Sacubitril/valsartan in heart failure with reduced ejection fraction: cost and effectiveness in the Italian context

The use of angiotensin-converting enzyme inhibitors (ACEIs), angiotensin receptor blockers (ARBs), beta-blockers and mineralocorticoid receptor antagonists (MRAs) has led to significant improvements in mortality and morbidity in patients with heart failure (HF) with reduced ejection fraction (HFrEF). However, HFrEF is still associated with high rates of morbidity and mortality, and high health care costs.1,2 Recently, a treatment option for HFrEF, sacubitril/valsartan (Entresto®; Novartis International AG; Basel, Switzerland), was approved by the US Food and Drug Administration and by the European Medicines Agency, and recommended in the new guidelines of the European Society of Cardiology and American College of Cardiology/American Heart Association as a replacement for an ACEI to further reduce the risk for HF hospitalization and death in ambulatory patients with HFrEF who remain symptomatic despite optimal treatment with an ACEI/ARB, beta-blockers and an MRA. However, the cost-effectiveness of sacubitril/valsartan has been studied mainly in US settings3–5 and few data are available for cost-effectiveness in European countries.6 Given the differences in the characteristics of the respective HFrEF populations in the USA and Europe established in the PARADIGM-HF trial7,8 and variations in the health systems, management of patients and health care costs, we performed a cost-effectiveness analysis of the use of sacubitril/valsartan in Italy, which is one of the most populated European countries. A previously published Markov model3 was used to simulate the course of HF and the impact of treatment options based on data derived from the PARADIGM-HF trial.7 Using this model, a cost-effectiveness analysis was conducted to compare treatment with sacubitril/valsartan b.i.d. in the Italian HFrEF population (Table 1). Expected clinical benefits and costs were estimated for each simulated treatment based on Italian National Health System practice and using a lifetime horizon in the base case analysis. The model estimated the costs related to the treatments (primary therapy and background therapy), costs associated with HF management (specialist evaluation and diagnostics) and hospitalizations, life years, quality-adjusted life years (QALYs) and the incremental costeffectiveness ratio (ICER). Future costs and clinical benefits were discounted at a rate of 3.5% per year. The details of the model have already been described in a previous article.3 Briefly, it is a two-state Markov model in which a patient with HFrEF has a monthly risk for surviving without further complication, becoming hospitalized or dying. Cardiovascular (CV) mortality and hospitalization for patients with HFrEF follows a Gompertz hazard function based on the PARADIGM-HF trial adjusted for the characteristics of the Italian HFrEF population (Table 1).1,3,4 The probability of non-CV mortality was derived from mortality tables for the general Italian population. In the PARADIGM-HF trial, sacubitril/valsartan was found to have higher efficacy than enalapril in reducing CV mortality and hospitalizations; these effects were applied to the model (Table 1). The model assumed that treatment effects would last until the patients discontinued the treatment or died. A probability of discontinuation of drugs during the first 10 years simulated was assumed in the model (Table 1). When sacubitril/valsartan is discontinued, patients are switched to enalapril treatment; when enalapril is discontinued, patients are switched to losartan. Only direct medical costs associated with hospitalization, pharmacological therapies and HF management were considered in the analysis (Table 1). A health-related quality of life (HRQoL) decrement was applied to the model based on data collected using the EuroQol five dimensions questionnaire (EQ-5D) in the PARADIGM-HF trial (Table 1). Sacubitril/valsartan was associated with an increment of utility value of 0.011 compared with enalapril.5 At 10 years, the base case analysis showed cumulative incidences of CV mortality of 37.2% in patients treated with enalapril and 33.8% in those treated with sacubitril/valsartan. Over the course of the entire simulation, for every 1000 patients, approximately 243 hospitalizations were avoided in those treated with sacubitril/valsartan compared with those treated with enalapril. When hospitalizations for HF diagnoses only and for HF plus other CV diagnoses were considered, the numbers of admissions avoided were 94 and 184, respectively. In the base case analysis performed with a lifetime horizon, average survival times in patients treated with sacubitril/valsartan or enalapril were 10.70 years and 10.24 years, respectively. The discounted QALYs gained and the overall discounted costs per patient were 6.16 QALYs and €35 943, respectively, in patients treated with sacubitril/valsartan, and 5.88 QALYs and €30 581, respectively, in patients treated with enalapril. The administration of sacubitril/valsartan increased overall costs, although this increase was partially offset by the reduction in hospitalization costs. Compared with enalapril, sacubitril/valsartan had an ICER of €19 487 per QALY gained, which is below the usually accepted willingness-to-pay (WTP) threshold of €40 000 per QALY gained (Figure 1). Variables with the highest impact on the ICER were the effect of sacubitril/valsartan on CV mortality and hospitalization rate, and the probability of hospitalization with enalapril. The results remained favourable for sacubitril/valsartan in all sensitivity analyses performed. The alternative scenario analysis conducted in specific subpopulations categorized according to New York Heart Association (NYHA) functional class showed that sacubitril/valsartan provided an ICER of less than €20 000 per QALY gained in both patients with NYHA class I or II status and those with NYHA class III or IV status (Figure 1). Assuming a short time horizon (10 years) in the analysis, the use of sacubitril/valsartan remained cost-effective with an ICER of €26 180 per QALY gained (Figure 1). In the probabilistic sensitivity analysis, the probability that sacubitril/valsartan would be cost-effective was >85% at a WTP threshold