Prospective Multicenter Study of Early Antiviral Therapy in Liver and Kidney Transplant Recipients of HCV-Viremic Donors

Study Oversight

The study protocol was reviewed and approved by the relevant institutional review boards of participating centers. The study was performed in accordance with the Declaration of Helsinki, the principles of Good Clinical Practice, and local regulatory requirements. The trial was designed and conducted by the site-study investigators, with provision of study drug and support for study procedures from Gilead Sciences (Foster City, CA). Trial data management was contracted to United Network of Organ Sharing business services with regular monitoring of trial conduct. An independent data and safety monitor reviewed the progress of the trial. All listed authors had access to the data and assumed responsibility for the integrity and completeness of the reported data from their centers and the fidelity of the trial to the protocol.

Study Design and Participants

This is a prospective multicenter, open-label, proof-of-concept study conducted at six US transplant centers (Supporting Table S1). The study sought to enroll up to 30 study patients undergoing liver and/or kidney transplantation, at least 10 of whom were undergoing liver transplant and at least 10 were undergoing kidney transplantation. Study enrollment was stopped after 24 study patients were enrolled, due to slower than expected enrollment, reflecting the broader use of HCV-viremic donors outside of study protocol. Additionally, it was determined that the study’s main findings would not be expected to change substantially by the enrollment of 6 additional patients.

Eligible patients were adults (age ≥18 years) on the waiting list for kidney or liver (alone or combined with kidney) transplant at one of the study sites. Eligible patients were without current HCV viremia at enrollment, although patients with a prior history of chronic HCV who were previously treated and HCV RNA–negative at least 12 weeks after completing treatment were eligible. As sofosbuvir was not approved in patients with estimated glomerular filtration rate (eGFR) < 30 mL/min at the time of the study start, recipients who were anticipated to have a need for prolonged dialysis following transplant, thereby delaying the start of SOF-VEL, were excluded. Patients treated with recent amiodarone before organ transplant were excluded due to risk of bradyarrhythmia when administered with sofosbuvir. In addition, individuals needing to continue other drugs with potential to interact with SOF-VEL, such as high-dose proton-pump inhibitors, were not eligible (see Supporting Table S2 for all inclusion/exclusion criteria). Finally, as the intent of the protocol was to offer transplant to patients with anticipated long waiting times, all site investigators were encouraged to particularly enroll those patients not expected to get access to a deceased donor during the study period. Recipients with a living donor were excluded. For similar reasons, kidney transplant candidates who were on dialysis more than 5 years at the time of screening were excluded, as well as those with panel-reactive antibody of more than 80%.

Deceased donors were tested for HCV-RNA positivity per protocol by the local organ procurement organizations using a sensitive nucleic-acid test (NAT) with reported qualitative results (positive vs. negative). Additional donor testing confirming the absence of human immunodeficiency virus (HIV) and hepatitis B surface antigen (HBsAg) was required. Finally, kidney donors were required to have a kidney donor profile index (KDPI) of ≤85%. Liver donors were required to have an absence of significant fibrosis (<F2 on scale of 4). These restrictions were used to reduce the use of donor organs of suboptimal quality.

Study Procedures

All participants underwent a two-part informed consent prior to any study procedures. The first part was signed at the time of screening on the wait list, and the second within 24 hours before the transplant event. Review of the potential risks and benefits of participating in this trial included a discussion regarding HCV treatment and complications of unsuccessfully treated HCV following transplant, including graft loss, fibrosing cholestatic hepatitis, cirrhosis, and death. In addition to reconfirming patient consent, study investigators also assessed and confirmed recipient and donor eligibility in the 24 hours before transplant. Transplant-related procedures at each center were performed as local standard of care, including immunosuppression protocols.

Recipient HCV RNA was checked on postoperative days 3 and 7, and every 7 days until detected. Once recipient HCV viremia was detected and study staff deemed the subject clinically stable to start antiviral therapy, SOF-VEL (400 mg/100 mg) was administered once daily. Day 1 was defined as the day that the study drug commenced. The total treatment duration was 12 weeks. If participants missed a dose (or doses) of study drug, they would continue to complete the full 84 days of study drug. For patients who did not achieve viral eradication with this 12-week course of therapy, retreatment with sofosbuvir-velpatasvir-voxilaprevir (400/100/100 mg) once daily for 12 weeks was available.

Assessments

Screening assessments of the patients included demographics, serologies for HCV, hepatitis B virus and HIV, standard laboratory testing, and clinical assessments. Liver transplant candidates had laboratory tests for Model for End-Stage Liver Disease (MELD)-Na, MELD exception score (if any), and Child-Pugh score. For kidney transplant candidates, the panel-reactive antibody result was obtained. Donor variables included age, sex, cause of death, United Network of Organ Sharing (UNOS) region, HBsAg, antibody to hepatitis B core antigen, KDPI, and fibrosis stage (for livers).

For recipients who had quantifiable HCV RNA, SOF-VEL was started when patients were deemed clinically stable to start antiviral therapy and had an eGFR > 30 mL/min. On-treatment HCV-RNA levels were measured at weeks 1 and 4, end of treatment, and weeks 4, 12, and 24 after completing therapy. For recipients who were HCV RNA–negative on day 7 after transplant, HCV RNA was retested weekly until day 28 and then monthly for 2 additional months.

The schedule for measurement of blood cell count, renal function, liver tests, as well as immunosuppression during treatment and follow-up are provided in Supporting Table S3.

Outcomes

The primary outcome was sustained virologic response, defined as HCV RNA below the lower limit of quantification 12 weeks after treatment completion (SVR12). Secondary outcomes included the proportion of patients with SVR24, defined as HCV RNA < lower limit of quantification (LLOQ) 24 weeks after the end of treatment. Viral relapse was defined as HCV RNA < LLOQ at end of treatment with subsequent quantifiable HCV RNA and on-treatment virologic breakthrough was defined as >1-log increase in viral RNA after treatment week 1.

Safety was measured as the adverse events (AEs) and serious adverse events (SAEs) attributed by the investigator to HCV infection or antiviral therapy; the proportion of recipients who prematurely discontinued antiviral therapy before the planned end of treatment; and patient and graft survival at 6 months following transplant. An exploratory endpoint was the change in wait time for kidney and/or liver transplant based on comparison with UNOS data for the participating regions.

AEs were graded using the Common Terminology Criteria for Adverse Events, Version 4.03, published June 2010. The investigator was responsible for assessing AEs and SAEs for causality and severity. Treatment-emergent AEs (TEAEs) were assessed for relationship to study drug and severity. TEAEs leading to premature discontinuation of the study drug were captured. AEs were recorded for all patients who received a NAT-positive HCV organ within 4 weeks after the last dose of study drug. SAEs were reported to the lead investigators and the independent data and safety monitor within 24 hours of recognition by study staff. SAEs were collected for the entire duration of the study. After the 10th study patient had reached treatment week 4, the lead investigators and the independent data and safety monitor conducted a safety and efficacy review of all available data. When these same patients reached follow-up week 12 (corresponding to 24 weeks after starting the study drug), the lead investigators and the independent data and safety monitor conducted another safety and efficacy review.

Statistical Analysis

No formal hypothesis testing was performed, as this is a pilot study. No inferential statistics or statistical comparisons were planned for efficacy endpoints. All participants who underwent liver and/or kidney transplant with an HCV NAT-positive organ were included in the analysis. If a quantitative PCR was missing, the value was imputed as LLOQ if the viral load before and after were both LLOQ. We assumed that none of the transplant recipients would achieve SVR12 without any antiviral therapy. The existing data for SOF/VEL have demonstrated an SVR rate above 94% in patients without cirrhosis. Along with the percentage of patients with SVR12, a two-sided 95% confidence interval was constructed using the Clopper-Pearson method. With a sample size of 24 patients, if the SVR12 rate was 95%, the lower and upper bound of the 95% CI would be 82% and 100%, respectively.

Treatment-emergent data from the first dose of study drug through the date of the last dose of study drug plus 30 days were included. TEAEs were summarized by relationship to study drug and severity. In addition, TEAEs leading to premature discontinuation of the study drug were summarized.

HCV-RNA kinetics were described as the rate of HCV-RNA log₁₀ change per week (1) from treatment start to 1 week on-treatment and (2) from 1 week to 4 weeks on-treatment. These intervals were selected based on the biphasic decline in viral load and the largely undetectable viral load by week 4 for liver recipients, effectively resulting in a slope of zero after 4 weeks. The distribution of the rate of HCV-RNA log₁₀ was reasonably normal. Therefore, HCV-RNA kinetics were evaluated as means (SDs) and compared by organ using linear regression. To account for differences in HCV RNA at the start of the evaluation period (treatment start or week 1), the comparison was adjusted for HCV RNA at the corresponding start time. For comparison of proportions achieving undetectable HCV RNA at weeks 1 and 4 of treatment, missing values were assumed to be HCV RNA–detectable. Analyses were completed using SAS version 9.4 (SAS Institute Inc., Cary, NC).

Disposition

From July 13, 2018, to December 21, 2019, 122 patients provided the first part of the informed consent and were enrolled in this trial, and 24 (20%) of these received organs from HCV-viremic donors (Fig. 1): 13 liver transplants and 11 kidney transplants. Of the remaining eligible patients, 47 (39%) received organs from HCV-negative donors. For potential liver recipients, 13 of 74 (18%) were transplanted with organs from HCV-viremic donors, and 36 of 74 (49%) were transplanted with organs from HCV-negative donors. Two potential liver recipients died on the wait list. The median (interquartile range [IQR]) time from activation to receiving an HCV-viremic liver was 116 (56, 169) days.

For potential kidney recipients, 11 of 40 (28%) were transplanted with organs from HCV-viremic donors, and 7 of 40 (18%) were transplanted with organs from HCV-negative donors. The median (IQR) time from activation to being able to receive an HCV-viremic kidney to transplant with an HCV-viremic kidney was 48 (34, 124) days. None of the listed liver-kidney candidates received HCV-viremic donors.

Demographics and Transplant Data

Baseline characteristics of the liver and kidney recipients of HCV-viremic organs are found in Table 1. All of the donors were located in the same UNOS regions as their recipient. The etiologies of end-stage liver disease were nonalcohol steatohepatitis (n = 5, 38%), alcohol (n = 4, 31%), HCV/alcohol (n = 2, 15%), alpha-1-anti-trypsin deficiency (n = 1, 8%), and hemochromatosis (n = 1, 8%). The median (IQR) MELD-Na score for all liver recipients was 19 (14, 22), and in 7 patients with MELD exception, the score was 22 (16.5, 24.5). The median (IQR) Child-Pugh score was 9 (7, 10). The etiologies of end-stage renal disease for the kidney recipients were hypertension (n = 4, 36%) with diabetes (n = 1, 9%), polycystic disease (n = 2, 18%), chronic nephritis (n = 2, 18%), and immunoglobulin A nephropathy (n = 2, 18%). The median (IQR) index hospitalization length of stay for the liver recipients was 10 (7, 12) days, and for kidney recipients was 4 (4, 6) days.

Efficacy

Recipient serum HCV RNA was quantifiable on day 3 after transplant in all 13 liver recipients and in 9 of 10 (82%) kidney recipients, with 1 kidney recipient with HCV RNA not tested until day 7 after transplant with quantifiable results at that time point. In another kidney recipient, HCV RNA remained negative at all time points following transplant; this recipient did not require antiviral treatment.

The median (IQR) time from transplant to start of antiviral therapy was 7.0 (6.0, 12.0) days for liver transplant recipients and 16.5 (9.8, 24.5) days for kidney transplant recipients (Fig. 2). More specifically, 7 of 13 (54%) liver recipients and only 1 of 10 (10%) kidney recipients started antiviral treatment in the first week. Conversely, only 1 of the 13 (8%) liver recipients and 3 of the 10 (30%) kidney recipients started antiviral treatment in the fourth week after transplant. Reasons for different starting times of antiviral therapy beyond the first week are included in Table 2. In 1 of 13 liver recipients and in 5 of 10 kidney recipients, the start of antiviral therapy was delayed beyond the first week by eGFR < 30 mL/min.

The median (IQR) HCV-RNA level 3 days after transplant was 6.5 (3.9, 7.1) log₁₀ IU/mL in liver recipients versus 3.6 (2.9, 4.0) log₁₀ IU/mL in kidney recipients (Fig. 3). The median (IQR) HCV RNA at the start of antiviral therapy was 6.6 (5.7, 7.2) in kidney recipients and 5.8 (5.3, 6.9) log₁₀ IU/mL in liver recipients (Table 3). The median (IQR) HCV RNA was higher in the kidney recipients at day 7 after starting treatment, 3.7 (2.1, 4.) versus 1.9 (1.5, 2.4) log₁₀ IU/mL in liver recipients. By week 4 of treatment, 10 of 13 (77%) liver recipients, but only 2 of 10 (20%) kidney recipients, had undetectable virus (P = 0.01). All of the liver recipients had undetectable virus at the end of treatment, while 3 (30%) of the kidney recipients still had detectable, but not quantifiable, virus at the end of the treatment period. Still, all 23 treated recipients achieved SVR12. All 22 subjects who completed 24 weeks of follow-up after treatment had SVR24. One subject died of complications of graft-versus-host disease (GVHD) prior to SVR24. Adjusted for baseline HCV-RNA levels, the mean change in HCV RNA from day 1 to week 1 was greater in liver than kidney recipients (−4.03 vs. −3.07, P = 0.008), but not statistically different from week 1 to week 4 (−0.85 vs. −1.74, P = 0.32) (Table 3). Individual viral kinetics for liver and kidney transplant recipients are shown in Supporting Fig. S1.

Safety

AEs and SAEs possibly or probably related to study participation are given in Table 4. One liver recipient required intraoperative amiodarone. In this case, due to the known drug interaction between amiodarone and sofosbuvir, the study drug was not started until 25 days after transplant and with telemetry monitoring for 72 hours. No bradyarrhythmias or other cardiac events occurred. SAEs were reported in 9 of 13 (69%) liver recipients and in 4 of 11 (36%) kidney recipients (Table 4).

SAEs deemed potentially related to receipt of an HCV-viremic donor or HCV therapy occurred in 4 (31%) of the liver recipients. One subject who received an A2→O graft and began the study drug 5 days after transplant developed early antibody-mediated rejection that was treated with intravenous immunoglobulin G with normalization of liver biochemistries by day 36 after transplant. Another subject who began the study drug 6 days after surgery developed a progressive, intrahepatic biliary sclerosis starting by day 31 after an uneventful transplant from a 36-year-old donor after brain death. An ischemic etiology was excluded. Other than abnormal-appearing bile ducts on cholangiography and an alkaline phosphatase 3 times the upper limit of normal, there have been no other clinical sequelae 16 months after transplant. A third subject was diagnosed with GVHD beginning day 19 following transplant. The recipient was a 62-year-old male with blood type O, and the donor was male. Study drug was started 8 days after transplant but was permanently discontinued prematurely after only 7 weeks of treatment per investigator discretion due to multisystem organ complications. Still, the subject had a SVR. However, despite treatment for GVHD with corticosteroids and ruxolitinib, the subject succumbed to multisystem organ failure 152 days after transplant. The last subject who started study treatment 6 days after surgery developed an idiopathic pericardial effusion with decreased left-ventricular ejection fraction about 1 month after transplant that resolved with pericardiocentesis.

Two (20%) of the kidney recipients had delayed graft function, requiring transient hemodialysis. Renal function before, during, and after SOF-VEL therapy are shown in Supporting Fig. S2. None of the kidney recipients had acute cellular rejection or clinical evidence of glomerulopathy following transplant. Three kidney recipients with normal serum liver tests at the time of transplant developed serum transaminase elevations 4-7 times the upper limit of normal before HCV treatment was initiated that persisted throughout the SOF/VEL treatment and into the posttreatment period (Fig. 4). One patient, without known risk factors, had biopsy-proven nonalcoholic steatohepatitis on liver biopsy performed after the SVR12 visit. In another patient, the abnormal liver tests were attributed to a statin. One kidney recipient had normal alanine aminotransferase (ALT) at the time of transplant and during the first 6 weeks of antiviral treatment, but then developed mild elevation in ALT, which peaked at two times normal at the end of treatment, with normalization by the SVR12 visit. None of these 4 kidney recipients with abnormal ALT during treatment had undetectable HCV at week 4 of treatment, although all had SVR12.