Effectiveness of Elbasvir/Grazoprevir in Patients with Hepatitis C Virus Genotype 1 Infection and Chronic Kidney Disease in the United States Veterans Population

Abstract

Background & Aims:

Randomized controlled trials of EBR/GZR have reported high treatment efficacy, safety and tolerability in patients undergoing dialysis. However, real world effectiveness data for EBR/GZR in this population is lacking. We evaluated the effectiveness of EBR/GZR in an HCV-infected population with all stages of CKD including dialysis compared with control patients with estimated glomerular filtration rate (eGFR) ≥60 in the US Department of Veterans Affairs (VA).

Methods:

We conducted a retrospective cohort study of patients with chronic HCV genotype 1 infection with EBR/GZR prescriptions dispensed during February 1, 2016-August 31, 2017 in 128 VA Medical Centers. We collected patient information regarding history of dialysis, end stage renal disease (ESRD), and/or eGFR values. We measured SVR based on undetectable HCV RNA at least 4 weeks after the completion of treatment. We examined SVR rates by CKD stage compared to control patients and within patient subgroups using logistic regression models.

Results:

We identified 5,961 patients (42.5% genotype 1a, 55.0% genotype 1b) who met eligibility criteria and completed a EBR/GZR treatment course (≥11 weeks). Approximately 73.2% (n=4,361) had eGFR ≥60 who served as control patients, 14.4% (n=860) had Stage 3 CKD, and 12.4% (n=740) had Stage 4-5 CKD or ESRD. Of patients with Stage 4-5 CKD/ESRD, 76.1% underwent dialysis (n=563). The overall SVR was 96.7% in all patients, 96.4% for eGFR≥60, 98.3% in Stage 3 CKD, and 96.5% in Stage 4-5 CKD/ESRD. No statistically significant differences were found in the SVR rates in patients with or without dialysis in the Stage 4-5 CKD/ESRD patients (adjusted OR 0.91; 95% CI 0.56-1.47 and OR 1.74; 95% CI 0.63-4.81) compared with those with eGFR≥60.

Conclusion:

We found EBR/GZR was effective in patients with HCV GT1 infection regardless of CKD severity or receipt of dialysis in the US VA population.

INTRODUCTION

Chronic hepatitis C virus (HCV) infection affects an estimated 71 million persons worldwide including 2.7-3.9 million in the United States.¹ In the U.S., more than 19,000 deaths were associated with HCV annually from 2013 to 2015, mainly due to progressive liver fibrosis, cirrhosis, liver failure and hepatocellular carcinoma.^1–4 Compared to the general population (prevalence of 1.8%), veterans are three times more likely (5.4%) to be infected with HCV.^4,5 The risk of having HCV is five times higher in veterans born between 1945 and 1965 compared to the general population.⁵ HCV infection is also widely prevalent and especially complex in patients with end-stage renal disease (ESRD) receiving dialysis since HCV infection can accelerate deterioration in kidney function.^6,7 Until recently patients who receive dialysis are rarely treated due to few treatment options available and have a higher risk of hospitalization, anemic complications, and mortality compared to HCV patients without ESRD.⁷

Elbasvir/grazoprevir (EBR/GZR) combination therapy is a once daily, all-oral, direct-acting antiviral agent (DAA) that was approved in 2016 by the U.S. Food and Drug Administration (FDA) for the treatment of chronic HCV genotype (GT) 1 infection.^8,9 Elbasvir is a HCV NS5A inhibitor and grazoprevir is a HCV NS3/4A protease inhibitor.^2,8,10,11 EBR/GZR has been recommended to treat HCV among patients with severe renal impairment (eGFR<30).^10,11 Randomized controlled trials of EBR/GZR have reported high treatment efficacy, safety and tolerability in patients with advanced chronic kidney disease (CKD) or ESRD, including those undergoing dialysis.^{2,6,9,12–14} EBR/GZR is not cleared by dialysis, as it is hepatically metabolized and highly protein-bound.¹⁵ Low rates of adverse events were reported in clinical trials in patients with HCV genotype 1 infection and CKD.⁶ Clinical trial of EBR/GZR among patients with HCV GT 1 reported low adverse-event rates (2.8% for immediate-treatment group, 11.5% for placebo followed by active treatment group) and SVR rates of 94%.¹⁶ The real world effectiveness data for EBR/GZR in the CKD population with and without dialysis is limited in small sample size.^12,17

The aim of this study was to evaluate the effectiveness of EBR/GZR in a real-world HCV-infected population with CKD across all stages including those with ESRD who are receiving dialysis in the US Department of Veterans Affairs (VA).

MATERIALS AND METHODS

1.1. Data Source

We conducted a retrospective cohort study using electronic administrative and clinical records obtained from the Department of VA Corporate Data Warehouse (CDW). The VA is the largest integrated health system in the U.S. with comprehensive administrative patient claims data files of over 9 million eligible Veterans.^18,19 The CDW was developed in 2006 as a repository for patient-level data across the national health system and includes data from pharmacy, laboratory, and inpatient and outpatient encounters including International Classification of Diseases, Ninth/Tenth Revision, Clinical Modification ICD-9/10 CM diagnosis codes.²⁰

We captured dialysis information from VA and non-VA facilities. To account for dialysis outside of VA facilities, we utilized the Fee Basis files, which captures payment and justification (medical care including dialysis, pharmacy services, and travel costs) to non-VA providers.²¹ In addition, VA facilities may contract with outside providers to provide dialysis within VA facilities, which is captured in the Veterans Information Systems and Technology Architecture (VISTA) files and directed to VA utilization databases of CDW.²¹

1.2. Study Population and Inclusion/Exclusion Criteria

We identified all patients aged 18 years or older with chronic HCV genotype 1 infection undergoing EBR/GZR treatment defined by ≥1 prescription dispensed from February 1, 2016 to August 31, 2017 from all 128 VA Medical Centers nationwide. Inclusion criteria for this analysis consisted of at least one positive HCV RNA test, at least one inpatient or outpatient visit in the VA within one year prior to treatment initiation, at least 11 weeks of EBR/GZR treatment, and an available HCV RNA test at least 4 weeks after end of treatment to determine SVR. Index date was defined as the release date for the first prescription of EBR/GZR. Patients were also required to have at least 2 eGFR values ≥3 months apart within two years prior to the index date or a code for dialysis or ESRD as defined below.

We excluded patients if they had an acute kidney injury (AKI) within one year prior to HCV treatment initiation. AKI was defined by ICD-9 CM codes 584.X, which has >90% sensitivity and negative predictive value for AKI.²² Patients who had undergone kidney transplantation any time prior to treatment initiation were also excluded from the study as well as patients who had previously received an NS5A-inhibitor containing treatment regimen or undetermined EBR/GZR regimens. Finally, patients with indeterminable CKD as defined below were excluded. See Figure 1 for study flow diagram.

Figure 1.

Sample selection flow chart

1.3. Chronic Kidney Disease (CKD) Severity Groups

To identify CKD and its severity, we used patient information regarding history of dialysis, ESRD, and/or estimated glomerular filtration rate (eGFR). We defined dialysis by at least one inpatient or outpatient procedure or diagnosis code in the one year prior to the index date, and ESRD by ICD code any time prior to index date. eGFR was ascertained from automated reporting of VA laboratory tests that used the Modification of Diet in Renal Disease (MDRD) equation.²³

We examined 2 eGFR test results ≥3 months apart but within two years prior to the index date. We classified patients into the following groups based on their eGFR, diagnosis of ESRD, and use of dialysis. The second and third groups were based on CKD stage criteria by the National Kidney Foundation²⁴: 1) Controls: eGFR ≥60mL/min/1.73m²; 2) Stage 3 CKD: eGFR of 30-59 without prior medical claims for dialysis in year prior or ESRD diagnosis ever; or 3) Stages 4-5 CKD/ESRD: eGFR<30 or presence of dialysis in year prior or ESRD diagnosis ever. If two criteria classified a patient in discrepant stages, the patient was classified using the stage closest to the treatment date.²⁴ Among Stages 4-5 CKD or ESRD, we stratified patients into those with or without dialysis. (see Supplemental Table 1 for codes used for inclusion/exclusion criteria).^21,22

1.4. Patient Characteristics

We examined baseline HCV viral load, genotype and history of any previous treatment for HCV. Previous treatment included history of interferon, boceprevir, telaprevir, sofosbuvir, and/or simeprevir. Given the patient sample size for these regimens were too small after stratifying by CKD stages, we combined EBR/GZR with RBV and EBR/GZR + sofosbuvir (SOF) ± RBV into other regimen for analysis. We also collected information on patient demographics, including age, sex, and race/ethnicity (White non-Hispanic, White Hispanic, Black and Other) and BMI. Cirrhosis and associated complications (e.g. ascites, hepatic encephalopathy, varices) and other comorbidities including diabetes, history of alcohol and/or drug abuse, and HIV were identified by the presence of at least one ICD-9/10 CM code any time before the index date.^25,26

We measured SVR based on undetectable HCV RNA at least 12 weeks after the completion of treatment. In the absence of HCV RNA ≥ 12 weeks after treatment, SVR was defined based on HCV RNA testing available between 4 to 12 weeks after treatment completion.^27,28 Data to determine SVR12 was available in 90.6% of the study population. Treatment completion date was determined as the last day covered by the medication dispensed (i.e., total number of day supply). We calculated duration of treatment by adding prescriptions until there was a break of at least 45 days without EBR/GZR. We considered patients to have completed 12 weeks of EBR/GZR if they had received between 77 to 91 days of medication, and 16 weeks of EBR/GZR if they had received 105 to 109 days of medication.

1.5. Statistical Analysis

We used a Chi-Square test to generate p-values to determine if the CKD groups were significantly different than the controls for the demographic and clinical characteristics. We then used multivariate logistic regression, where the primary efficacy outcome variable was SVR and the main independent variable of interest was CKD severity groups (3 groups: 1) Controls with eGFR ≥60; 2) Stage 3 CKD; and 3) Stage 4-5 CKD/ESRD with or without dialysis), to determine differences in SVR while adjusting for patient demographic, liver performance, comorbidities, baseline viral load and HCV treatment history. This model is shown in Table 3: Model A. To assess the effectiveness of EBR/GZR in patients with advanced CKD with and without dialysis, we further examined patients in a sensitivity analysis who were classified as Stages 4-5 CKD with dialysis compared to patients with Stages 4-5 CKD without dialysis (4 groups: 1) Controls with eGFR ≥60; 2) Stage 3 CKD; 3) Stage 4-5 CKD/ESRD with dialysis; and 4) Stage 4-5 CKD/ESRD without dialysis). This model is shown in Table 3: Model B. Likelihood ratio tests were used to determine the fit of our model and multicollinearity was tested between pairs of coefficients to identify any collinearity issues using variance inflation factor scores.

Table 3.

Multivariate logistic regression results for predictors of SVR in HCV GT1 patients who received EBR/GZR in the VA (n=5,961)

Variable	Model A (combining Stage 4-5 CKD/ESRD)			Model B [Stage 4-5 CKD/ESRD group was classified as: a) with dialysis and b) without dialysis]
	Adjusted OR	95% CI	P-Value	Adjusted OR	95% CI	P-Value
Age (ref: 25-65 years)
>65 years	1.01	0.74-1.37	0.96	1.01	0.74-1.38	0.95
Race/ethnicity (ref: Non-black)
Black	0.96	0.72-1.29	0.80	0.97	0.72-1.30	0.82
Genotype (ref: 1a)
1b	1.51	1.11-2.06	0.20	1.51	1.11-2.06	0.20
Cirrhosis	0.76	0.56-1.03	0.08	0.77	0.56-1.05	0.09
BMI (ref: <30 kg/m2)
≥30 kg/m2	0.82	0.61-1.11	0.44	0.82	0.60-1.11	0.42
Baseline HCV RNA (ref: <800,000 IU/mL)
≥800,000 IU/mL	0.39	0.52-0.60	0.01	0.39	0.25-0.59	0.005
Any previous treatment	0.80	0.54-1.20	0.28	0.80	0.54-1.19	0.27
Regimen (ref: Other)
EBR/GZR without RBV 12 weeks	3.04	2.17-4.25	<0.001	3.03	2.17-4.24	<0.001
CKD Status (ref: eGFR ≥60)
Stage 3 CKD	2.23	1.30-3.84	0.01	2.23	1.29-3.84	0.04
Stage 4-5 CKD/ESRD (all)	1.04	0.67-1.63	0.15	N/A	N/A	N/A
Stage 4-5 CKD/ESRD with dialysis	N/A	N/A	N/A	0.91	0.56-1.47	0.06
Stage 4-5 CKD/ESRD without dialysis	N/A	N/A	N/A	1.74	0.63-4.81	0.54

All analyses were conducted using SAS 9.4 (SAS Institute Inc., Cary, NC). The study protocol was approved by the Baylor College of Medicine Institutional Review Board and the Michael E. DeBakey VA Research & Development Committee.

RESULTS

2.1. Patient Characteristics

We identified 5,961 patients with HCV GT1 that met all inclusion criteria (Figure 1) and completed their EBR/GZR treatment course (80.5% completed at least 11 weeks of treatment; n=6,079/7,550). The mean age was 63.5 years (S.D. = 6.2 years) and approximately 97.0% (n=5,780) were male. More than half of patients (56.3%) were Black, followed by patients who were White Non-Hispanic (34.1%), White Hispanic (3.5%), and other race/ethnicity (1.9%).

Of 5,961 patients in the study population, 73.2% (n=4,361) had eGFR ≥60, followed by 14.4% (n=860) of patients with Stage 3 CKD, and 12.4% (n=740) of patients who had Stages 4-5 CKD and ESRD receiving dialysis. Of patients with Stages 4-5 CKD, 76.1% underwent dialysis in the prior year (n=563). More than half of patients had genotype 1b infection (55.0%; n=3,276) and 42.5% had genotype 1a infection (n=2,533). Approximately 68.4% (n=4,077) of patients had a baseline HCV RNA of ≥800,000 IU/mL. Most patients were treatment naïve (88.7%; n=5,286), while 11.3% of patients had a history of previous treatment (n=675). In terms of treatment, 87.1% of the patients were treated with EBR/GZR without RBV for 12 weeks, followed by EBR/GZR + RBV 12 weeks (3.2%), and EBR/GZR + RBV for 16 weeks (2%). Others included EBR/GZR +Sofosbuvir +/−RBV and other off-label regimens.

Overall, more than a fourth of patients had cirrhosis (28.0%; n=1,667), 38.1% had diabetes (n=2,270), almost three fifths had a history of alcohol abuse (57.0%; n=3,399) and half had a history of drug abuse (49.6%; n=2,957). About 2.1% (n=125) of patients had HCV/HIV coinfection. Patient characteristics are described in further detail in Table 1.

Table 1.

Characteristics of HCV GT1 infected patients treated with EBR/GZR by CKD status (n=5,961)

Variable	Controls with eGFR ≥60 (%) (n=4,361)	Stage 3 CKD (%) (n=860)	Stage 4-5 CKD/ESRD (%) (n=740)
Age (years)	-	-	-
18-65	2,973 (68.2)	418 (48.6)**	431 (58.3)**
>65	1,388 (31.8)	442 (51.4)	309 (41.8)
Sex	-	-	-
Male	4,222 (96.8)	831 (96.6)	727 (98.2)*
Female	139 (3.2)	29 (3.4)	13 (1.8)
Race/ethnicity	-	-	-
White Non-Hispanic	1,659 (38.0)	263 (30.6)**	109 (14.7)**
White Hispanic	163 (3.7)	18 (2.1)	27 (3.7)
Black	2,257 (51.8)	527 (61.3)	569 (76.9)
Other	84 (1.9)	19 (2.2)	9 (1.2)
HCV Genotype	-	-	-
1a	1,678 (38.5)	391 (45.5)**	464 (62.7)**
1b	2,593 (59.5)	438 (50.9)	245 (33.1)
BMI (kg/m2)	-	-	-
<25	1,377 (31.6)	270 (31.4)*	289 (39.1)**
25-<30	1,655 (38.0)	290 (33.7)	250 (33.8)
≥30	1,294 (29.7)	297 (34.5)	192 (26.0)
% baseline HCV RNA	-	-	-
<800,000 IU/mL	1,099 (25.2)	233 (27.1)	295 (39.9)**
≥800,000 IU/mL	3,061 (70.2)	600 (69.8)	416 (56.2)
Previous HCV treatment	-	-	-
Naïve	3,856 (88.4)	778 (90.5)	652 (88.1)
Experienced	505 (11.6)	82 (9.5)	88 (11.9)
Cirrhosis	1,103 (25.3)	253 (29.4)*	311 (42.0)**
History of alcohol abuse	2,549 (58.5)	488 (56.7)	362 (48.9)**
History of drug abuse	2,166 (49.7)	440 (51.2)	351 (47.4)
HIV coinfected	54 (1.2)	22 (2.6)*	49 (6.6)**
Diabetes	1,384 (31.7)	398 (46.3)**	488 (66.0)**
Dialysis	-	-	-
Yes	0 (0.0)	0 (0.0)	563 (76.1)**
No	4,361 (100.0)	860(100.0)	177 (23.9)
Regimen	-	-	-
EBR/GZR without RBV 12 weeks	3,904 (89.5)	770 (89.5)	663 (89.6)
Other	457 (10.5)	90 (10.5)	77 (10.4)

Note:

^*p<0.05 ,

^**p<0.001 for comparisons of Stage 3 CKD with controls and Stage 4-5 CKD with controls

2.2. Overall SVR

The overall SVR rate in all patients was high at 96.7% (n=5,762/5,961; 95% CI 96.2-97.1). SVR was high regardless of patient subgroups (Figure 2) including both male and female sex, patients in any age group (18 years and over) and patients who were Black and White Non-Hispanic. Patients who were White Hispanic had a slightly lower SVR rate of about 95%.

Figure 2.

SVR and 95% confidence interval overall and by subgroup in GT1 HCV patients treated with EBR/GZR (n=5,961)

SVR rates in the genotype 1b subtype was slightly higher in comparison to genotype 1a, 97.5% versus 95.5% respectively. In treatment naïve patients, the SVR rate was 96.8% (n=5,224/5,961) and in treatment-experienced patients, SVR was achieved by 95.0% (n=650/5,961) of patients.

Similarly, the SVR rate was >96% across comorbidities categories including alcohol and drug abuse, HCV/HIV coinfection and diabetes. For patients with a BMI≥25kg/m², the SVR rate was about 96.5% compared to 97.4% in patients with a BMI<25kg/m². Patients with a baseline HCV RNA ≥800,000 IU/mL had a lower SVR rate of 95.9% compared to patients with a baseline HCV RNA <800,000 IU/mL of almost 98.5%. The SVR rate for patients with cirrhosis and diabetes was also high at 96.2%, and roughly 97% for patients with a history of alcohol or drug abuse, and HIV.

2.3. Unadjusted SVR by CKD Seventy

The unadjusted SVR rates were >95% across all eGFR and CKD severity groups. SVR rates ranged from 96.4% (n=4,203/4,361) in controls with eGFR ≥60 to 96.5% (n=714/740) in patients with Stages 4-5 CKD and ESRD requiring dialysis, and 98.3% (n=845/860) in patients with Stage 3 CKD. SVR by eGFR and CKD severity groups in various clinical and demographic subgroups are described in Table 2. By eGFR and CKD severity, SVR rates were higher than 95% regardless of age, sex, genotype, BMI level, or baseline viral load. The SVR for patients who were White Hispanic with Stage 3 CKD was 88.9% (n=16/18) compared to 92.6% (n=25/27) and 95.7% (n=156/163) for patients with Stages 4-5 CKD and controls with eGFR ≥60, respectively. In patients with HIV and Stages 4-5 CKD, the SVR rate ranged from 93.9% to 100%. Of patients who had any history of previous HCV treatment, patients with Stages 4-5 CKD had an SVR rate of 92.1%, while controls with eGFR ≥60 and Stage 3 CKD both had an SVR rate of roughly 96%. Among patient who were treated with EBR/GZR for 12 weeks, we found the SVR rate to be 97.3% (n=5,194/5,337). The SVR in controls and by CKD stage in GT1a and GT1b ranged from 95% to 100% (see Supplemental Tables 2 and 3).

Table 2.

SVR in HCV GT1 patients treated with EBR/GZR by CKD status (n=5,961)

Variable	Controls with eGFR ≥60 (SVR%) n=4,361	Stage 3 CKD (SVR%) n=860	Stage 4-5 CKD/ESRD (SVR%) n=740
Total	96.4%	98.3%*	714/740 (96.5%)
Age (years)	-	-	-
25-65	2,866 (96.4)	411 (98.3)	411 (95.4)
>65	1,337 (96.3)	434 (98.2)	303 (98.1)
Sex	-	-	-
Male	4,068 (96.4)	816 (98.2)	702 (96.5)
Female	135 (97.2)	29 (100.0)	12 (92.3)
Race/ethnicity	-	-	-
White Non-Hispanic	1,602 (96.6)	255 (97.0)	106 (97.3)
White Hispanic	156 (95.7)	16 (88.9)	25 (92.6)
Black	2,175 (96.4)	522 (99.1)	548 (96.3)
Other	81 (96.4)	19 (100.0)	9 (100.0)
HCV Genotype	-	-	-
1a	1,593 (94.9)	379 (96.9)	447 (96.3)
1b	2,525 (97.4)	435 (99.3)	237 (96.7)
BMI (kg/m2)	-	-	-
<25	1,337 (97.1)	268 (99.3)	281 (97.2)
25-<30	1,596 (96.4)	283 (97.6)	238 (95.2)
≥30	1,236 (95.5)	291 (98.0)	186 (96.9)
% baseline HCV RNA	-	-	-
<800,000 IU/mL	1,082 (98.5)	232 (99.6)	288 (97.6)
≥800,000 IU/mL	2,927 (95.6)	586 (97.7)	398 (95.7)
Previous HCV treatment	-	-	-
Naïve	3,720 (96.5)	766 (98.5)	633 (97.1)
Experienced	483 (95.6)	79 (96.3)	81 (92.1)
Regimen	-	-	-
EBR/GZR without RBV 12 weeks	3,787 (97.0)	763 (99.1)	644 (97.1)
Other	416 (92.1)	82 (91.4)	70 (91.6)
Cirrhosis	1,054 (95.6)	248 (98.0)	297 (96.0)
History of alcohol abuse	2,465 (96.7)	479 (98.2)	352 (97.2)
History of drug abuse	2,097 (96.8)	432 (98.2)	340 (96.9)
HIV coinfected	53 (98.2)	22 (100.0)	46 (93.9)
Diabetes	1,324 (95.7)	390 (98.0)	469 (96.1)
Dialysis	-	-	-
Yes	0 (0.0)	0 (0.0)	541 (96.1)
No	4203 (96.4)	845 (98.3)	173 (97.7)

Note:

^*p<0.05 for stage 3 CKD patients vs. controls

2.4. Association Between CKD Severity, Dialysis, and SVR

We conducted multivariate logistic regression analyses to assess risk factors associated to the SVR achievement (Table 3: Model A and Model B). Overall, there was no difference in adjusted SVRs across eGFR and CKD Stages, however patients who had Stage 3 CKD (adjusted OR 2.23; CI 1.30-3.84) had slightly increased odds of achieving SVR compared to patients with controls with eGFR ≥60 (Table 3). Compared with other regimens, those who were prescribed EBR/GZR without RBV for 12 weeks (adjusted OR 3.04; CI 2.17-4.25) also had increased odds of achieving SVR. Patients with high baseline viral load were less likely to achieve SVR (adjusted OR 0.39; 0.25-0.59) compared to patients with low baseline viral load (Table 3). In the sensitivity analysis examining the impact of having dialysis on the likelihood of achieving SVR (Model B), there was no statistical difference in likelihood of SVR between patients with or without dialysis (adjusted OR 1.74; 95% CI 0.63-4.81). There were no statistical differences in achieving SVR across age group, CKD groups with and without dialysis, race, genotype, BMI, and previous HCV treatment.

DISCUSSION

We found in a large retrospective cohort study of 5,961 patients that EBR/GZR was highly effective in patients with HCV GT1 infection across CKD stages III-V, including patients who were on dialysis as well in several patient subgroups in the large national VA clinical practice setting. The overall SVR rate was 96.7%, ranging from 96.4% (controls with eGFR ≥60) to 98.3% (Stage 3 CKD) and 96.1% in dialysis.

Overall, SVR rates among patients with stages III-V CKD were high regardless of age, sex, GT1a/1b genotype, cirrhosis, history of alcohol or drug abuse and diabetes. However, we found that patients treated with EBR/GZR only regimen for 12 weeks were more likely to achieve SVR than other regimens. It is possible that these patients were less likely to have a resistance associated substitution (RAS) than patients on the other regimens; however, we were unable to test this hypothesis in this study. Although the SVR rates for all observed CKD stages were overall high, SVR in patients with Stage 3 CKD was only slightly higher than controls with eGFR≥60 in comparison (1.9% crude difference). This finding may be attributed to patients with Stage 3 CKD having slightly lower baseline viral load and more likely to be treatment naive when compared to controls with eGFR≥60. Although the difference remained after adjustment for these factors.

In the phase 3 multicenter, randomized, placebo-controlled C-SURFER study (n=215), response rates of EBR/GZR in patients with Stage 4-5 CKD/ESRD were similar and exceeded 98%.²⁹ The results of the phase 3 multicenter study showed EBR/GZR was an effective, safe treatment option, comparable to the patients in the study who received placebo.²⁹ Similar to our study, SVR rates remained high across all patient subgroups, including patients with cirrhosis and diabetes.²⁹In comparison to other real-world studies, such as the retrospective analyses from the TRIO network study (n=470), 89% of patients with CKD stage 4 or 5 or ESRD requiring dialysis and HCV genotypes 1 or 4 received EBR/GZR for 12 weeks.¹² Response rates for EBR/GZR in patients with genotype 1 and among patients with Stages 4-5 CKD was slightly higher than our findings, at 99% (n=113/114 and n=108/109 respectively), while the SVR rate for patients with HCV genotype 4 was 95% (21/22).¹² However, race/ethnicity was not reported for most patients in this study. Similar findings were reported in another real-world, multicenter study in Japan that evaluated EBR/GZR for 12 weeks in patients with HCV genotype 1 and chronic kidney disease.¹⁷ In this real-world study by Ogawa et al., 31.6% (n=89) were CKD Stage 3-5 and 7.4% (n=21) were CKD stage 5D (hemodialysis dependent).¹⁷ The overall and CKD Stage 3-5D SVR rates of SVR in patients receiving EBR/GZR treatment was 98.6% (n=272) and 98.1% (n=101) respectively.¹⁷

Our study had few limitations. First, the number of patients who received dialysis may be under-estimated because some dialysis claims from outside the VA healthcare system may not be completely captured by our databases.²¹ However, these missing claims would have minimal impact on our results because the SVR rate was high in patients with CKD regardless of dialysis receipt. We also did not have data on RAS in these patients because it is not consistently reported in the CDW data. Furthermore, data for safety and adverse events was not evaluated in this study because data on adverse events and related abnormalities were not systematically collected. Symptoms such as headache and fatigue related to treatment are not readily available in CDW. Finally, our results may lack generalizability outside the VA system because our cohort was predominately male, African American with low socioeconomic status.

In the largest, most racially diverse real-world effectiveness study of EBR/GZR among the CKD population to date, we found SVR rates in patients with HCV GT1 and CKD including those undergoing dialysis were similar to controls with eGFR≥60 and those reported in clinical trials.^{2,6,9,12–14,17,29,30} Further studies should include the evaluation of the long-term effects on renal function outcomes and reinfections, after the elimination of HCV.

RESEARCH HIGHLIGHTS.

• We examined the real-world effectiveness of EBR/GZR in HCV GT1 infected patients with CKD and receipt of dialysis.

• We utilized electronic medical and laboratory records from a racially diverse VA population.

• EBR/GZR regimens were highly effective in HCV GT1 infected patients, regardless of CKD severity or receipt of dialysis.

Abbreviations:

CKD

Chronic Kidney Disease

C-SURFER

Hepatitis C: Study to Understand Renal Failure’s Effect on Responses

eGFR

Estimated Glomerular Filtration Rate

ESRD

End Stage Renal Disease

FDA

U.S. Food and Drug Administration

GT

Genotype

HCV

Hepatitis C Virus

SVR

Sustained Virologic Response

VA

Veterans Affairs

VBA

Veterans Benefits Administration

VHA

Veterans Health Administration