Predictors of Erythropoiesis-Stimulating Agent Failure in Lower-Risk Myelodysplastic Syndromes: Systematic Literature Review

Introduction: Patients with lower-risk myelodysplastic syndromes (LR-MDS) commonly experience chronic progressive anemia, which is initially treated with erythropoiesis-stimulating agents (ESAs). However, many patients do not respond to ESAs, thereby putting them at risk of earlier disease progression, which can require invasive and/or costly treatments, including blood transfusions or hypomethylating agents. A systematic literature review (SLR) was conducted to assess quantitative associations between prognostic factors and response to treatment with ESAs in patients with LR-MDS.

Methods: An electronic search for relevant literature was conducted on February 25, 2022, across Embase and MEDLINE via OvidSP. Also, proceedings from the American Society of Hematology, European Hematology Association, American Society of Clinical Oncology, Academy of Managed Care Pharmacy, and European Society for Medical Oncology meetings from 2019 to 2022 were searched via OvidSP or manually if not indexed in databases. Articles written in languages other than English were excluded. Adult patients with LR-MDS were included in the analysis. Studies reporting measures of effect between prognostic factors and ESA response were included, as were studies reporting quantitative data pertaining to differences in patient characteristics between ESA responders and non-responders. Studies reporting both univariate and multivariate analyses were included. No geographic or temporal limits were applied to the search strategy, except for congress proceedings, which were limited to 2019 onwards.

Results: In total, 38 studies were eligible for inclusion in this SLR, and 8 prognostic factors of interest were most frequently reported: baseline serum erythropoietin (EPO), hemoglobin (Hb), transfusion independence, bone marrow blast count, ferritin levels, International Prognostic Scoring System (IPSS) risk status, age, and karyotype status. Regarding study design, most were retrospective cohort studies (N = 15) or single-arm trials (N = 12), and were conducted in Europe (N = 24). Within the identified publications, ESA response was often assessed using the International Working Group (IWG) MDS 2000 or 2006 criteria in publications after 2000. Studies published prior to 2000 did not use a standardized method of refining response, but still classified patients according to changes in Hb levels or transfusion requirements.

The literature available in LR-MDS demonstrated statistically significant relationships suggesting that patients with low serum EPO levels (vs higher EPO levels), high Hb levels (vs lower levels), and transfusion independence (vs transfusion dependent) are more likely to respond to ESA treatment (Table). This relationship was consistent across both univariate and multivariate analyses. The evidence for bone marrow blasts and ferritin levels was inconclusive, but could potentially be influenced in the future by the publication of studies with larger patient populations. There was no published evidence that ESA response was related to IPSS risk status, age, and karyotype status. The associated data for these prognostic factors in responders versus non-responders showed similar patterns.

Conclusion: All studies assessing quantitative associations between ESA response and prognostic factors were published after 2000, hence ESA response was assessed according to IWG 2000 or 2006 criteria. The studies demonstrate a clear link between response to ESA treatment and lower serum EPO levels, higher Hb levels, and transfusion independence. The other prognostic factors were either not significant or inconclusive, and more research is needed to clarify whether they have a prognostic impact on ESA response in patients with LR-MDS. Some of the studies comparing prognostic factors for responders versus non-responders to ESA treatment were published before 2000 and did not use standardized methods to describe response. Due to changes in ESA treatment response definitions over time, some patients in the included studies may be classified differently with a more recent version of the IWG MDS criteria.

Boccia:Rigel: Consultancy, Honoraria, Research Funding, Speakers Bureau; AbbVie: Consultancy, Honoraria, Speakers Bureau; Genmab: Consultancy, Honoraria, Speakers Bureau; Sanofi: Speakers Bureau; Regeneron: Speakers Bureau; AstraZeneca: Consultancy, Honoraria, Research Funding, Speakers Bureau; Janssen: Consultancy, Honoraria, Research Funding, Speakers Bureau; Bristol Myers Squibb: Honoraria, Speakers Bureau; Incyte: Consultancy, Honoraria, Research Funding, Speakers Bureau; Karyopharm: Speakers Bureau; Genentech: Consultancy, Honoraria, Research Funding, Speakers Bureau; Daiichi Sankyo: Consultancy, Honoraria, Research Funding, Speakers Bureau; Amgen: Consultancy, Honoraria, Research Funding, Speakers Bureau. Deshpande:BMS: Consultancy. von Wilamowitz-Moellendorff:Evidera: Current Employment. Raorane:Evidera: Current Employment. Klijn:OPEN Health: Ended employment in the past 24 months; BMS: Current Employment. Xiao:Bristol Myers Squibb: Current Employment, Current equity holder in private company. Yucel:Bristol Myers Squibb: Current Employment, Current holder of stock options in a privately-held company.