| |
Modelling the impact of deferring HCV treatment on liver-related complications in HIV coinfected men who have sex with men (Swiss Cohort)....Deferring Therapy Increased 10-fold mortality & morbidity
|
| |
| |
Download the PDF here
"HCV therapy should be accessible to everyone at an early stage.....deferring treatment until advanced fibrosis increased liver-related morbidity and mortality... This [early therapy] may decrease further HCV transmissions in those with high-risk behavior......Our findings support current recommendations to start HCV treatment irrespective of fibrosis stage in those with risk factors for accelerated fibrosis progression including HIV-coinfected MSM, and in persons at elevated risk of HCV transmission"
"The effect of deferring HCV treatment until later stages of liver fibrosis is shown in Figure 3a. The percentage of simulated individuals who died of liver-related complications was 2% if treatment was initiated in F0 or in F1. It rose to 3% if treatment was deferred until F2, 7% if deferred until F3, and 22% if deferred until F4.
Of individuals who died of liver-related complications, less than 1% died after clearing HCV if they were treated as they reached F0 or F1. This percentage increased to 2% if treatment was deferred until F2, 6% if it was deferred until F3, and 17% if it was deferred until F4 (Figure 3b). A large proportion of liver-related deaths occurred in individuals without replicating HCV if treatment was deferred until advanced fibrosis or cirrhosis as the model assumed that SVR substantially reduces the risk of liver disease progression but does not eliminate it [10-15]. The median time spent with replicating HCV increased from 5 years if treatment was initiated in F2 to almost 15 years if treatment was deferred until F4 (Figure 4). The percentages of individuals who died from liver-related complications depending on the follow up time since HCV infection are shown in Supplementary Table S2.
We show that initiating HCV therapy in F2, instead of F3 or F4, reduced the time individuals spent with replicating HCV by 47-64% as compared to when therapy is started in F1. Initiating therapy in F1, instead of waiting until F3 or F4 reduced the median time spent with replicating HCV by 85-90%. Early treatment reduced the median time with replicating HCV
Our model showed that initiating HCV therapy in METAVIR stage F2 instead of deferring treatment until stage F3 or F4 could prevent 4-19 liver-related deaths per 100 HCV infections. In the scenario where all diagnosed individuals are treated with DAAs in METAVIR stages F3 or F4, most liver-related deaths were caused by liver disease progression after HCV clearance, rather than because of treatment failure or a lack of diagnosis. Thus, if treatment is deferred until advanced fibrosis or cirrhosis has developed, most liver-related deaths will occur after HCV is cleared. HCV clearance is often associated with fibrosis regression, but liver fibrosis may progress in some individuals after HCV clearance [10, 12, 16, 17, 39-42]. Accordingly, deferring treatment until advanced fibrosis increased liver-related morbidity and mortality in all scenarios except when we assumed that liver fibrosis never progressed after SVR, or in a scenario with an extremely fast fibrosis progression."
Journal of Hepatology Feb 2016
Cindy Zahnd, Luisa Salazar-Vizcaya, Jean-François Dufour, Beat Müllhaupt,
Gilles Wandeler, Roger Kouyos, Janne Estill, Barbara Bertisch, Andri Rauch,
Olivia Keiser, The Swiss HIV and the Swiss Hepatitis C Cohort Studies
Abstract
Background and aims
Hepatitis C (HCV) is a leading cause of morbidity and mortality in people who live with HIV. In many countries, access to direct acting antiviral agents to treat HCV is restricted to individuals with advanced liver disease (METAVIR stage F3 or F4). Our goal was to estimate the long term impact of deferring HCV treatment for men who have sex with men (MSM) who are coinfected with HIV and often have multiple risk factors for liver disease progression.
Methods
We developed an individual-based model of liver disease progression in HIV/HCV coinfected men who have sex with men. We estimated liver-related morbidity and mortality as well as the median time spent with replicating HCV infection when individuals were treated in liver fibrosis stages F0, F1, F2, F3 or F4 on the METAVIR scale.
Results
The percentage of individuals who died of liver-related complications was 2% if treatment was initiated in F0 or F1. It increased to 3% if treatment was deferred until F2, 7% if it was deferred until F3 and 22% if deferred until F4. The median time individuals spent with replicating HCV increased from 5 years if treatment was initiated in F2 to almost 15 years if it was deferred until F4.
Conclusions
Deferring HCV therapy until advanced liver fibrosis is established could increase liver-related morbidity and mortality in HIV/HCV coinfected individuals, and substantially prolong the time individuals spend with replicating HCV infection.
INTRODUCTION
Liver disease has become a leading cause of mortality in people who live with HIV (PWLH); it is often caused by infection with the Hepatitis C virus (HCV) [1, 2]. In high-income countries, about 30% of HIV-positive individuals are coinfected with HCV, though the proportion varies by risk group. As many as 70-90% of HIV-positive intravenous drug users are coinfected with HCV [3]. In the population of HIV-positive men who have sex with men (MSM) [4-6], HCV incidence has increased in recent years. The accelerated fibrosis progression observed in some studies [7-9], and the high incidence of HCV seroconversions and reinfections underscore the need for reliable predictions of the HCV disease burden and of the optimal therapeutic interventions in this population. Successful HCV treatment greatly reduces the risk of decompensated cirrhosis, hepatocellular carcinoma (HCC) and extrahepatic complications, but does not eliminate it [10-15]. Because HIV connected individuals have multiple risk factors for liver disease, including drug toxicity and metabolic liver disease, they
might be at increased risk to have liver-related complications even after they clear HCV [12, 14, 16].
We do not know if treatment can be deferred until METAVIR stages F3 without increasing the risk of liver-related complications [17].
For the last decade, the standard of care for people infected with HCV has been treatment with pegylated-interferon- (PEG-IFN) plus ribavirin (RBV). This IFN-based regimen is challenging to use, especially in HIV coinfected individuals who are at high risk for serious side-effects and have a low probability of cure [18-20]. Recently, new direct acting antivirals (DAAs) have revolutionized the treatment of HCV. These compounds are very effective, easy to use, and have few contraindications. These are factors that greatly increase the proportion of PWLH eligible for HCV treatment [21-24]. Yet the very high cost of the DAAs represents a major barrier to widespread treatment scale up and is a matter of debate [25]. Although the European Association for the Study of the Liver (EASL) now recommends that individuals coinfected with HIV are prioritized for treatment regardless of their fibrosis stage [26], reimbursement of HCV therapy is often restricted to individuals with advanced liver fibrosis [17, 27-29] .
We set out to estimate the impact of deferring HCV treatment on of liver-related complications in HIV coinfected individuals by using a model of liver disease progression and care. Our main outcomes of interest were liver-related morbidity and mortality as well as the time spent with replicating HCV.
DISCUSSION
Principal findings
Over a lifetime, deferring HCV treatment until advanced liver disease stages is likely to substantially increase liver-related complications, increase the time individuals spend with replicating HCV, and may not save money.
In many settings, cost considerations and related limitations in reimbursement by health insurances have led the authorities to recommend that HCV treatment be deferred until METAVIR stage F3 or more. Our model showed that initiating HCV therapy in METAVIR stage F2 instead of deferring treatment until stage F3 or F4 could prevent 4-19 liver-related deaths per 100 HCV infections. In the scenario where all diagnosed individuals are treated with DAAs in METAVIR stages F3 or F4, most liver-related deaths were caused by liver disease progression after HCV clearance, rather than because of treatment failure or a lack of diagnosis. Thus, if treatment is deferred until advanced fibrosis or cirrhosis has developed, most liver-related deaths will occur after HCV is cleared. HCV clearance is often associated with fibrosis regression, but liver fibrosis may progress in some individuals after HCV clearance [10, 12, 16, 17, 39-42]. Accordingly, deferring treatment until advanced fibrosis increased liver-related morbidity and mortality in all scenarios except when we assumed that liver fibrosis never progressed after SVR, or in a scenario with an extremely fast fibrosis progression. This is plausible since many risk factors associated with fibrogenesis, including drug toxicity, alcohol use, coinfections or metabolic liver disease, persist after cure. HCC can occur in those with cirrhotic livers even after they clear HCV [10]. Reinfections have been observed in up to 22% of patients following
spontaneous or treatment-induced HCV clearance [38]. As expected, the benefit of treating individuals earlier was partially offset through reinfections and the proportion of patients who experienced liver related events was higher if reinfections were considered (see Figures S2 and S3). However, even in a worst-case-scenario assuming a very high reinfection rate and no pretreatment, treating earlier reduced liver-related complications.
We show that initiating HCV therapy in F2, instead of F3 or F4, reduced the time individuals spent with replicating HCV by 47-64% as compared to when therapy is started in F1. Initiating therapy in F1, instead of waiting until F3 or F4 reduced the median time spent with replicating HCV by 85-90%. Early treatment reduced the median time with replicating HCV even in our worst case scenario where reinfected individuals were not retreated. This may decrease the risk of further HCV transmission in those with high-risk behavior. This is particularly important for HIV-positive MSM since this population is in the midst of an increase in HCV transmissions. Earlier initiation of treatment could be a valuable preventive strategy, akin to the concept of treatment-as-prevention in HIV, which was established as a very effective measure to reduce HIV transmissions [43]. A recent study in the SHCS found that increased treatment uptake and efficacy can reduce the proportion of individuals with replicating HCV infection [37].
Our cost calculations suggest that, despite the very high cost of treatment, early treatment might not increase total spending, since the increase in treatment cost is balanced by the savings in health care costs. This is assuming that prices of DAA therapy do not decrease in the coming years.
Comparison with other studies
Three other studies modelled the effect of timing of HCV therapy. The first investigated the effect of deferring HCV therapy in HCV-genotype 1 monoinfected individuals [44]. Researchers compared the cost-effectiveness of initiating therapy in different stages of liver disease and found it did not have much impact on the life expectancy. The second study examined the cost-effectiveness of early HCV treatment for individuals with HCV monoinfection and concluded that treating those with moderate or advanced fibrosis was cost-effective; the cost-effectiveness of treating those with minimal or no fibrosis depended on the cost of treatment [45]. The third study estimated the quality-adjusted life years for a 40 years old patient to increase from 23.9 if treatment was started in F4 to 33.7 if treatment started in F0 assuming and SVR rate of 90% [46]. These studies considered only cohorts of HCV monoinfected individuals. The first assumed that successful HCV treatment would eliminate further risk of liver disease progression, the second and the third assumed that only individuals treated in F4 were still at risk of liver-disease progression after HCV clearance. In contrast, our model assumes that liver fibrosis progresses in some individuals [13, 14, 16, 40], which led to an increase in liver-related events if therapy was deferred until F3.
Of note, a recent cost-effectiveness analysis among HIV/HCV coinfected patients suggested that IFN-free regimens will be cost-effective if treatment costs were below 109'000 USD, which is now the case in many settings [47]. For the Swiss setting, a recent study suggested that DAA-based therapies were cost-effective even at current prices if a threshold of 100,000 CHF per QALY was assumed [48].
Another study published in 2015 demonstrated that cost-effectiveness is highly sensitive to drug prices and that treating patients in F0 would be cost-effective if treatment costs were below 50'000 USD [49]. A recent study [50] showed that the life-expectancy of HCV-monoinfected individuals who had been successfully treated in an advanced liver-disease stage was comparable to that of the general population. The apparent inconsistency between this finding and our results can be explained by the differences between the cohorts, including different patient characteristics and very different follow-up times. The median follow-up time in that study was 8.4 years, while we make predictions over a life-time. In fact, when we simulated a cohort of individuals cured in an advanced stage of the disease, our model predicted a very low percentage (1.2%) of liver-related deaths after 8.4 years of follow-up (see appendix). A recent meta-analysis [10] estimated a 5-year risk of HCC after SVR of 2.9% in the overall population, and 5.3% among cirrhotic individuals.
Strengths and limitations
Our study was strengthened by our access to observed data from a large and nationally representative cohort of PWLH. Data were collected prospectively during regular follow-up visits and include detailed demographic, clinical and laboratory data on HIV and HCV infections. The individual-based design of our model enabled us to exploit this detailed information. The use of the R package 'gems' allowed us to model time-dependant transition rates for spontaneous HCV clearance and treatment success. The very flexible structure of the model also allowed us to adapt parameters quickly as new data became available.
Implications of Findings
Deferring HCV therapy until advanced liver fibrosis is established may increase the percentage of liver-related complications in people who have multiple risk factors for liver disease progression, such as HIV-coinfected MSM. Our model predicts that the time individuals spend with replicating HCV can be greatly shortened by early treatment. This may decrease further HCV transmissions in those with high-risk behavior. Both findings support arguments that HCV therapy should be accessible to everyone at an early stage. To make this affordable for health insurances and governments, the costs for DAA drugs need to be lowered substantially. Our findings support current recommendations to start HCV treatment irrespective of fibrosis stage in those with risk factors for accelerated fibrosis progression including HIV-coinfected MSM, and in persons at elevated risk of HCV transmission [26].
|
|
| |
| |
|
|
|
