COST-EFFECTIVENESS OF NOVEL HEPATITIS C REGIMENS IN THE VETERANS HEALTH ADMINISTRATION

Purpose: Novel therapies for hepatitis C (HCV) are highly effective and very costly.  Current guidelines recommend sofosbuvir, a recently approved RNA polymerase inhibitor, with or without injectable interferon.  Trials suggest that more effective all-oral regimens will become available in 2015.  We compare the cost-effectiveness of treating HCV with current sofosbuvir-based regimens versus waiting to treat with future all-oral regimens.

Method: We developed a Markov model with 1-year cycle length for a cohort of 50-year old treatment-naïve Veterans with genotype 1, 2, or 3 HCV to compare: (1) treating all with current sofosbuvir regimens, (2) treating advanced disease with sofosbuvir and treating less advanced disease in one year with future all-oral regimens, and (3) treating all in one year with future all-oral regimens.  For comparison, we included the previous standard-of-care (interferon-based regimen with telaprevir/boceprevir) and no treatment. Patients could progress through stages of HCV and cirrhosis, develop hepatocellular carcinoma, undergo transplantation, or die.  Patients with sustained virologic response (SVR) had lower rates of progression, complications and mortality.  Analyses were performed from the VA healthcare system perspective, using a lifetime time horizon and discounting 3%/year. We performed one-way and probabilistic sensitivity analyses.

Result: In the base case, treating all genotype 1 patients with current sofosbuvir-based regimens was more costly but more effective ($105,151, 15.4 QALYs) than waiting to treat all patients with future all-oral regimens ($95,894, 15.1 QALYs), costing $36,234/QALY gained.  These strategies dominated all other strategies.  Treating all had an incremental cost-effectiveness ratio of $3,922/QALY for patients with genotype 2 and $23,813/QALY for genotype 3.  In sensitivity analysis, if sofosbuvir/simeprevir combination therapy’s SVR was <84%, treating now was no longer favorable at a $100,000/QALY threshold.  In probabilistic sensitivity analysis for genotype 1, waiting to treat with all-oral regimens was favored in 56% of iterations at $50,000/QALY; treating all patients with current sofosbuvir-based regimens was preferred in 44%.  Treating with sofosbuvir-based regimens was preferred above $75,000/QALY.

Conclusion: Treating patients with genotype 1, 2, or 3 HCV with currently available sofosbuvir-based regimens is more costly, but more effective than waiting one year to treat with future all-oral therapies.  This strategy is preferred above a willingness-to-pay threshold of $75,000/QALY.