The Relationship between Anemia, Transfusion Dependence, Quality of Life and Survival in Myelofibrosis

Introduction: Most patients with myelofibrosis (MF) are anemic within 1 year of diagnosis, and nearly all become dependent on red blood cell transfusions over time. Anemia and transfusion dependence (TD) carry a substantial humanistic burden and are prognostic factors routinely incorporated in clinicopathologic risk models for MF. However, the extent to which they affect health-related quality of life (HRQoL) and overall survival (OS) is not well characterized in published literature. To address this, we aimed to identify, explore, collate, and synthesize data on the potential inter-relationships that either anemia or TD has with either OS or HRQoL in patients with MF.

Methods: We conducted a targeted literature review within Embase, MEDLINE, MEDLINE In-Process and grey literature sources to identify studies relevant to the research objective. Additionally, we searched for eligible studies included in previously conducted systematic literature reviews of clinical trials and humanistic burden for MF. The retrieved publications were screened. Selected studies underwent quality assessment, data extraction and meta-analysis.

Results: The review identified 51 publications relating to 44 unique studies providing data on the relationship between anemia or TD and either OS or HRQoL. Of the 44 unique studies, there were 38 observational studies, 3 randomized controlled trials (RCTs), 1 single-arm trial, 1 pooled analysis of RCTs and 1 open-label extension of RCTs. 15 studies reported univariate Cox proportional regression results, and 24 studies reported relative effect estimates of multivariate Cox proportional regression.

There was substantial heterogeneity in study and patient characteristics across the studies. We used qualitive synthesis and presented hazard ratios (HRs) in forest plots to show univariate and multivariate relationships between anemia/TD and OS across the studies. 23 studies demonstrated a statistically significant association between shortened OS and either anemia or TD (univariate and multivariate HR ranges, for anemia: 1.20-3.28 and 1.27-37.15; for TD: 1.80-4.82 and 1.48-4.43). A similar trend was observed in 11 studies, demonstrating the negative prognostic value of anemia and TD in MF.

A feasibility assessment was conducted to explore whether the extracted results could be pooled in a meta-analysis. There was considerable heterogeneity in the definition of TD (14/22 studies provided no clear definition; only 2 studies used the same definition), while some differences in anemia definition across studies were noted (9 studies reported Hb<10g/dL as a defining cut-off level for anemia). Based on these findings, a quantitative analysis of the relationship between anemia and OS was conducted. Univariate and multivariate random effects model Paule-Mandel and DerSimonian-Laird estimators (anemia vs no anemia) were 1.73 (1.25-2.39; p=.005), 1.74 (1.29-2.34; p<.001), 2.56 (0.93-7.02; p=.062) and 1.80 (1.37-2.37; p<.001), respectively.

3 studies reported on anemia and HRQoL. Among them, one identified a significant correlation between anemia and increased symptom scores; another observed a numerical but not statistically significant difference in symptom reduction or HRQoL improvement between anemic and non-anemic groups; and the third study noted a greater improvement for those who responded to anemia treatment compared to non-responders.

Only 1 study reported on TD and HRQoL and found a significant association between requiring transfusions and greater symptom burden.

Conclusion: Evidence collated in this study demonstrates that anemia and TD have a detrimental association with survival in MF across variations in patient characteristics and clinical settings and should be considered as part of treatment decision-making. Limited available evidence indicates a trend between anemia, TD and poorer HRQoL and increased symptom burden in MF, however additional research is required to confirm and quantify this relationship.

Gerds:Agios: Consultancy; Disc Medicine: Consultancy; PharmaEssentia: Consultancy; Rain Oncology: Consultancy; GSK: Consultancy; AbbVie: Consultancy; BMS: Consultancy. Sen Nikitas:GSK plc: Current Employment, Current equity holder in publicly-traded company. Zhang:GSK: Current Employment, Current equity holder in publicly-traded company. Purser:GSK: Current Employment, Current equity holder in publicly-traded company. Conlon:GSK: Current Employment, Current equity holder in publicly-traded company. Pandey:GSK plc: Consultancy. Iheanacho:GSK plc: Consultancy. Dobi:GSK plc: Consultancy. Kapetanakis:GSK plc: Consultancy.