ISCL is a Intelligent Information Consulting System. Based on our knowledgebase, using AI tools such as CHATGPT, Customers could customize the information according to their needs, So as to achieve

All-Oral, Interferon-Free Treatment for Chronic Hepatitis C

3
All-Oral, Interferon-Free Treatment for Chronic Hepatitis C

Materials and Methods


Using TreeAge Pro 2012 software (TreeAge Software, Inc., Williamstown, MA, USA), we constructed a decision-analytic Markov model to simulate the progression of a 50-year-old cohort through CHC natural history and possible treatment with either SOC or an all-oral regimen. Cohort age was chosen based on CDC estimates of peak US HCV seroprevalence in the current 50–59 age group (4.3%), coupled with the expected rise in CHC-related healthcare costs as infected individuals in this cohort progress to late-stage liver disease.

Markov Model


A Markov model (Fig. 1) is a recursive decision tree that guides a cohort through a series of probabilities representing disease natural history, medical care, possible treatment and treatment outcomes. Based on the probability-driven pathways in our model (Fig. 2 and 3), subjects accrued costs and quality-adjusted life years (QALYs) at the end of each model year (stage), depending on their disease state and treatment profile. Cumulatively, these accruals were used to calculate the incremental cost-effectiveness ratio (ICER), which measures the cost per QALY gained by implementing all-oral treatment compared to SOC. Death was possible from any model stage. Subjects alive at the end of a given stage continued cycling through the model as determined by their health or treatment outcome in the preceding stage. Analysis terminated when the cohort reached its average life expectancy. Base case values for all model parameters, as well as ranges used in sensitivity analyses, can be found in Tables S1–S5.



(Enlarge Image)



Figure 1.



Simplified Markov model. As the HCV-positive cohort progresses through the model, subjects accrue medical costs and QALYs based on probabilities for disease progression and treatment outcome. Those who do not die from any cause during a given stage continue through the model for an additional year. Aggregate costs and QALYs are summed and used to calculate the ICER. HCV = hepatitis C virus; CHC = chronic hepatitis C; QALY = quality-adjusted life year; ICER = incremental cost-effectiveness ratio.







(Enlarge Image)



Figure 2.



Model schematic of CHC natural history and treatment. Model subjects progress through fibrosis stages F0–F4, decompensated cirrhosis and HCC based on annual probabilities (white circles, solid lines). Those who initiate SOC or all-oral treatment (light grey squares) either achieve SVR (black stars) or fail/discontinue and continue progressing through CHC natural history without retreatment (dark grey hexagons). Further fibrosis progression after SVR is possible for subjects in stages F3, F4 and decompensated cirrhosis (dotted lines). F3 subjects can progress directly to decompensated cirrhosis or HCC within one year, bypassing F4 (dotted lines). Subjects with decompensated cirrhosis can be treated in the all-oral treatment pathway, but not in the SOC pathway. Subjects with decompensated cirrhosis and HCC receive liver transplants according to published annual probabilities. Death is possible from any cause at any stage in the model. See Table S1 for specific progression probabilities. CHC = chronic hepatitis C; F0–F4 = Metavir fibrosis stages; Tx = treatment; SVR = sustained virologic response; HCC = hepatocellular carcinoma.







(Enlarge Image)



Figure 3.



Decision tree excerpt. A selection of the decision tree underlying the Markov model, where M represents the starting point for each annual model cycle, and individual probabilities are included under each branch. Subjects who reach terminal branches (boxes with thick borders) begin untreated progress through CHC natural history in the subsequent model year (solid black boxes). Chronic hepatitis C natural history progression probabilities are not depicted but are summarized in Table S1. Branches involved in treatment uptake and outcome are illustrated under F2 only, but this subtree was included in all fibrosis stages in the model. HCV = hepatitis C virus; SOC = standard of care treatment; CHC = chronic hepatitis C; F0–F4 = Metavir fibrosis stages; Tx = treatment.




Background Mortality Rates


Age-specific background morality rates were applied throughout the model, estimated at 2.37 times the rates for non-CHC-infected individuals, After SVR, subjects were assigned lower mortality rates, estimated at 1.4 times non-CHC rates based on evidence that virus clearance improves overall health outcomes. Subjects with advanced liver disease were assigned higher mortality rates based on published literature (Table S1).

Screening


This hypothetical cohort included only screened, laboratory confirmed HCV-positive subjects. It did not include HCV-positive individuals unaware of their infection because they would not have the opportunity to be treated with either drug regimen.

All-oral Parameter Estimates


Aside from SVR rates, which were estimated from clinical trial data, published values for parameters associated with all-oral CHC treatment were not available because these drugs are not yet approved. Estimates for specialist referral and attendance, contraindications to treatment, treatment uptake and discontinuation, and QALYs associated with all-oral treatment and SVR were derived from interviews with clinical hepatologists practicing at Emory University in Atlanta, Georgia, hereafter referenced as 'expert opinion' (Karpen S, Spivey J, Ford R, Parekh S. Personal communication). Due to the uncertainty of these estimates, most sensitivity analyses involving all-oral treatment parameters used ranges of at least ± 20%. To remain conservative, ranges often included the corresponding SOC parameter's base case (Tables S3–S5).

Fibrosis Progression


Probabilities of treatment uptake and SVR were dependent on CHC genotype and fibrosis stage, defined by METAVIR score (F0 = no fibrosis; F1 = portal fibrosis without septa; F2 = portal fibrosis with few septa; F3 = numerous septa without cirrhosis; F4 = compensated cirrhosis). Initial distribution of subjects across fibrosis stages was based on a meta-analysis of 111 clinical studies including over 33 000 individuals with CHC in fibrosis stages F0–F4, adjusted to include decompensated cirrhosis. Subjects progressed to later fibrosis stages, HCC, liver transplant and death based on annual progression probabilities from the same meta-analysis and other published estimates (Tables S1–S2). At the end of each model year in which a disease state transition occurred (e.g. progression from F0 to F1, or from F4 to decompensated cirrhosis), subjects accrued the costs and QALYs associated with the disease state in which they began the year; the following year, they accrued the costs and QALYs associated with the state to which they had transitioned.

Treatment Probabilities


Subjects had multiple opportunities to be treated. In the initial model year, a subject's probability of treatment was the product of individual probabilities for specialist referral, specialist attendance, likelihood of contraindications and acceptance of treatment that was offered, which varied by treatment pathway (Table S3; Fig. 3). Subjects eligible for treatment (i.e. with no immutable contraindications) but not treated in the initial model year transitioned to treatment in subsequent years at annual rates varying from 1–10% depending on genotype, fibrosis stage and treatment pathway. These probabilities were based on published literature for SOC and expert opinion for all-oral treatment,

Due to lower SVR rates for genotype 1 compared to genotypes 2 and 3 with SOC treatment, combined with physicians' expectations for improved treatments in the near future, many genotype 1 individuals in clinical care with little or no fibrosis delay treatment. To remain consistent with clinical practice, this model did not treat genotype 1 F0 subjects in the SOC pathway during the initial model year. However, these subjects transitioned to treatment at a higher rate in subsequent years compared to subjects in later fibrosis stages, as modelled in a recent cost-effectiveness analysis of CHC screening strategies by Coffin et al.. Because of high expectations for SVR with all-oral treatment, all genotype 1 subjects in the all-oral pathway were treated at the same rate in the initial model year, regardless of fibrosis stage. Subjects were assumed treatment-naïve. Those who failed or discontinued treatment were not retreated in the model.

Liver Biopsy


In the clinic, approximately 70% of genotype 1 subjects who attend a specialist visit undergo a liver biopsy to determine fibrosis stage, which helps determine whether to initiate treatment or to wait. If SVR rates rise as expected with the adoption of all-oral treatment, most individuals diagnosed with CHC will likely be treated regardless of genotype or fibrosis stage, reducing the need for liver biopsy. Based on expert opinion, we assumed that only 40% of genotype 1 individuals would receive a liver biopsy once all-oral treatment becomes widespread.

Treatment of Subjects With Advanced Liver Disease


Because few individuals with decompensated cirrhosis can tolerate interferon, treatment with SOC is rare in this group, One of the benefits of an eventual all-oral regimen is greater expected compatibility with late-stage disease, increasing treatment access for cirrhotic individuals. Therefore, this model allowed subjects with decompensated cirrhosis to be treated with all-oral therapy, but not with SOC. Estimates of treatment uptake and associated QALYs for decompensated subjects were derived from expert opinion. Treatment discontinuation rates were set at higher levels for these subjects compared to those in stages F0–F4 due to the possibility of more frequent adverse events.

Fibrosis Regression After SVR


To account for evidence of liver regeneration after viral eradication, this model allowed post-SVR fibrosis regression according to probabilities from clinical literature. Regression rates for decompensated subjects after all-oral treatment were assumed equal to rates for those with compensated cirrhosis.

Costs and QALYs


Most costs and QALYs associated with SOC treatment and nontreatment-related CHC care were derived from two recent cost-effectiveness studies on CHC screening and treatment strategies, which provide comprehensive reviews of these parameters from a variety of published sources, Treatment costs incorporate both the cost of the drugs themselves and the costs of medical care during treatment, including adverse events. Base case parameters were chosen as mid-range estimates (Tables S4-S5). Costs were adjusted to 2012 US dollars, and costs and QALYs were discounted 3% per year.

We used $70 000 as the base case cost for one course of all-oral treatment based on the anticipated market entry cost ($47 000) for sofosbuvir (GS-7977), one of the leading direct-acting antiviral candidates currently in advanced clinical trials, combined with expectations that eventual FDA-approved all-oral regimens will include more than one drug. Due to the uncertainty of this estimate, we used a wide range of all-oral drug costs in sensitivity analyses (Figs 4,5).



(Enlarge Image)



Figure 4.



Top ten most influential parameters for cost-effectiveness. A series of one-way sensitivity analyses, generated by TreeAge Pro software, depicts the influence that individual model variables exert on the ICER. The length of a given bar indicates the magnitude of change effected by variations in a model parameter. Variables were tested over ranges listed in Tables S1–S5. ICER = incremental cost-effectiveness ratio; QALY = quality-adjusted life year; CHC = chronic hepatitis C; F0–F4 = Metavir fibrosis stages; SVR = sustained virologic response.







(Enlarge Image)



Figure 5.



Maximum all-oral drug costs at three WTP thresholds. Cost of all-oral drugs was plotted against the ICER to determine the maximum drug cost at which all-oral treatment (dashed line) can remain cost-effective compared to SOC treatment (solid line) at various WTP thresholds. Maximum costs of all-oral drugs at WTP thresholds of $50 000/QALY, $80 000/QALY and $100 000/QALY are shown in black boxes. WTP = willingness-to-pay; SOC = standard of care treatment; ICER = incremental cost-effectiveness ratio; QALY = quality-adjusted life year.




ICER and Sensitivity Analyses


To assess cost-effectiveness, we calculated the incremental cost-effectiveness ratio (ICER), which measured the average cost per QALY gained by using all-oral treatment instead of SOC. We conducted one-way sensitivity analyses to determine which model parameters had the greatest impact on the ICER and ran sub-analyses to explore differences in cost-effectiveness by viral genotype and age at treatment.

Source...
Subscribe to our newsletter
Sign up here to get the latest news, updates and special offers delivered directly to your inbox.
You can unsubscribe at any time

Leave A Reply

Your email address will not be published.