HCV-infected Deceased Donor Kidney Transplantation – Time to Take Up the Offer

Kidney transplantation (KT) offers the best quality of life and survival for patients with end stage kidney disease (ESKD). While the number of KTs performed annually has only grown modestly, the prevalence of ESKD has risen substantially, resulting in a proportional increase in organ shortage.¹ Consequently, over 95,000 patients are waiting for KT, and wait times exceed 5 years for many patients in the United States.^1,2 With 5-year mortality of 42–52% on dialysis, many die due to the shortage of available kidneys.¹ High rates of discard of kidneys acquired for transplantation contribute to organ shortage. In 2017, 18.9% of kidneys were discarded, amounting to approximately 4,770 potential missed opportunities for KT.² With the movement to increase KT under the Advancing American Kidney Health Initiative, identifying ways to reduce organ discard could improve clinical outcomes for patients with ESKD.

Hepatitis C virus (HCV)-infected kidneys have historically been transplanted exclusively into HCV-infected recipients. With limited HCV-infected recipients, these organs have been discarded at higher rates than HCV-uninfected donors.³ Because of the devastating opioid epidemic, there has been a 227% increase in deceased donors from drug overdose, with 19% having HCV.⁴ With the advent of direct acting antivirals (DAA) there is growing interest in transplanting HCV-infected kidneys into HCV-uninfected recipients.⁵ Prospective single-center trials and retrospective studies have demonstrated excellent short-term clinical outcomes of transplanting HCV-infected kidneys into HCV-uninfected recipients followed by DAA treatment to eradicate HCV.^6–8 These studies showed near universal HCV transmission with KT, but nearly 100% cure rates of HCV infection with DAAs. Because of these favorable reports, more transplant centers are offering HCV-infected kidneys to uninfected recipients.⁷ Before implementing this strategy nationally, it is important to understand its impact on clinical outcomes and its cost-effectiveness, particularly given the current high costs of DAAs. In this issue of AJKD, Eckman et al. project these outcomes through a Markov state transition simulation model (illustrated in Figure 1).⁹

Figure 1:

Simplified schematic of the simulation model of Eckman et al. of kidney transplantation using HCV-infected and HCV-uninfected donor kidneys

*At each transition point in the model, patients are at risk of entering the absorbing health state of death.

**Failed KT is represented in the Eckman et al. model by the 30-day mortality following KT.

Abbreviations: DAA – Direct Acting Antiviral; HCC – Hepatocellular Carcinoma; HCV - Hepatitis C Virus; HD - Hemodialysis; KT- Kidney Transplantation.

In the model, the authors simulated a population of HCV-uninfected maintenance hemodialysis patients. They compared two KT strategies for an HCV-uninfected recipient: 1) an HCV-uninfected donor kidney; and 2) an HCV-infected donor kidney with a 12-week course of glecaprevir/pibrentasvir (G/P) starting just prior to KT. Within the model, they defined health states that incorporated life on dialysis, life after KT (with or without HCV), liver disease, and death. They assumed that patients who failed to respond to G/P were at risk of progressing through the stages of liver disease (fibrosis, cirrhosis, hepatocellular carcinoma and liver transplantation). For input parameters, they incorporated data from the United States Renal Data System, the Scientific Registry of Transplant Recipients (SRTR) and literature review.^1,2,7,8 Parameters included time-dependent mortality rates while on dialysis or after transplantation, transplantation rates, treatment efficacy of G/P, probability of hepatic complications, and quality of life and costs in each health state. They assumed an average wait time for HCV-uninfected and HCV-infected donors of 4 years and 1.56 years, respectively, and a monthly cost of G/P of $9,830.¹⁰ The model generated lifetime outcomes in terms of quality-adjusted life years (QALYs) and costs associated with each strategy. The authors calculated the incremental cost-effectiveness ratio (ICER, the difference in discounted costs divided by the difference in discounted QALYs) of the HCV-infected KT strategy versus the HCV-uninfected KT strategy. The HCV-infected KT strategy was considered cost-effective if its ICER was below $50,000/QALY (i.e., higher costs and more QALYs in an economically efficient manner), and it was considered cost-saving if it resulted in lower costs and more QALYs.

Their model demonstrated that the transplantation of HCV-infected kidneys into HCV-uninfected recipients would increase QALYs by 0.91 and decrease costs by $37,918 compared to the strategy of transplanting only HCV-uninfected kidneys into HCV-uninfected recipients. To account for uncertainty, they performed sensitivity analyses. By varying key input parameters in deterministic sensitivity analysis, they found that 1) the HCV-infected donor strategy was cost-saving or cost-effective as long as the wait time advantage (the difference in wait time between receiving an HCV-uninfected kidney and an HCV-infected kidney) was at least 0.9 years (~11 months) and 2) if G/P costs decreased to $7,000 per month, average lifetime cost savings would increase to $44,671 in the HCV-infected KT strategy. Through probabilistic sensitivity analysis, in which multiple key input parameters were varied simultaneously across a range of plausible values, the HCV-infected donor strategy was cost-effective over 99.9% of the time.

A model-based analysis is a simplification of a complex disease; it depends on model parameters and structure. Kadatz et al. previously reported a model of an HCV-infected KT strategy compared to an HCV-uninfected KT strategy for HCV-uninfected recipients and showed cost-savings when the wait time advantage was at least 2 years.¹¹ Eckman et al. used the most recent SRTR data and the lower costs of G/P compared to other DAAs, as well as the prior time on dialysis for patients who received HCV-infected kidneys as a proxy for the wait time for HCV-uninfected recipients willing to accept HCV-infected kidneys. While actual wait times will change as HCV-infected KT expands, the sensitivity analysis by Eckman et al. projects cost-effectiveness when the wait time advantage is at least 0.9 years.

The results of this study should be interpreted in the context of three important clinical considerations. First, current guidelines from the American Association for the Study of Liver Diseases recommend that treatment-naive, non-cirrhotic patients receive only eight weeks of G/P. In clinical trials of patients undergoing HCV-infected KT ( NCT03781726 and NCT03627299), eight-week and four-week courses of G/P are being studied. If shorter courses of G/P provide similar clinical outcomes, it will decrease costs and increase the attractiveness of an HCV-infected donor strategy. Second, in the model, patients who failed G/P were not retreated. In clinical practice, most patients would be retreated with a second-line, more expensive, DAA regimen, and could expect another >95% chance of achieving cure.¹² As the model does not account for second-line DAAs, it cannot estimate the clinical and economic outcomes of this practice. Third, the authors modeled a strategy of “pre-emptive” G/P, begun shortly before the transplant procedure. Compared to clinical trials with pre-approved DAA, retrospective studies have reported delays in the start of DAA therapy for weeks to months due to delays in insurance approval. These studies reported higher rates of complications, including cytomegalovirus infection, polyoma virus infection, and fibrosing cholestatic hepatitis.^13,14 Thus, unless DAAs receive streamlined insurance approval, the model will underestimate complication rates.

Eckman and colleagues’ study is an important advance to the field with a key conclusion: transplanting HCV-infected kidneys into HCV-uninfected recipients is a cost-effective (and often cost-saving) strategy that improves clinical outcomes under the model assumptions, even if the difference in wait times between the two strategies is reduced to ~11 months. While larger trials with longer follow-up are needed to fully understand the spectrum of possible complications, currently, over 80% of HCV-uninfected patients waiting for KT would consider receiving an HCV-infected kidney depending on factors such as DAA efficacy, wait time, and quality of the organ.¹⁵ With this widespread interest and existing excellent 1-year clinical outcomes reported, we should utilize shared decision-making with patients, involving them in a full and upfront discussion of the potential benefits and risks.⁷ It is time to implement the HCV-infected donor strategy more widely, which will require guideline support and streamlined approval of DAAs by insurers to avoid delays in therapy for patients.