Optimal timing for hepatitis C therapy in US patients eligible for liver transplantation: a cost-effectiveness analysis
B. Njei, T. R. McCarty, B. E. Fortune, J. K. Lim
First published: 19 September 2016
Full publication history DOI: 10.1111/apt.13798
Recurrence of hepatitis C virus (HCV) following liver transplantation (LT) is universal for those with ongoing viraemia and is associated with higher rates of allograft failure and death. However, the optimal timing of HCV treatment for patients awaiting transplant remains unclear.
To evaluate the comparative cost-effectiveness of treating HCV pre-LT vs. post-LT (pre-emptive or after HCV recurrence).
Methods
A Markov state-transition model was created to simulate the progression of a cohort of HCV-genotype 1 or 4 cirrhotic patients from the time of transplant listing until death. We then used this model to study the cost-effectiveness of ledipasvir–sofosbuvir (LDV/SOF) with ribavirin for 12 weeks, administered for three separate treatment strategies: (i) pre-LT; (ii) post-LT preemptively prior to HCV recurrence; or (iii) post-LT after HCV recurrence.
Results
In the base-case analysis using a median model for end-stage liver disease (MELD) score <25 at the time of transplant, we found that pre-LT treatment of HCV led to more QALYs for fewer dollars compared to other strategies. Analysis limited to living donor LT recipients revealed that pre-LT treatment was also the most cost-effective strategy. When the analysis was repeated for MELD ≥25, decompensated disease (Child–Pugh class B or C), and hepatocellular carcinoma cases, preemptive post-LT strategy was more cost-effective.
Conclusions
Treatment of HCV prior to liver transplantation appears to be the most cost-effective strategy for patients with a MELD score <25. For patients with a MELD ≥25 or decompensated cirrhosis, preemptive post-liver transplantation treatment before HCV recurrence is the most cost-effective strategy.
Discussion Only
Full Text \
Alimentary Pharmacology and Therapeutics - September 19, 2016
The results of this cost-effectiveness analysis demonstrated that treatment of HCV prior to LT in patients with a MELD score <25 is the most cost-effective strategy. For living donor LT recipients, HCV treatment prior to LT was also the most cost-effective approach. However, for patients with decompensated cirrhosis (i.e. Child–Pugh class B and C disease), HCC, and MELD score ≥25, a preemptive post-LT HCV treatment strategy was the most cost-effective.
Chronic HCV affects over 174 million individuals worldwide with an estimated number of over 3 million people in the US alone.[
49] Although many people living with chronic HCV may be asymptomatic or unaware of their diagnosis, HCV has become the leading cause for liver cirrhosis and HCC and is the most common indication for LT in the US.[
4, 5] Furthermore, one of the major themes common to the varied HCV treatments available is that all appear to be less effective in patients with progressive fibrosis and worsening underlying liver disease.[
50] Given this significant disease burden and potential cost on the US healthcare economy, the need to treat these individuals prior to progression of their disease remains paramount.
A recent study by Chidi et al. examined the cost-effectiveness of DAA treatment in a Veteran Affairs population.[
51] From this study, it was demonstrated that managing any treatment-naïve, genotype 1 HCV patient with DAA therapy, irrespective of the stage of fibrosis, appeared to be cost-effective. Additional studies evaluating DAA treatment in a non-Veteran population have demonstrated improved patient-reported outcomes as well as favourable short and long-term clinical and health economic results as compared to other regimens.[
8, 52-56] While these findings suggest that all HCV patients should be treated, complex questions remain regarding a preferred algorithm for patients awaiting LT.
Should HCV positive patients defer treatment to preserve eligibility for both HCV negative and HCV positive organs? Patients who have been cured of chronic HCV may lose eligibility for HCV positive organs for potential transplantation. In addition, we acknowledge that in high MELD regions with low transplant volume, this percentage may be higher and represent a more important consideration in the decision to pursue or defer HCV treatment given favourable SVR rates in patients treated post transplant.[
9]
Furthermore, another question remains regarding the treatment of patients in centres or regions that commonly utilise HCV positive donor livers. While wide variation exists in the practice of LT with positive donors, estimates have ranged from 1.68% to 3.45%.[
57, 58] At present, no universal policy is in place with many centres choosing not to adopt this strategy given concerns surrounding progression of fibrosis in immunocompromised hosts and the potential of inducing a new virus to the recipient.[
58] However, regions with organ shortages (i.e. region 1, 5, 7 and 9) may consider donors with extended criteria and increasingly rely upon HCV positive donors.[
58] In such centres, the strategy is to treat these recipients with a post-LT strategy. For these patients, a preemptive post-LT treatment should be considered as compared to waiting for HCV recurrence.
Can HCV eradication in patients with decompensated cirrhosis lead to worse outcomes due to a potential delay to transplantation? Current DAA regimens, such as LDV/SOF, have been associated with a decrease in MELD score for some pre-LT patients achieving SVR and thereby reduce their transplant prioritisation despite ongoing hepatic decompensation, often described as ‘MELD limbo’ or ‘MELD purgatory’.[
9] In high MELD regions, the prospect of interval decreases in MELD score may potentially influence this treatment decision. Policy changes in UNOS organ allocation to stop-gap or freeze MELD scores in context of HCV treatment may mitigate concerns regarding impact of HCV eradication in pre-LT patients.
With regard to HCV positive patients with HCC, subgroup analysis demonstrated preemptive post-LT treatment to be the most cost-effective strategy. Nevertheless, in patients with a MELD score <25 there is the potential of coming off the LT list particularly in well-compensated cirrhotics (i.e. Child–Pugh classification A) with HCC who are successfully ablated and do not desire LT. This strategy may be attractive for these select patients with well-controlled underlying liver disease; however, patients with Child–Pugh classification B or C disease would still likely need LT in the long term regardless of HCC control.
Inherent to cost-effectiveness studies is the realisation that some assumptions must be made to assess clinical utility or effectiveness. This is true in part with regard to cost, intention-to-treat, and the intrinsic heterogeneity of included studies used to formulate decision tree probabilities and modelling. In addition, although mathematical and statistical modelling has been designed to simulate a true clinical setting with regard to this hypothetical cohort, there is no substitute for real-world patient populations, and the true probability of liver disease, fibrosis progression, and response to treatment for individualised clinical practice. This was in part addressed using UNOS data to directly quantify paramount probabilities for the analysis.
Our analyses have several additional limitations. First, the model was restricted to one commonly used treatment regimen: LDV/SOF with ribavirin; future analyses evaluating the cost-effectiveness of alternative regimens are warranted. Second, only patients with genotypes 1 or 4 were included in this model, reflecting the paucity of data addressing the safety and efficacy of all-oral DAA regimens for pre-transplant and post-transplant patients with other genotypes. The emergence of all-oral pangenotypic regimens may permit future analyses across all six genotypes. Third, our model was based on wholesale acquisition cost (WAC) without consideration of payor discounts. Competition across emerging all-oral DAA regimens is expected to prompt further decreases in both WAC and discounted prices over the next several years, and will directly impact estimates of cost-effectiveness in future models. Finally, although statistically significant, we observed small differences in total costs and quality-adjusted life years saved for the various treatment strategies that may be perceived as less clinically relevant. Therefore, each individual patient may possess certain variables outside the realm of this analysis that result in a discrepant physician-drawn conclusion regarding when to initiate HCV therapy. Despite these limitations, this analysis was strengthened by the use of real-world outcomes data from the 2015 UNOS database and available published data regarding treatment outcomes with all-oral DAA regimens in pre- and post-transplant settings.
In summary, interferon-free DAA therapy combinations have revolutionised the ability to treat HCV-associated liver disease. With the impressive SVR rates attained, clinicians are faced with the complex decision regarding when to treat HCV positive patients awaiting LT. Based upon our base-case analysis, pre-LT treatment appears to be the most cost-effective strategy for patients with HCV-cirrhosis and a median MELD score <25. Additional sensitivity analysis revealed that pre-LT treatment was associated with improved liver-related outcomes compared to delaying treatment until post-LT HCV recurrence. Subgroup results of our cost-effective analysis demonstrated that treatment of HCV 3–6 months preemptively during the post-LT course appears to be the most cost-effective strategy for patients with a MELD score ≥25 and those with decompensated disease or HCC. Future prospective studies are needed to validate these findings and adjust for alternative HCV treatment regimens.
