The impact of direct acting antivirals on hepatitis C virus disease burden and associated costs in four european countries

Abstract Background and Aims We assessed the clinical and economic impact of direct‐acting antiviral (DAA) therapy for hepatitis C virus (HCV) in England, Italy, Romania and Spain. Methods An HCV progression Markov model was developed considering DAA eligibility and population data during the years 2015‐2019. The period of time to recover the investment in DAAs was calculated as the cost saved by avoiding estimated clinical events for 1000 standardized treated patients. A delayed treatment scenario because of coronavirus disease (COVID‐19) was also developed. Results The estimated number of avoided hepatocellular carcinoma, decompensated cirrhosis and liver transplantations over a 20‐year time horizon was: 1,057 in England; 1,221 in Italy; 1,211 in Romania; and 1,103 in Spain for patients treated during 2015‐2016 and 640 in England; 626 in Italy; 739 in Romania; and 643 in Spain for patients treated during 2017‐2019. The cost‐savings ranged from € 45 to € 275 million. The investment needed to expand access to DAAs in 2015‐2019 is estimated to be recovered in 6.5 years in England; 5.4 years in Italy; 6.7 years in Romania; and 4.5 years in Spain. A delay in treatment because of COVID‐19 will increase liver mortality in all countries. Conclusion Direct‐acting antivirals have significant clinical benefits and can bring substantial cost‐savings over the next 20 years, reaching a Break‐even point in a short period of time. When pursuing an exit strategy from strict lockdown measures for COVID‐19, providing DAAs should remain high on the list of priorities in order to maintain HCV elimination efforts.


| INTRODUTI ON
Hepatitis C virus (HCV) is a leading cause of liver-related morbidity and mortality worldwide. 1,2 The effect of antiviral therapy, in terms of the impact on clinical-related outcomes rather than virological efficacy, changes over time according to the epidemiological profile of the disease, specific patient characteristics including fibrosis stage, and antiviral treatment efficacy. [3][4][5][6][7] These factors vary by country, meaning country-specific epidemiology of HCV infection, DAA treatment guidelines, and treatment access are expected to impact the burden of HCV-liver related outcomes following viral eradication. No clinical studies have assessed the impact of antiviral therapy on long-term morbidity and mortality, however, as it is unethical to maintain patients without therapy.
Thus, only a modelling approach can address this issue and predict its impact on a population. 5 In this study, we built country-specific models using real-life data of fibrosis stage, genotype distribution and treatment eligibility for

| MATERIAL S AND ME THODS
A country-specific Markov model was designed to estimate the clinical and economic outcomes of expanded access to DAA therapy, considering the direct costs of HCV treatment in the European context (England, Italy, Romania and Spain). The model inputs are shown in Tables 1 and 2.

Key points
• Despite the country-specific dynamics and natural history of hepatitis C virus (HCV) infection in Italy, Spain, Romania, and England, expanding access to treatment will lead to a positive return on investment and is cost saving in <10 years.
• Not treating or delaying treatment of infected individuals will result in higher disease burden and costs, which could be avoided with immediate screening, linkage to care, and treatment of all HCV infected individuals.
• As lockdown orders for coronavirus disease 2019 (COVID-19) begin to lift, providing direct acting antivirals (DAAs) should remain a priority for public health officials in order to maintain HCV elimination efforts.
Conclusion: Direct-acting antivirals have significant clinical benefits and can bring substantial cost-savings over the next 20 years, reaching a Break-even point in a short period of time.
When pursuing an exit strategy from strict lockdown measures for COVID-19, providing DAAs should remain high on the list of priorities in order to maintain HCV elimination efforts.

K E Y W O R D S
break-even, DAAs, HCV elimination, hepatitis C infection TA B L E 1 Transition probabilities and efficacy of treatment (base-case, deterministic and probabilistic sensitivity analysis parameters)

| Transition probabilities
The probabilities of progressing through the various stages of disease were based on the results of a literature review ( All transition probabilities are adjusted for competing probabilities of death from other causes according to the official data from each country (Supporting Information Table A1). 16 For the HCC state, the probability of death because of HCV and the probability of transplant were assumed to be independent.

| Epidemiological and clinical parameters
The model simulates a cohort of 1000 standardized treated patients according to real-life fibrosis stage and genotype distribution data for England, Italy, Romania and Spain. 19-23 Fibrosis stage by age and genotype distribution of treated patients by year for each of the countries included in the study were collected. The fibrosis distribution of treated patients for each country is reported in Figure 1, while the genotype distribution for each country during the two time periods evaluated is shown in Table A2  outcomes were compared to a cohort of non-treated patients.

| Treatment efficacy
The efficacy of second-generation DAA regimens used in each period was stratified by genotype distribution and the presence or absence of cirrhosis (F0-F4, DC or HCC), as summarized in Table 1 and   Tables A3-A5 in the Supporting Information 3

| Scenarios
The model simulates two different scenarios considering 1000 standardized patients over a 20-year time horizon:

| Economic parameters and analysis
Direct healthcare costs were those associated with the manage-  (Table 2). 10,17,24,29 The average treatment cost of DAAs was estimated based on expert opinion and non-official sources.
The associated per-patient cost by disease stage post-SVR was assumed to be null, presuming a state of full health after SVR for the patient until F3 fibrosis stage (Table 2 and Supporting Information Figure A1). The associated per-patient cost by disease stage post-SVR from ILD remained associated to the costs of ILD prior to therapy, independently of the achievement of SVR (Table 2)  Costs were expressed in Euros and were discounted at a rate of 3% annually.

| Sensitivity analysis
To estimate the uncertainty of the economic results, probabilistic sensitivity analysis (PSA) and deterministic sensitivity analysis (DSA) were performed. The probabilistic distribution choice for cost was made by applying a gamma distribution and for epidemiological parameters, a beta distribution. 30 Tables 1 and 2

| Delayed treatment scenario
A delayed treatment scenario because of the COVID-19 pandemic was developed to estimate the incremental cases associated with the subsequent scenario: 1. Base-case 2020: all patients were treated during the first year and followed for 5 years;  Spain and Italy, but only a 25% increase in Romania. (Figure 1).

| Evaluation of clinical outcomes from 2015-2035
By expanding access to DAA therapies in 2017-2019, the model es-   In all countries, it would take less than 10 years to reach a Breakeven point (Table 3) Considering the confidence intervals around the base case values (Table 3)

| Delayed treatment because of COVID-19
As shown in Figure 4, there will be a progressive increase in HCV disease-related outcomes because of the delay in treatment caused by COVID-19. Generally, there will be an increase in the number of As COVID-19 has become a global pandemic, we considered the impact that the disease may having on delaying access to DAA therapy. Despite major differences in HCV epidemiology and disease burden among the evaluated countries, the delayed treatment scenario generates comparable liver related mortality in the four countries evaluated. Over time, severe clinical outcomes because of delayed treatment will be evident in England, Spain and Romania, which will however result in less mortality than seen in Italy. In these three countries, while progressive liver disease will be evident across the next 5 years if treatment is delayed, the HCV infected population, as seen by the treatment data, is more likely to be in earlier stages of disease. On the contrary, in Italy, the liver morbidity is generally equal to the liver mortality because among individuals who