Abstract Background and Aims Anaemia is a common complication of chronic kidney disease (CKD), affecting up to 53.4% of patients with Stage 5 CKD compared to 8.4% of patients with Stage 1 CKD [1]. Anaemia of CKD is associated with increased risk of cardiovascular events and death, increased healthcare resource utilisation (HRU), and reduced health-related quality of life (HRQoL) [2]. Established treatments for anaemia of CKD include erythropoiesis-stimulating agents (ESAs) [1,2]; however, hypoxia-inducible factor prolyl hydroxylase inhibitors (HIF-PHIs), targeting both erythropoietin production and iron metabolism (the two key mechanisms of anaemia of CKD), are an emerging class of agents in this indication [3]. The cost and HRQoL burden associated with anaemia of CKD treatment is not well characterised.1 Thus, we conducted a systematic literature review (SLR) to evaluate published economic models (including the cost inputs, time horizon, subgroups, and clinical assumptions used) and HRU data in patients with anaemia of CKD. Findings of the SLR will help guide future economic model development and inform economic evaluation strategies as part of health technology assessment submissions to support daprodustat, an HIF-PHI under development for use in both dialysis-dependent and non-dialysis-dependent patients. Method Relevant publications were identified using structured searching of MEDLINE, Embase, MEDLINE In-Process, Cochrane Controlled Trials Register, and the Cochrane Database of Systematic Reviews from database inception to 10 April 2022 using disease- (e.g. CKD, chronic kidney failure, anaemia) and economic-associated (e.g. economics, economic evaluation, quality-adjusted life-years, model structure) search terms. Additional supplemental searches (conference proceedings [2016–2021], grey literature, and bibliographies) were conducted. Studies were assessed for inclusion or exclusion by two independent reviewers, with any discrepancies resolved by a third reviewer. Objectives were to determine the structure of any published economic models and the availability of health utility data in patients with anaemia and additional complications of CKD, and to examine what cost inputs, time horizons, subgroups, and clinical assumptions are used to inform economic modelling. Results From 1397 citations identified in the searches, 40 primary studies were included in the final analysis. Review of the data revealed an established approach for modelling the ESA cost-effectiveness (i.e. a Markov model with health states defined according to CKD treatment received or anaemia status): most studies were conducted from a North American or European perspective, time horizons varied, and reporting of cycle length was poor. Only one economic evaluation included an HIF-PHI model: given the different mechanisms of action of ESAs and newer treatments, a modelling strategy focused mainly on a Markov approach may not be the most optimal way of assessing cost-effectiveness in anaemia of CKD where HIF-PHIs are used. Considerations of modelling factors that could influence cost-effectiveness of HIF-PHIs are shown in Fig. 1. While existing models typically only stratified patient populations according to age (<65 vs >65 years), for appropriate modelling of treatment effectiveness, the impact of prior treatments (including response to prior ESAs) was also considered. As in every model, comorbidities may also have a role in effect modification and this impact will need to be considered in economic models of anaemia of CKD treatment. Due to the natural history of anaemia in CKD, innovative modelling techniques that retain the Markov approach and which account for respective haemoglobin levels, CKD stage, and potential kidney transplantation should be incorporated into the economic evaluation (Fig. 2). Conclusion Adjustment of existing models using the approaches described may provide more reliable estimates of treatment efficacy in a highly heterogenous population of patients with anaemia of CKD.
Read full abstract