Early Treatment of Hepatitis C Virus Improves Health Outcomes and Yields Cost-Savings: A Modeling Study in Argentina

Objective: Most untreated hepatitis C virus (HCV) patients develop chronic infection and severe complications, including death. Direct-acting antivirals in early stages of liver fibrosis reduce complications and healthcare costs. However, therapy is often delayed, and patients in early stages have limited access to effective treatments. We assessed the clinical and economic effect of treating chronic HCV at early versus late stages of disease in Argentina. Methods: A Markov model of the natural HCV history was used to forecast lifetime liver-related and economic outcomes from social security sector perspective. Healthcare use and transition probabilities were drawn from literature. Demographic characteristics of the patients and treatment attributes were based on data from registrational trials of glecaprevir/pibrentasvir. Results: Lower rates of all hepatic complications and liver-related mortality were predicted when treatment was initiated in mild versus advanced disease. Sustained virologic response rates were similar among all stages. Higher quality-adjusted life years (QALYs) were predicted when treatment was initiated in mild (F0-F1) versus moderate (F2-F3) or advanced (F4) liver disease (11.5, 9.9, and 7.5 QALYs, respectively). Delaying treatment increased long-term total lifetime costs (F4: AR$ 1 437 816; F2-F3: AR$ 967 673; F0-F1: AR$ 954 018; 37.10 AR$=1 USD, Nov 2018 exchange rate) and provided fewer QALYs. Conclusions: Our study show early treatment was a dominant strategy compared with treatment in advanced stages of liver disease. These results may help health policy makers take actions to reduce health and economic burden of HCV in Argentina.


Introduction
Hepatitis C virus (HCV) infection is considered a highly important healthcare problem (World Health Organization [WHO], 2017). This transmissible viral infection usually remains asymptomatic in the disease early stages but may lead to serious complications, including compensated cirrhosis (CC) and decompensated cirrhosis (DCC), hepatocellular carcinoma (HCC), and liver failure linked to significant morbidity and mortality rates (Moreno et al., 2017). The cost burden derived from HCV is associated with hepatic complications, as well as severe extrahepatic complications (eg, type 2 diabetes mellitus, depression and cognitive impairment, renal insufficiency, non-Hodgkin lymphoma) (Reau, Vekeman, Wu, & Sánchez-González, 2017). prevalence may be higher. Most data have been obtained by extrapolation and may not represent the real HCV infection prevalence (Ridruejo et al., 2016).
The main goal of antiviral treatment of HCV is to achieve sustained virologic response (SVR), defined as an absence of detectable HCV RNA≥12 weeks after treatment (Pearlman & Traub, 2011). The choice of the best treatment scheme for each individual patient depends on the clinical situation, presence of comorbidities, use of concomitant treatments, evaluation of liver damage, HCV genotype/subtype, and treatment efficiency. Currently, direct-acting antivirals (DAAs) are the reference therapies for chronic HCV in Argentina (Asociación Argentina para el Estudio de las Enfermedades del Hígado [AAEEH], 2020). HCV treatment can be very effective, especially if HCV is diagnosed in the early stages of the infection (Alomar et al., 2018). In addition, HCV treatment in early stages of the disease was associated with decreased liver-related complications and healthcare costs when compared with treatment in later stages of liver fibrosis (Pearlman & Traub, 2011;Reau, Vekeman, Wu, & Sánchez-González, 2017). Nevertheless, treatment start is often delayed due to several factors (patient level: lack of awareness, fear of side effects, poor adherence, and comorbidities; provider level: limited knowledge, lack of availability, and communication difficulties; payer level: lack of promotion, surveillance, and costs) (Mendizábal, Alonso, & Silva, 2019). As a consequence, patients in early fibrosis stages may have restricted access to effective treatment regimens. Specific actions to increase accessibility to HCV diagnosis and timely treatment are still pending in most Latin American countries, including Argentina. It is estimated that less than one-third of infected patients have been diagnosed (Kershenobich et al., 2011); according to recent epidemiologic data, only 8.9% of diagnosed patients currently are being treated in Latin America (Viola et al., 2020).
The combination of glecaprevir (GLE 300 mg, a NS3/4A protease inhibitor) and pibrentasvir (PIB 120 mg, a NS5A viral protein inhibitor) administered once daily is indicated for the treatment of adult patients with genotypes 1 to 6 HCV chronic infection with or without CC. According to current guidelines, this pangenotypic scheme is indicated in an 8-week schedule in selected populations (AAEEH, 2020).
Availability of optimized treatment schemes may lead to even better responses with the goal of a complete eradication of this disease. However, there is a need to evaluate the financial healthcare impact of DAA schemes. In the present study, we assessed the clinical and economic consequence of treating patients with HCV at early versus late stages of liver disease in Argentina.

Method
A Markov model of the natural history of HCV was used to forecast liver-related and economic outcomes over a lifetime; the model scheme has been described elsewhere (Liu et al., 2012). The model was run from the perspective of Argentina's social security sector, which represents approximately one-third of the total healthcare system. The social security sector in Argentina covers all employers of the formal economy and their families, is financed with payroll contributions of employers and employees, and often operates through contracts with private providers (Bello & Becerril-Montekio, 2011). Healthcare use and transition probabilities were drawn from published literature (Bardach et al., 2019;Saraswat et al., 2015). Demographic characteristics of the patients and treatment attributes were based on data from clinical trials of GLE/PIB (Forns et al., 2017;Wyles et al., 2018;Asselah et al., 2018;Zeuzem et al., 2018). Analyses were performed for patients with all HCV genotypes and distinct fibrosis stages of liver disease, defined by Metavir score: mild (F0: no fibrosis; F1: portal fibrosis without septa), moderate (F2: portal fibrosis with few septa; F3: numerous septa with incomplete nodules), and severe disease (F4: CC), both in naive or treatment-experienced patients (Bedossa & Poynard, 1996). DCC, HCC, and need for liver transplantation were also included. Without treatment, spontaneous viral clearance is only possible from F0 stage. Death may occur in any model stage ( Figure 1). The main demographic and virologic data of naive and treatment-experienced patients are shown in Table 1. Probability rates of fibrosis progression and health condition were obtained from available literature (  Health outcomes included lifetime risk of CC, DCC, HCC, liver transplantation, and liver-related death. Other outcomes included lifetime costs and quality-adjusted life years (QALYs), both discounted at a 5% rate, with a time horizon of 70 years (Augustovski et al., 2010;MERCOSUR, 2009). An analysis strategy with a portfolio approach was performed, considering a pangenotypic HCV patient population, regardless of treatment history and presence of cirrhosis. Costs of the model were obtained from reimbursement tariffs of the social security gjhs.ccsenet.org Global Journal of Health Science Vol. 14, No. 11;2022 30 institutions and expressed in Argentine pesos (exchange rate 37.10 AR$ = 1 USD, November 2018) (Table 3). Probabilistic sensitivity analysis (PSA) was carried out to examine the robustness of model results. In the PSA, it was assumed that each input parameter are associated with a certain probabilistic distribution. The parameters of the variables considered in the PSA are shown in Table 2 and Table 3. For the PSA, 500 simulations were drawn from the variables' distributions. The results are summarized graphically using the cost-effectiveness plane, where the horizontal axis indicates the total QALYs, and the vertical axis indicates the discounted total cost of each fibrosis stages of liver disease.

Results
In base-case, when treatment was initiated in milder versus advanced stages, lower rates of DCC, HCC, liver transplant, and liver-related death were predicted in HCV patients (Figure 2). Rates of SVR were similar among patients with mild (F0-F1: 97.9%), moderate (F2-F3: 97.9%), and severe disease (F4-CC: 98.9%). Nevertheless, higher QALYs were predicted in the Markov model when GLE/PIB scheme was started at early stages, accounting for 11.5, 9.9, and 7.5 lifetime QALYs for mild, moderate, and advanced liver disease, respectively (Figure 3). Lower direct medical costs were also predicted in a context of early treatment with GLE/PIB when compared with patients starting therapy in moderate or advanced liver disease (F0-F1: AR$ 954 018; F2-F3: AR$ 967 673; and F4: AR$ 1 437 816).  Vol. 14, No. 11;2022 Exchange rate 37.10 AR$ = 1 USD, November 2018. Direct costs included extrahepatic complication. F0-F1: mild liver disease (no fibrosis-portal fibrosis without septa); F2-F3: moderate liver disease (portal fibrosis with few septa-numerous septa with incomplete nodules); F4: severe liver disease /compensated cirrhosis; GLE/PIB, glecaprevir/pibrentasvir; QALY, quality-adjusted life year. Results of the PSA prove the robustness of the base-case findings. These confirm that treating chronic HCV infection at early fibrosis stages has better outcomes and lower total healthcare costs than treatment in the severe liver disease stage (Figure 4).

Discussion
Hepatitis C is considered a global public health problem. According to the Centers for Disease Control and Prevention, up to 70% of people infected with HCV will develop chronic liver disease, 5% to 20% will develop cirrhosis in a period of 20 to 30 years, and 1 to 5 subjects will die from cirrhosis or HCC (Stasi, Silvestri, & Voller, 2020). The World Health Organization recommends treating all persons with chronic HCV infection older than 12 years with pangenotypic DAAs, with the goal of cure (WHO, 2021). Evaluation of costs is highly relevant for inclusion of these treatment schemes in healthcare systems.
Although in Argentina there is no formal centralized and consolidated instance for health technology assessment, this type of analysis (integrating clinical information with economic data in connection with a new clinical intervention) is essential for rational decision-making that is not based only on implicit health coverage. In our study, which used a Markov model for the social security sector of the Argentine healthcare system, early treatment with GLE/PIB generated more QALYs at lower cost and was therefore a dominant strategy.
Our data showed that treating chronic HCV infection at early fibrosis stages has better health outcomes in terms of lifetime risks of DCC, HCC, liver transplantation, and liver-related mortality. In addition, early pangenotypic treatment with GLE/PIB was related to a reduction of total healthcare costs in Argentina, especially when comparing both mild and moderate stages with patients with F4 Metavir score or CC. This benefit in medical costs included extrahepatic complications of HCV.
Our results are in line with previously published research by other authors from developed countries (Ahmed, (Ahmed, Gordon, Saab, & Younossi, 2014). Furthermore, Buti et al. reported similar results using a model in the Spanish setting. They showed that, when compared with delayed administration of therapy at the F4 stage, initiating DAA treatment at stages F2-F3 reduced the incidence of liver-disease complications and was associated with cost-savings for the Spanish National Health System in previously untreated patients with genotype 1 HCV (Buti et al., 2016).
In our model, the GLE/PIB treatment regimen has been associated with very high overall SVR rates in all stages of the disease, as previously reported in real-world studies (Hsu et al., 2019;Lampertico et al., 2020;Ridruejo et al., 2020). However, higher QALYs were predicted when treatment was started in mild stages versus moderate or advanced liver disease. The QALYs estimation is useful to assess both the effect of a treatment on life expectancy and quality of life (Ogden, 2017). As a consequence, QALYs represent time alive, scaled to reflect health state desirability (Neumann & Cohen, 2018).
Considering these data and that the coronavirus disease 2019 has placed a significant strain on national healthcare systems at a critical moment in the context of hepatitis elimination (WHO, 2016), this analysis may help decision makers to reprioritize programs and resources to achieve HCV elimination targets and reap the full benefits of early HCV treatment (Blach et al., 2021).
In conclusion, from a perspective of social security of Argentina, treating chronic HCV infection at early fibrosis stages improves health outcomes and reduces total healthcare costs. Our results may help public health authorities take actions to reduce the relevant health and economic burden of HCV in Argentina by avoiding delays and restrictions in HCV treatment accessibility.