1 Survival (in years) of patients with LR‐MDS was simulated using the exponential survival function. The exponential distribution parameter, modeled as a function of age, sex, and IPSS‐R risk score, was estimated by Cox modeling using published survival curves and hazard ratios. Patient characteristics and survival statistics from the simulation sample were verified against those published by de Swart et al. 1 Real‐world life‐year loss was defined as the difference between simulated survival time and the population‐level life expectancy specific to patient age, sex, and country. Results: A total of 5,000 patients were included in the simulation; 1,000 patients from each country. The simulated sample generated similar patient characteristics, median survival, and hazard ratios as found by de Swart et al. 1 As shown in Table 1, the mean life‐year loss was estimated at 6.3 years (median 6.8 years; 5 and 95 percentile [−13.9, 23.4]), a 40% relative survival loss for patients with LR‐MDS compared with the general population. Overall, 77% of patients were estimated to experience life‐year loss. Patients experiencing more pronounced life‐year losses than the general population were those <60 years (mean 13.4 years, 43% relative survival loss, 80% with life‐year loss) and those with Intermediate IPSS‐R risk score (mean 11.6 years, 76% relative survival loss, 96% with life‐year loss). The results were consistent among the EU5 countries and when other similar European LR‐MDS real‐world studies were used as a data source. Summary/Conclusion: LR‐MDS patients in 5 European countries experience significant real‐world life‐year loss when compared with the general population.