OBJECTIVES

Ravulizumab, a treatment for paroxysmal nocturnal hemoglobinuria (PNH) administered every 8 weeks (q8w), demonstrates clinical effectiveness and a sustained safety profile. Previously, the only approved therapy for PNH was eculizumab, administered every 2 weeks (q2w). We evaluated public preferences in Australia, Canada, the Netherlands, Sweden and the UK for PNH treatment attributes and estimated disutilities for economic modelling (without requiring EQ-5D data).

METHODS

A stated preference discrete choice experiment survey evaluated PNH treatment attributes, including overall survival, administration, risk of hemolysis and meningitis and the need for blood transfusions. Attributes and defined levels were combined using a published orthogonal array into choice sets. The survey was administered to public participants aged ≥18 years, and attribute characteristics were explained (e.g. need for meningococcal vaccination, risk of hemolysis and transfusion needs). The mixed logit model estimated preference strength and disutilities for each attribute. Marginal substitution rates were estimated between overall survival and other attributes to estimate disutilities, weighted against average life expectancy. Data were analyzed following logical consistency checks, reflecting respective country demographics.

RESULTS

Choice data (n=1,764 participants) indicated that maximizing life expectancy was most important. Typically, only the most severe level was statistically significant for other attributes. Participants preferred q8w infusions versus q2w (disutility = –0.058 [Australia]; –0.044 [Canada]; –0.070 [Netherlands]; –0.069 [Sweden]; −0.057 [UK]), and were concerned about: risk of meningitis (–0.034 [Australia]; –0.036 [Canada]; –0.046 [Netherlands]; –0.047 [Sweden]; −0.040 [UK]); severe hemolysis (–0.140 [Australia]; –0.132 [Canada]; –0.155 [Netherlands]; –0.153 [Sweden]; −0.158 [UK]), and the need for annual blood transfusions (–0.071 [Australia]; –0.016 [Canada]; –0.053 [Netherlands]; –0.084 [Sweden]; −0.073 [UK]).

CONCLUSIONS

We evaluated public preference for treatment attributes and estimated disutilities for events difficult to measure clinically. In this context, the study demonstrated the value that the general public places on attributes of treatment for PNH.