Modelled cost-effectiveness of high cut-off haemodialysis compared to standard haemodialysis in the management of myeloma kidney

Abstract

Background:Approximately 10–20% of multiple myeloma patients experience dialysis-dependent renal failure. This is principally due to myeloma kidney, a tubulointerstitial injury caused by high circulating concentrations of monoclonal free light chains. Studies have found that between 3% and 37% of patients with myeloma kidney requiring dialysis recover renal function. In-vivo studies indicate that extended haemodialysis using high cut-off dialysers (HCO-HD) can remove significant quantities of free light chains and is associated with a renal recovery rate of 63–74% in these patients.Objective:The objective of this study was to assess the cost-effectiveness of HCO-HD compared to standard HD in the management of myeloma kidney.Methods:The study used a lifetime Excel-based decision tree model that followed all patients from treatment of the initial presentation with myeloma kidney requiring dialysis to death. It was populated with published clinical data, United Kingdom costs and expert opinion, using a National Health Service perspective and 3.5% annual discounting.Results:HCO-HD was dominant to standard HD, meaning it was both more effective (greater life years and quality adjusted life years) and less costly, due to a greater increase in the proportion of patients recovering renal function. The model projected lifetime costs of £31,345 per patient for patients treated with standard haemodialysis only and £24,845 for the new treatment (discounted). The model predicted an average survival of 19.92 months for patients on standard HD and 33.90 months for the new therapy (discounted).Conclusions:The analysis found that treatment of myeloma kidney using an extended schedule of HCO-HD may substantially improve renal recovery in multiple myeloma patients compared to standard HD, resulting in greater life expectancy and cost savings due to avoided chronic dialysis. Limitations of the study include those common to rare diseases including small study sizes and limited natural history data.