| |
|
Should we treat acute hepatitis C? A decision and cost-effectiveness analysis
|
| |
| |
Download the PDF here
Download the PDF here
"In clinical practice, the treatment of acute HCV infection remains controversial. Current guidelines recommend deferring treatment to the chronic stage or monitoring patients with acute HCV for at least 12-24 weeks before initiating therapy. One of the driving factors in the decision to delay treatment is the desire to avoid needlessly administering therapy to patients who would have otherwise spontaneously cleared the virus. This was particularly true in the era of IFN-based HCV treatment regimens, when the risks associated with therapy were far greater than what we now see with the advent of highly effective and well-tolerated DAAs. In addition, because of the high cost of DAAs, payers are equally concerned about providing unnecessary treatment to patients who may ultimately not require therapy. No past analyses have looked specifically at weighing the savings associated with monitoring for spontaneous HCV clearance against the clinical benefits and cost reduction seen with shorter duration therapy in the treatment of acute HCV infection....In conclusion, our modeling-based analysis finds that treatment of acute HCV infection can improve outcomes and be highly cost-effective or cost-saving compared with deferring treatment to the chronic phase. Given the need to halt the rising incidence of HCV infection, if future studies demonstrate that effective acute HCV cure can be achieved with shorter treatment durations, then it may be time to revisit treatment guidelines to incorporate recommendations that account for the clinical and economic benefits of acute HCV treatment in the era of DAAs.
In patients at risk of transmitting HCV, using a per-person transmission rate estimate of 0.2, and therefore implying that treating 5 people with acute HCV could prevent one new HCV infection, QALYs increased by 0.03 and costs decreased by $3655, demonstrating that treatment of acute HCV was cost-saving (Table 2). Indeed, treating acute HCV was cost-saving even when the number of new infections transmitted was as low as 0.02 (Supporting Fig. S1). Greater cost-saving was achieved as the number of new infections transmitted by people with acute HCV infection increased. In the scenario analyses, evaluation of a range of acute and chronic treatment durations continued to demonstrate that acute HCV treatment remained either cost-effective or cost-saving (Table 2). In addition, we assessed possible scenarios in which patients may not adhere to treatment for acute HCV, or fail to return for chronic HCV therapy after the 6-month monitoring period. Treating acute HCV remained cost-effective with a loss to follow-up probability as high as 90%....Probabilistic sensitivity analysis showed that treating acute HCV was either cost-effective or cost-saving with a very high probability of 98.6% using a willingness-to-pay threshold of $100,000/QALY (Supporting Fig. S3)."
--------------------------------------------
Hepatology March 2018 - Emily D. Bethea,1-3* Qiushi Chen,1,2* Chin Hur,1-3 Raymond T. Chung,2,3 and Jagpreet Chhatwal1-3
from Jules: there is a little debate on when to start DAA therapy but by week 6-8 appears to be reasonable, some say immediately. HCV clearance has been reported at 20% to 40% and in this study 11% in HIV coinfected.
FUELING THE EPIDEMIC: Low Spontaneous HCV Clearance in Big European HCV/HIV Cohort - (03/15/18)
Abstract
It is not standard practice to treat patients with acute hepatitis C virus (HCV) infection. However, as the incidence of HCV in the United States continues to rise, it may be time to re-evaluate acute HCV management in the era of direct-acting antiviral (DAA) agents. In this study, a microsimulation model was developed to analyze the trade-offs between initiating HCV therapy in the acute versus chronic phase of infection. By simulating the lifetime clinical course of patients with acute HCV infection, we were able to project long-term outcomes such as quality-adjusted life years (QALYs) and costs.
We found that treating acute HCV versus deferring treatment until the chronic phase increased QALYs by 0.02 and increased costs by $483 in patients not at risk of transmitting HCV. The resulting incremental cost-effectiveness ratio was $19,991 per QALY, demonstrating that treatment of acute HCV was cost-effective using a willingness-to-pay threshold of $100,000 per QALY. In patients at risk of transmitting HCV, treating acute HCV became cost-saving, increasing QALYs by 0.03 and decreasing costs by $3,655.
Conclusion: Immediate treatment of acute HCV with DAAs can improve clinical outcomes and be highly cost-effective or cost-saving compared with deferring treatment until the chronic phase of infection. If future studies continue to demonstrate effective HCV cure with shorter 6-week treatment duration, then it may be time to revisit current HCV guidelines to incorporate recommendations that account for the clinical and economic benefits of treating acute HCV in the era of DAAs.
Abbreviations
DAAs
direct-acting antivirals
HCC
hepatocellular carcinoma
HCV
hepatitis C virus
ICER
incremental cost effectiveness ratio
IFN
interferon
LT
liver transplantation
PWID
people who inject drugs
QALYs
quality-adjusted life years
QoL
quality of life
SVR
sustained virological response
Hepatitis C virus (HCV) infection remains one of the main causes of chronic liver disease worldwide.1 An estimated 71 million people across the globe are currently infected with HCV, with approximately 3-4 million new infections occurring each year.2, 3 Persons with HCV virus can be classified as being in either the acute or chronic phase of infection. The acute phase of HCV infection is generally defined as occurring from the time of exposure up through 6 months of infection. If a patient with acute HCV does not spontaneously clear the virus within this 6-month time frame, then in the majority of cases chronic infection will subsequently develop.4
People who inject drugs (PWID) account for the largest portion of acute HCV infections in the United States.5, 6 Upfront prevention of virus acquisition is the best way to limit new cases of HCV infection; however, despite attempts at targeted interventions, such as needle exchange and opiate substitution programs, the incidence of HCV continues to rise.7 The annual number of reported HCV cases has tripled since 2010, with an estimated 34,000 newly infected Americans in the year 2015.8 Alongside continued efforts to minimize exposure risks in these populations, it could be worth revisiting current management approaches for patients with acute HCV infection.
Treatment options for HCV infection have undergone a major transformation with the introduction of direct-acting antivirals (DAAs). The success of these new agents has led to significant changes in consensus guidelines and recommended treatment options for patients with chronic HCV.9-11 Of note, these guidelines do not routinely recommend the initiation of therapy in patients with acute HCV infection. Current reasons for deferring treatment to the chronic stage are the (1) desire to avoid treating patients who would have otherwise spontaneously cleared the infection on their own, (2) limited data on the efficacy of DAAs in the acute phase of HCV infection, and (3) lack of studies evaluating the cost-effectiveness of early treatment strategies.9-11
During the era of interferon (IFN)-based regimens, patients being treated for acute HCV infection achieved higher sustained virological response (SVR) rates with shorter-duration therapy.12 A number of recent and ongoing pilot trials evaluating the efficacy of DAAs in acute HCV infection are beginning to demonstrate similar positive results. In particular, two studies-the German HepNet Acute HCV IV Study and the US SLAM C Study-showed 100% SVR with 4-6 weeks of treatment with ledipasvir/sofosbuvir.13, 14
With these recent developments, there is a need to revisit the trade-offs between the initiation of therapy in the acute versus chronic phase of HCV infection. Because the duration of treatment in the acute phase is potentially shorter (4-6 weeks) compared with treatment in the chronic phase (8-12 weeks), the cost of treatment in the acute phase is lower. An additional benefit of acute HCV treatment is to reduce potential transmission.15, 16 Conversely, if we defer treatment, approximately 25% of patients with acute HCV infection will spontaneously clear the virus,17 creating the opportunity to save on cost by monitoring for clearance and treating only those who develop chronic infection. To evaluate such trade-offs, we conducted a model-based analysis on acute versus chronic HCV treatment, providing some of the first cost-effectiveness data on treating acute HCV infection in the current era of DAA therapy.
Materials and Methods
We developed a microsimulation (individual-level state transition) model to compare treating HCV in acute versus chronic phase of the infection. This was done by extending our previously validated model18 to incorporate recent clinical data on the efficacy of DAAs in treating acute as well as chronic HCV infection. We simulated the clinical course of patients diagnosed with acute HCV infection and projected long-term outcomes such as quality-adjusted life years (QALYs) and costs. A weekly cycle length was utilized in the model. Simulated patients were followed until death.
PATIENT COHORT
Our base-case cohort was representative of the acute HCV patient population in the United States. Patient age at the start of the analysis was 26.4 All patients were assumed to be genotype 1 because this was the genotype included in the currently available studies investigating the efficacy of DAAs in the acute phase of HCV infection.13, 14
TREATMENT STRATEGIES
We simulated and compared two management strategies: (1) immediately treat acute HCV infection (referred to as "treat acute HCV" arm) versus (2) wait for 6 months and treat only those who develop chronic HCV infection (referred to as "treat chronic HCV" arm; Fig. 1). All patients received an HCV-RNA test and genotype test at initial diagnosis. In the treat acute HCV arm, after the diagnostic tests, patients received immediate treatment with 6 weeks of ledipasvir/sofosbuvir.13 Although two pilot studies showed 100% SVR with 4-6 weeks of treatment with ledipasvir/sofosbuvir,13, 14 these were small trials and we did not want to overestimate SVR efficacy by incorporating rates of 100%. We therefore made the assumption that the SVR rate in acute HCV treatment would at a minimum be equivalent to the SVR rate in chronic treatment and utilized SVR values from published studies on chronic HCV therapy.11, 19
In the treat chronic HCV arm, patients were followed for 6 months before receiving a second HCV-RNA test to determine whether spontaneous clearance had occurred. For our base-case analysis, we used a 25% spontaneous clearance rate obtained from a systematic review17 and further conducted sensitivity analysis on a wide range of spontaneous clearance values (0%-60%).11, 20 If patients demonstrated spontaneous clearance, then no further treatment was needed. Patients with a positive HCV-RNA test after 6 months were deemed to have developed chronic HCV infection and were subsequently treated with 8 weeks of ledipasvir/sofosbuvir, as recommended by the American Association for the Study of Liver Disease (AASLD-IDSA) guidance.11 All patients in our base case were treated for 8 weeks, acknowledging that there may be a subset of patients in whom 12 weeks of therapy is still the recommended treatment duration based on drug therapy choice. In both arms, patients who failed to achieve SVR were eligible for retreatment, and those who failed second-line therapy were assumed to follow the natural history of HCV. SVR status was confirmed by an RNA test 12 weeks following therapy completion.
NATURAL HISTORY OF HCV INFECTION
To account for the higher risk of non-liver-related death in the acute HCV patient population, we adjusted the background mortality with sex-specific hazard ratios (2.58 for men and 1.97 for women).21-23 Patients who failed to achieve SVR started at F0 fibrosis stage and progressed through different stages of liver fibrosis (F0-F4). Patients with F4 fibrosis stage or compensated cirrhosis could further progress to decompensated cirrhosis, hepatocellular carcinoma (HCC), or death from liver-related mortality.18 The rate of fibrosis progression, decompensation, HCC, and liver-related death were estimated from published systematic reviews24 and observational studies.25, 26 Patients with decompensated cirrhosis and HCC had higher excess mortality26, 27 and were eligible for liver transplantation (LT).28, 29 The likelihood of LT was estimated from previous studies.18, 30
COST AND QUALITY-OF-LIFE ESTIMATES
Our analysis was conducted from a third-party payers' perspective and included only direct medical costs. The recent listing price of the highly effective glecapravir/pibrentasvir is $3,300 per week; to ensure that our model cost input reflected the current DAA price environment, we used a 50% discount on the wholesale acquisition costs for ledipasvir/sofosbuvir of $7,875 per week (i.e., $3,937.50) for our base-case analysis.31 We also conducted one-way sensitivity analysis on drug price by further applying an additional discount up to 50% on top of our base-case drug price. The cost of HCV genotype and RNA testing were $435 and $98, respectively.32 We estimated health state-specific costs from our published studies30, 33 and conservatively assumed no additional costs for the acute HCV state. All cost estimates were converted to 2016 U.S. dollar values using the health care component of the Consumer Price Index.34
We assigned health-related quality-of-life (QoL) utility values to each health state in the model; utilities were bounded between a minimum value of 0 (denoting death) and a maximum value of 1 (denoting perfect health). All utility weights were extracted from previously published studies that used the EuroQol-5D instrument.35, 36 We assumed the QoL for the acute phase of HCV was equivalent to that for the F0 fibrosis state, and the QoL for post-SVR patients was equivalent to that of the general population. We also normalized all utilities with age- and sex- specific QoL weights representative of the U.S. population.
EFFECT OF ACUTE HCV TREATMENT ON HCV TRANSMISSION
We examined the potential impact of acute HCV treatment on preventing HCV transmission.15, 16, 37 In particular, we considered a cohort of patients who are at high risk of transmitting HCV, such as PWIDs, who account for the largest portion of acute HCV infections in the United States. It is possible that immediate treatment of acute HCV in these patients (vs. waiting until the chronic phase) can reduce potential HCV transmission. We conducted an analysis by assuming that an untreated acute HCV person could infect another 0.2 (range, 0.0-1.0) persons during the acute phase of the infection. We made this conservative assumption given that the equivalent estimates based on published parameters are typically greater than 0.5.38-43 Our assumption implied that treating 5 people with acute HCV could prevent one new HCV infection.
MODEL OUTCOMES
Our model simulated the lifetime clinical course of the treat acute HCV and treat chronic HCV strategies and projected expected average QALYs and costs for the base-case cohort. We then calculated the incremental cost-effectiveness ratio (ICER) of treating acute versus chronic HCV infection. All future QALYs and costs were discounted at 3% per year. To reduce the effect of the first-order uncertainty, the model was run 100,000 times.
SCENARIO ANALYSIS
We conducted several scenario analyses to evaluate model outcomes under different clinical settings. First, we considered 4- and 6-week treatment durations in acute HCV and 8- and 12-week treatment durations in chronic HCV infection. Second, we included the possibility of loss to follow-up in both the acute and chronic treatment arms. Because of limited evidence for this variable, we varied the probability of loss to follow-up from 0% to 90%.
SENSITIVITY ANALYSIS
We performed one-way sensitivity analysis to assess the impact of uncertainty in transition probabilities, QoL weights, costs, initial age, model time horizon, and background mortality on model outcomes (Table 1). Because both spontaneous viral clearance rate and DAA efficacy in acute HCV infection are two key parameters in our model, we conducted two-way sensitivity analysis by evaluating different combinations of these parameters. We also performed probabilistic sensitivity analysis to simultaneously account for uncertainty in all model parameters. We used the recommended statistical distributions for each parameter (Table 1) and conducted the analysis using 10,000 second-order samples (and 100,000 first-order samples).
Results
COST-EFFECTIVENESS OF TREATING ACUTE HCV
In patients who are not at risk of transmitting HCV, treating acute HCV increased QALYs by 0.02 and increased total costs by $483 as compared to deferring treatment to the chronic phase (Table 2). The resulting ICER was $19,991/QALY, substantially below the $100,000/QALY willingness-to-pay threshold, demonstrating that treatment of acute HCV is cost-effective.
In patients at risk of transmitting HCV, using a per-person transmission rate estimate of 0.2, and therefore implying that treating 5 people with acute HCV could prevent one new HCV infection, QALYs increased by 0.03 and costs decreased by $3655, demonstrating that treatment of acute HCV was cost-saving (Table 2). Indeed, treating acute HCV was cost-saving even when the number of new infections transmitted was as low as 0.02 (Supporting Fig. S1). Greater cost-saving was achieved as the number of new infections transmitted by people with acute HCV infection increased.
In the scenario analyses, evaluation of a range of acute and chronic treatment durations continued to demonstrate that acute HCV treatment remained either cost-effective or cost-saving (Table 2). In addition, we assessed possible scenarios in which patients may not adhere to treatment for acute HCV, or fail to return for chronic HCV therapy after the 6-month monitoring period. Treating acute HCV remained cost-effective with a loss to follow-up probability as high as 90% (Supporting Fig. S2).
SENSITIVITY ANALYSIS
One-way sensitivity analysis showed that for patients not at risk of transmitting HCV, the ICER was most sensitive to the spontaneous clearance rate, the QoL utility values of the acute HCV state and the post-SVR state, and the probability of lost to follow-up (Fig. 2A). When the drug price was changed by ±50% of the base-case value, the ICER of treating acute versus chronic HCV ranged from $9,000 to $30,000/QALY. For patients at risk of transmitting HCV, treatment in acute HCV was found to be cost-saving across all parameter ranges except for high values of the spontaneous clearance rate and probably of lost to follow-up (Fig. 2B). Model results were not sensitive to the starting age and the model time horizon (Supporting Table S1).
Figure 3 shows a two-way sensitivity analysis for the spontaneous clearance rate and the SVR rate in acute HCV treatment. Displayed on the graph are the regions in which varying combinations of these two parameters lead to cost-effective or cost-saving results. For instance, at 98% SVR, treating acute HCV is cost-effective/saving if the rate of spontaneous clearance is below 31% for patients not at risk of transmitting HCV (Fig. 3A) and below 44% for patients at risk of transmitting HCV (Fig. 3B).
Probabilistic sensitivity analysis showed that treating acute HCV was either cost-effective or cost-saving with a very high probability of 98.6% using a willingness-to-pay threshold of $100,000/QALY (Supporting Fig. S3).
Discussion
In clinical practice, the treatment of acute HCV infection remains controversial. Current guidelines recommend deferring treatment to the chronic stage or monitoring patients with acute HCV for at least 12-24 weeks before initiating therapy. One of the driving factors in the decision to delay treatment is the desire to avoid needlessly administering therapy to patients who would have otherwise spontaneously cleared the virus. This was particularly true in the era of IFN-based HCV treatment regimens, when the risks associated with therapy were far greater than what we now see with the advent of highly effective and well-tolerated DAAs. In addition, because of the high cost of DAAs, payers are equally concerned about providing unnecessary treatment to patients who may ultimately not require therapy. No past analyses have looked specifically at weighing the savings associated with monitoring for spontaneous HCV clearance against the clinical benefits and cost reduction seen with shorter duration therapy in the treatment of acute HCV infection.
In this study, we performed a modeling-based analysis and demonstrated that treatment with DAA therapy in acute HCV infection can be a cost-effective, and in fact cost-saving, strategy in the majority of scenarios. We sought to incorporate conservative estimates throughout our model to prevent overestimation of acute HCV treatment benefits. We used low values for SVR rates in acute treatment, and did not take into account the decrement in QoL that can be experienced by patients with acute HCV when therapy is not offered. In addition, all patients in our base cohort that developed chronic HCV were treated with 8 weeks of DAA therapy, despite a subset for which 12 weeks of therapy may still be recommended under current guidelines.44
Our study has several limitations. First, the duration of treatment in acute HCV infection was based on data obtained from a small number of pilot studies evaluating the treatment of acute HCV monoinfection with DAAs. However, given the robust data available on the efficacy seen with DAAs in chronic HCV, and the historic evidence that treatment of acute HCV results in higher SVR rates with shorter duration therapy, we anticipate that future investigations will find results that are comparable to the values used in our computations. Second, our analysis focused on a specific DAA regimen. We used price information for ledipasvir/sofosbuvir as model input, because this was the regimen utilized in the available studies evaluating DAAs in patients with acute genotype 1 HCV infection.13, 14 We realize that alternatives, such as highly effective pan-genotypic treatment options, may prove to be the therapy of choice for many patients, including genotype 3 infections found commonly in PWID. These regimens will have their own price profile that could impact model inputs; however, our sensitivity analysis on a wide range of drug prices did not alter study conclusions. If the duration of therapy difference holds true-4-6 weeks in acute infection versus 8-12 weeks in chronic infection-then the cost-effectiveness argument supporting treatment in acute HCV can still be generalized to other drugs.
Identifying patients with acute HCV remains challenging in clinical practice. Most patients remain asymptomatic during this phase, and the diagnosis of acute HCV infection requires both active testing to identify seroconversion and repeated testing over time. This task is often difficult at baseline, without considering the added hurdle of accessing and retaining some of the more marginalized populations that are not well connected to health or social services. Nevertheless, these challenges should not deter discussions on treatment initiation in patients with acute HCV infection who do present to care. Additionally, our study results provide important data that can be used to further support the development of strategies aimed at enhancing early HCV detection.
The treatment of HCV infection continues to be a rapidly evolving field. Most studies to date focus on the treatment of patients with chronic infection; however, further investigations into the treatment of acute HCV are needed. Furthermore, elimination of HCV is a stated objective made by the World Health Organization and the National Academy of Medicine in 2016.45, 46 Important to this goal will be recognizing that in the United States this virus is not just concentrated among baby boomers, but also spreading rapidly in a new generation of Americans.47-50 Our results suggest that incorporation of acute HCV treatment in those who drive the bulk of transmission should be considered as part of the management algorithm in HCV infection.
In conclusion, our modeling-based analysis finds that treatment of acute HCV infection can improve outcomes and be highly cost-effective or cost-saving compared with deferring treatment to the chronic phase. Given the need to halt the rising incidence of HCV infection, if future studies demonstrate that effective acute HCV cure can be achieved with shorter treatment durations, then it may be time to revisit treatment guidelines to incorporate recommendations that account for the clinical and economic benefits of acute HCV treatment in the era of DAAs.
| |
| |
| |
|
|
|
