OBJECTIVES: To evaluate the cost-effectiveness of using intravitreal aflibercept (IVT-AFL) in comparison to ranibizumab 0.5 mg treat-and-extend (T&E) in the treatment of diabetic macular edema (DME) under the Brazilian public healthcare system (SUS).

METHODS: The model population consisted of patients with DME without previous treatment with anti-vascular endothelial growth factor agents. It was assumed that 46.5% of the patients developed DME in the second eye as soon as they entered the model. The health states were defined by visual acuity (VA) classified from VA1 (best VA) to VA8 (worst VA). Differences between each stage of VA were denominated by 10 incremental letters. The perspective of the analysis was SUS and the time horizon was 15 years. Each cycle consisted of one month. A 5% discount in costs and outcomes was applied. As economic outcomes, only the direct costs were considered, including drug acquisition, treatment maintenance and monitoring. Clinical outcome was the quality adjusted life-years (QALY). The clinical data and dosage were based on the VISTA-DME/VIVID-DME studies and an indirect comparison conducted by Korobelnik et al. (2015). The utility estimation associated to loss of vision was obtained from the Czoski-Murray et al. (2009) study. A probabilistic sensitivity analysis was performed to verify the model uncertainties.

RESULTS: The incremental total treatment cost was higher for ranibizumab, and the incremental QALY result was positive for IVT-AFL in comparison to ranibizumab. These results demonstrated that IVT-AFL was a dominant alternative in this scenario. The probabilistic sensitivity analysis indicated that 99% of the scenarios resulted in IVT-AFL being less costly with higher efficacy compared to ranibizumab.

CONCLUSIONS: This economic evaluation demonstrated that IVT-AFL can promote resource economy with better clinical benefits when compared to ranibizumab in the treatment of DME under the SUS perspective.