Survival and quality of life in patients with lower risk myelodysplastic syndromes exposed to erythropoiesis-stimulating agents: an observational cohort study

Abstract

Background: In our previous study on erythropoiesis-stimulating agent (ESA) treatment in lower risk myelodysplastic syndromes from the European MDS (EUMDS) Registry, we showed that patients treated with ESAs had longer survival compared with patients who receive red blood cell transfusion (RBCT). In this study, with a longer follow up time and more patients included, we aimed to assess long-term effects on survival and health-related quality of life (HRQoL) of exposure to ESAs with or without RBCT in patients with lower risk myelodysplastic syndromes.

Methods: The EUMDS Registry is a non-interventional, longitudinal, real-world registry prospectively enrolling newly diagnosed patients older than 18 years with lower risk (International Prognostic Scoring System low or intermediate-1) myelodysplastic syndromes from 16 European countries and Israel. The analysis was restricted to patients with haemoglobin concentrations less than 100 g/L enrolled between Jan 1, 2008, and July 1, 2019, with last censoring of data on Dec 31, 2021. Patient management was recorded every 6 months, including treatment, transfusions, and HRQoL. ESA treatment followed local guidelines. The patients were separated into four groups at each study visit: no ESA or RBCT, ESA only, ESA plus RBCT, and RBCT only. The data were analysed longitudinally over time according to ESA and RBCT status during each 6-month interval, using propensity score matching. The main outcomes were median overall survival and leukaemia-free survival, and HRQoL. This study is registered with ClinicalTrials.gov, NCT00600860, as is ongoing.

Findings: 2448 patients (the ESA-unexposed group [n=1265] and ESA-exposed group [n=1183]) were diagnosed before July 1, 2019; 1520 (62·1%) were male and 928 (37·9%) were female. Median follow-up time was 3·9 years (IQR 1·6-6·5). After applying eligibility criteria and propensity matching, there were 426 patients in the ESA-unexposed group and 744 patients in the ESA-exposed group. Median overall survival in the ESA exposed group was 44·9 months (95% CI 40·2-50·5) compared with 34·8 months (28·6-39·2) in the ESA unexposed group; the absolute difference was 10·1 months (95% CI 2·2-18·0; hazard ratio [HR] 0·70 [95% CI 0·59-0·83]; p<0·0001). Patients without RBCT in the presence or absence of ESA exposure maintained significantly better HRQoL than those with RBCT, irrespective of ESA exposure (linear mixed effect model of EQ-5d-3L index score, RBCT coefficient -0·04 [95% CI -0·06 to 0·03], p<0·0001; linear mixed effect model of VAS, -4·57 [-6·02 to -3·13], p<0·0001).

Interpretation: ESA treatment in patients with lower risk myelodysplastic syndromes significantly improves overall survival when started before or early after the onset of regular transfusion therapy. Avoiding RBCT is associated with significantly better HRQoL.

Funding: H2020 European Research Council, Novartis Pharmacy B V Oncology Europe, Amgen, BMS/Celgene International, Janssen Pharmaceutica, Takeda Pharmaceuticals International, and Gilead Sciences.

Figure 1

Study profile ESA=erythropoiesis-stimulating agent. HRQoL=health-related quality of life. RBC=red blood cell. RBCT=red blood cell transfusion.

Figure 2

Overall survival and leukaemia-free survival by treatment status (A) Inverse probability of treatment weighted Kaplan–Meier curve for overall survival: median overall survival of patients in the ESA-exposed (time-varying covariate) group (44·9 months [40·2–50·5]) vs ESA-unexposed group (34·8 months [28·6–39·2]). (B) Overall survival of patients in the ESA-exposed group stratified by RBCT status before ESA exposure (p<0·0001). (C) Overall survival of the ESA-exposed group based on the number of pre-exposure transfusions given; no RBC units (n=396), 0–4 RBC units (n=146), and >4 RBC units (n=207). (D) Inverse probability of treatment weighted Kaplan–Meier curve for time to acute myeloid leukaemia transformation by ESA exposure (p=0·86). ESA=erythropoiesis-stimulating agent. RBC=red blood cell. RBCT=red blood cell transfusion.

Figure 3

Health-related quality of life by treatment status (A) Sankey diagram illustrating the flow of patients between the four groups. (B) Overall survival curves for patients included in the HRQoL analysis.(C) Mean HRQoL by ESA or RBCT status over time since reaching the eligibility criteria. RBCT and ESA status was defined at each visit. ESA=erythropoiesis-stimulating agent. HRQoL=health-related quality of life. RBCT=red blood cell transfusion. VAS=Visual Analog Scale.