Content uploaded by Lisa I Backus
Author content
All content in this area was uploaded by Lisa I Backus on Oct 18, 2017
Content may be subject to copyright.
Real-World Effectiveness of Ledipasvir/
Sofosbuvir in 4,365 Treatment-Naive,
Genotype 1 Hepatitis C-Infected Patients
Lisa I. Backus, Pamela S. Belperio, Troy A. Shahoumian, Timothy P. Loomis, and Larry A. Mole
Real-world effectiveness data are needed to inform hepatitis C virus (HCV) treatment decisions. The uptake of ledi-
pasvir/sofosbuvir (LDV/SOF) regimens across health care settings has been rapid, but variations often occur in clinical
practice. The aim of this study was to assess sustained virologic response (SVR) of LDV/SOF6ribavirin (RBV) in rou-
tine medical practice. This observational, intent-to-treat cohort was comprised of 4,365 genotype 1, treatment-naive,
HCV-infected veterans treated with LDV/SOF6RBV. SVR rates were 91.3% (3,191/3,495) for LDV/SOF and 92.0%
(527/573) for LDV/SOF1RBV (P50.65). African American race (odds ratio 0.70, 95% confidence interval 0.54-0.90,
P50.004) and FIB-4 >3.25 (odds ratio 0.56, 95% confidence interval 0.43-0.71, P<0.001) were independently associ-
ated with decreased likelihood of SVR; age, sex, body mass index, decompensated liver disease, diabetes, genotype 1
subtype, and regimen did not predict SVR. In models limited to those who completed 12 weeks of treatment, African
American race was no longer a significant predictor of SVR but FIB-4 >3.25 (odds ratio 0.35, 95% confidence interval
0.24-0.50, P<0.001) remained. Among those without cirrhosis (defined by FIB-4 3.25) and with baseline HCV
RNA<6,000,000 IU/mL, SVR rates were 93.2% (1,020/1,094) for those who completed 8 weeks of therapy and 96.6%
(875/906) for those who completed 12 weeks of therapy (P50.001). Conclusions: In this real-world cohort, SVR rates
with LDV/SOF6RBV nearly matched the rates reported in clinical trials and were consistently high across all sub-
groups; those without cirrhosis but with HCV RNA<6,000,000 IU/mL were less likely to achieve SVR with 8 weeks
compared to 12 weeks of therapy, although the numeric difference in SVR rates was small.(H
EPATOLOGY 2016;64:405-
414)
Antiviral therapy for chronic hepatitis C virus
(HCV) infection continues to rapidly evolve.
Sustained virologic response (SVR) rates
reported in clinical trials with the newest regimens are
consistently above 90% for most HCV-infected patient
populations. The promise of such results in clinical tri-
als has created similarly high expectations among both
providers and patients. However, studies examining
real-world effectiveness with interferon-based HCV
regimens and, more recently, sofosbuvir plus simepre-
vir regimens revealed differences in outcomes com-
pared to clinical trials.
(1-7)
These differences may
reflect the wide heterogeneity in clinical practice where
patient characteristics, care coordination, and manage-
ment are not uniform. Only when prescribed to a
broader population do these variances become appa-
rent. Understanding the effectiveness of antiviral regi-
mens in real-world settings is essential to provide
practical information to better inform HCV treatment
decisions.
The use of ledipasvir/sofosbuvir (LDV/SOF) for
varying treatment durations has been extensively eval-
uated in clinical trials of treatment-naive, genotype 1
HCV-infected individuals. The phase 3 ION-1 clinical
Abbreviations: APRI, aspartate aminotransferase-to-platelet ratio index; BMI, body mass index; CI, confidence interval; EOT, end of treatment;
FDA, US Food and Drug Administration; HCV, hepatitis C virus; IL, interleukin; LDV/SOF, ledipasvir/sofosbuvir; OR, odds ratio; RBV, ribavirin;
SVR, sustained virologic response; VA, Department of Veterans Affairs.
Received December 24, 2015; accepted April 20, 2016.
Additional Supporting Information may be found at onlinelibrary.wiley.com/doi/10.1002/hep.28625/suppinfo.
Published 2016. This article is a U.S. Government work and is in the public domain in the USA.
View this article online at wileyonlinelibrary.com.
DOI 10.1002/hep.28625
Potential conflict of interest: Nothing to report.
405
HEPATOLOGY, VOL. 64, NO. 2, 2016
A
HE STU
DYOFLIVERD I S E ASES
T
ME
RIC
A
N
A
SSOCI
ATIO
NFOR
trial comparing 12-week and 24-week regimens resulted
in SVR rates of 97%-100% in all treatment arms includ-
ing subgroups defined by race, genotype subtype, and
presence of cirrhosis.
(8)
The high SVRs observed in this
study led to the exploration of shortening treatment
duration even further. In ION-3, LDV/SOF 6riba-
virin (RBV) regimens of 8 and 12 weeks’ duration were
compared in treatment-naive, HCV genotype
1-infected patients without cirrhosis.
(9)
While SVR
rates were high in all treatment arms (93%-95%),
relapse rates were higher in the 8-week treatment arms
(5%) compared to the 12-week arms (1%). Post hoc anal-
ysis of baseline predictors revealed higher relapse rates
only in patients receiving the 8-week regimen with a
baseline HCV RNA above 6 million IU/mL. Based on
data from this trial, current US Food and Drug Admin-
istration (FDA) labeling indicates that an 8-week regi-
men of LDV/SOF can be considered in patients who
are treatment-naive without cirrhosis and have a base-
line HCV RNA below 6,000,000 IU/mL. Though the
generalizability of this approach may be viewed as lim-
ited, it is being widely used in clinical practice.
HCV disproportionately affects the veteran popula-
tion, and this population is often underrepresented in
clinical trials, largely due to complicating comorbid-
ities.
(10)
Thus, evaluating the effectiveness of HCV
antiviral regimens remains a priority for the Depart-
ment of Veterans Affairs (VA).
(11)
With the rapid
uptake of LDV/SOF regimens across health care set-
tings and variations that often occur in clinical practice,
we examined real-world outcomes in the diverse
HCV-infected veteran population receiving this regi-
men. Our aim was to assess the effectiveness of LDV/
SOF6RBV in treatment-na€ıve, genotype 1 HCV-
infected veterans treated in routine medical practice.
Materials and Methods
This was an observational intent-to-treat cohort
analysis of HCV-infected veterans receiving LDV/
SOF6RBV from the VA. Data for this study were
obtained from the VA’s Clinical Case Registry for
HCV, an extract of the VA electronic medical record
that contains demographics, laboratory results, phar-
macy information, and International Classification of
Diseases diagnosis codes from inpatient hospitaliza-
tions, outpatient visits, and problem lists of HCV-
infected veterans seen at all VA medical facilities.
(12)
Eligible subjects included all treatment-naive, geno-
type 1 HCV-infected veterans from any VA facility
nationwide who initiated VA-prescribed LDV/SOF6
RBV with a recommended duration of 8 or 12 weeks
between November 1, 2014, and March 31, 2015, and
stopped treatment by July 14, 2015. The choice of reg-
imen and timing of follow-up visits and laboratory
testing were at the discretion of the provider as patients
were treated in routine practice. Patients were excluded
if they changed regimens (n 58), had baseline HCV
RNA 1000 IU/mL (n 582), had undergone liver
transplantation (n 5113), or had received more than
91 days of LDV/SOF6RBV (n 5266).
TREATMENT OUTCOME
SVR was defined as an undetectable HCV RNA on
all HCV RNA tests after the end of treatment (EOT)
including at least one test 10 weeks or more after the
EOT. This 10-week time point was chosen to account
for variability of clinic visits and timing of lab draws in
clinical practice. Patients were categorized as not
achieving SVR if they had a detectable HCV RNA
after the EOT, had no viral load testing after the
EOT, and had a detectable HCV RNA on their last
HCV viral load test while on treatment or died while
on treatment or within 10 weeks of the EOT. Patients
with undetectable HCV RNA on their last HCV viral
load test, either on treatment or after the EOT, but no
test 10 weeks of more after the EOT were excluded
from the SVR analysis. The EOT was calculated as
the last day covered by prescriptions of LDV/SOF
using the dates the medication was dispensed and the
ARTICLE INFORMATION:
From the Department of Veterans Affairs, Palo Alto Health Care System, Palo Alto, CA.
ADDRESS CORRESPONDENCE AND REPRINT REQUESTS TO:
Lisa I. Backus, M.D., Ph.D.
Patient Care Services/Population Health Services
Veterans Affairs Palo Alto Health Care System
Palo Alto, CA 94304
E-mail: Lisa.Backus@va.gov
Tel: 11-650-849-0365
BACKUS ET AL. HEPATOLOGY, August 2016
406
number of days of supply. HCV RNA was categorized
as detectable or undetectable based on the locally
reported HCV RNA result of which 98% used assays
with a lower limit of detection of 7 IU/mL or less.
Patients were followed from the initiation of LDV/
SOF6RBV through February 29, 2016, allowing for
more than 32 weeks of follow-up after the EOT for all
patients in the cohort.
CONTROL VARIABLES
Demographic and other baseline variables were deter-
mined at the time of treatment initiation and included
age, sex, race/ethnicity, history of decompensated liver
disease (defined as esophageal variceal hemorrhage,
hepatic coma, hepatorenal syndrome, or spontaneous
bacterial peritonitis), diabetes, human immunodefi-
ciency virus coinfection, and HCV genotype 1 subtype.
Baseline values for height and weight, which were used
to calculate body mass index (BMI), and laboratory tests
for alanine aminotransferase, aspartate aminotransferase,
platelets, and baseline HCV RNA were defined as the
value within 1 year before and closest to the treatment
start date. A FIB-4 score >3.25 or an aspartate
aminotransferase-to-platelet ratio index (APRI) score
>2 at the start of treatment using baseline laboratory
values was used as a marker of advanced liver disease.
(13-
15)
Patients with FIB-4 3.25 or APRI 2 were con-
sidered to not have cirrhosis.
The percentage of patients on treatment was calcu-
lated from the cumulative days’ supply of all LDV/
SOF prescriptions starting with the first prescription
for LDV/SOF through the last day of treatment cov-
ered by LDV/SOF. In the VA, LDV/SOF prescrip-
tions are frequently filled for quantities less than 28
days, particularly for the first 1 or 2 months of treat-
ment. Patients were considered to have completed 8
and 12 weeks of LDV/SOF6RBV if they had received
between 49-63 and 77-91 days’ worth of medication,
respectively.
On-treatment HCV RNA at 4 weeks was deter-
mined using the locally reported result closest to and
within 2 weeks prior to and 2 weeks after the specified
time point. HCV RNA results at 4 weeks were catego-
rized as undetectable, detectable <15 IU/mL, and
detectable 15 IU/mL.
STATISTICAL ANALYSIS
Univariate comparisons used the Pearson chi-squared
test with Yates’ continuity correction for categorical var-
iables and analysis of variance for continuous variables.
Multivariate logistic regression models were constructed
to model SVR. Variables included in the multivariate
regression were selected a priori based on variables that
had been previously shown to impact SVR or were of
clinical interest. Models included age, sex, race/ethnic-
ity, history of decompensated liver disease, diabetes,
genotype 1 subtype, regimen, and FIB-4. A set of mod-
els with the above baseline variables was constructed
with patients who completed 12 weeks of treatment.
These models were rerun also including HCV RNA at
4 weeks. In addition, a set of models with the baseline
variables was constructed with only patients who com-
pleted 8 or 12 weeks of LDV/SOF and then further
restricted to those patients who met the criteria for not
having cirrhosis based on FIB-4 3.25 and baseline
HCV RNA <6,000,000 IU/mL.
For all comparisons, P<0.01 was considered statis-
tically significant. All analyses were performed using
R, version 3.1 (R Foundation for Statistical Comput-
ing, Vienna, Austria).
The protocol was approved by the Stanford Univer-
sity Institutional Review Board and the VA Palo Alto
Health Care System Research and Development
Committee.
Results
In total, 4,834 treatment-naive patients with HCV
genotype 1 initiated LDV/SOF6RBV at 124 VA
facilities. After applying exclusion criteria, 4,365
remained. The mean age for the cohort overall was
61.3 years, 96.1% were male, 36.6% were African
American, 3.3% had a history of decompensated liver
disease, 34.2% had a BMI 30 kg/m
2
, 30.5% had dia-
betes, and 29.2% had FIB-4 >3.25.
Baseline characteristics for the cohort by regimen
appear in Table 1. For the cohort, 86.3% (n 53,763)
received LDV/SOF and 13.7% (n 5602) received
LDV/SOF1RBV. Patients who received LDV/
SOF1RBV were more likely to have markers of
advanced liver disease including lower mean platelet
count, higher mean APRI score, APRI >2, higher
mean FIB-4 score, FIB-4 >3.25, and a history of
decompensated liver disease.
Among patients who received LDV/SOF, 3.6%
(n 5134) discontinued treatment before 8 weeks,
37.0% (n 51,392) received 8 weeks, 1.7% (n 565)
discontinued treatment between 8 and 12 weeks,
and 57.7% (n 52,172) received 12 weeks. In total,
HEPATOLOGY, Vol. 64, No. 2, 2016 BACKUS ET AL.
407
94.7% completed either an 8-week or a 12-week
course. Among people who received LDV/
SOF1RBV, 90.0% (n 5542) completed 12 weeks
of treatment.
SVR results were available for 4,068 patients,
including 12 patients who died while on treatment or
shortly after who were categorized as no SVR. Patients
whose last HCV RNA was undetectable but occurred
while still on treatment (n 586) or less than 10 weeks
after the EOT (n 5211) were excluded from the SVR
analysis. Patients with an undetectable HCV RNA
obtained 10 or 11 weeks after the EOT were included
in the SVR analysis for reasons described previously
(n 5197). In additional support for categorizing these
individuals as achieving SVR, 98 other patients in this
cohort had undetectable HCV RNA test results in
week 10 or 11 with subsequent HCV RNA testing at
12 weeks or later. All 98 patients remained undetect-
able on the later HCV RNA testing, further validating
the acceptance of testing at 10 weeks or later as a prac-
tical accommodation for the realities of clinical practice
while providing an accurate assessment of virologic
response that would be seen if testing were conducted
strictly at 12 weeks.
Overall SVR rates were similar for LDV/SOF and
LDV/SOF1RBV (91.3% versus 92.0% respectively, P
50.65) (Table 2). For patients who received LDV/
SOF, SVR rates were significantly lower in African
Americans compared to Caucasians, patients with
APRI >2, and patients with FIB-4 >3.25 but for no
other baseline characteristics examined. Significantly
lower SVR rates were observed in those receiving
LDV/SOF who had a 4-week on-treatment detectable
HCV RNA 15 IU/mL compared to those who were
TABLE 1. Baseline Characteristics and 4 Week On-Treatment Response
of Treatment-Naive Genotype1 Patients Receiving LDV/SOF-Based Regimens
LDV/SOF
(n 53,763)
LDV/SOF1RBV
(n 5602)
P
*
Age (years) 61.1 66.7 (25.3-90.8) 62.1 65.2 (37.0-86.2)
Sex, male 95.9% 97.2%
Race/ethnicity <0.001
African American 37.9% 28.1%
Caucasian 50.8% 57.1%
Hispanic 4.3% 6.8%
Other/multiple 7.0% 8.0%
Decompensated liver disease 2.1% 10.3% <0.001
Diabetes 29.6% 36.2% 0.001
HIV coinfected 5.4% 3.5%
BMI (kg/m
2
) 28.4 65.1 (15.8-65.2) 29.5 65.4 (17.9-60.1) <0.001
BMI (kg/m
2
)<0.001
<25 25.2% 20.1%
25-29 41.8% 38.0%
30 33.0% 41.9%
ALT (U/L) 72.3 656.6 (8-659) 83.4 656.9 (16-445) <0.001
AST (U/L) 61.2 643 (6-614) 80.2 649.7 (11-427) <0.001
Platelets (K/lL) 193.8 668.7 (6-661) 148.2 670.7 (22-759) <0.001
APRI 1.0 61.2 (0.1-20.4) 1.9 61.8 (0.1-11.5) <0.001
APRI >2 12.0% 31.3% <0.001
FIB-4 2.9 62.7 (0.1-51.4) 5.1 64.4 (0.6-34.7) <0.001
FIB-4 >3.25 24.8% 56.7% <0.001
HCV RNA (log IU/mL) 6.2 60.7 (3.1-7.9) 6.2 60.7 (3.9-7.8)
HCV RNA <6,000,000 IU/mL 83.2% 83.9%
HCV subtype 1b 24.0% 21.1%
IL28B polymorphism n 5462 n 569
CC 25.8% 14.5%
CT 50.2% 56.5%
TT 24.0% 29.0%
4-week HCV RNA n 53265 n 5557
Undetectable 76.4% 75.6%
Detectable <15 IU/mL 9.1% 8.4%
Detectable 15 IU/mL 14.6% 16.0%
Continuous variables reported as mean 6standard deviation (range). Categorical variables reported as percentage.
*Pvalues listed if <0.01 between LDV/SOF and LDV/SOF1RBV.
Abbreviations: ALT, alanine amino transferase; AST, aspartate aminotransferase; HIV, human immunodeficiency virus.
BACKUS ET AL. HEPATOLOGY, August 2016
408
undetectable by week 4. For patients who received
LDV/SOF1RBV, SVR rates were numerically lower
in those with high FIB-4 or APRI scores; but these
differences were not statistically significant. SVR rates
were also significantly lower in those receiving LDV/
SOF1RBV who had a 4-week on-treatment detecta-
ble HCV RNA 15 IU/mL compared to those who
were undetectable by week 4.
Among patients who completed a full 12 weeks of
treatment, SVR rates did not differ between LDV/
SOF and LDV/SOF1RBV (94.3% [1,924/2,040]
versus 94.2% [485/515], respectively; P51.00).
Among patients who completed a full 12 weeks of
LDV/SOF, SVR rates remained significantly lower in
those with APRI >2, FIB-4 >3.25, and 4-week on-
treatment HCV RNA 15 IU/mL compared to those
TABLE 2. SVR Rates by Regimen for Genotype 1, Treatment-Naive Patients Receiving LDV/SOF-Based Regimens
LDV/SOF
(n 53,495)
P
*
LDV/SOF1RBV
(n 5573)
P
*
Overall SVR 91.3% (3,191/3,495) 92.0% (527/573)
Age (years)
<55 93.8% (361/385) 95.1% (39/41)
55-64 90.3% (1,918/2,124) 90.7% (323/356)
65 92.5% (912/986) 93.8% (165/176)
Sex
Male 91.1% (3,052/3,349) 92.1% (512/556)
Female 95.2% (139/146) 88.2% (15/17)
Race/ethnicity
African American 89.8% (1,185/1,320) 0.003 93.2% (151/162)
Caucasian 92.8% (1,657/1,785) 92.0% (298/324)
Hispanic 88.2% (134/152) 90.0% (36/40)
Other/multiple 90.3% (215/238) 89.4% (42/47)
Decompensated liver disease
No 91.3% (3,125/3,421) 92.6% (476/514)
Yes 89.2% (66/74) 86.4% (51/59)
Diabetes
No 91.3% (2,236/2,448) 93.7% (344/367)
Yes 91.2% (955/1,047) 88.8% (183/206)
HIV coinfection
No 91.3% (3,013/3,300) 92.1% (510/554)
Yes 91.3% (178/195) 89.5% (17/19)
BMI (kg/m
2
)
<25 89.8% (791/881) 93.3% (111/119)
25-29 92.8% (1,356/1,461) 93.0% (200/215)
30 90.5% (1,044/1,153) 90.4% (216/239)
APRI
2 92.0% (2,814/3,059) <0.001 93.6% (368/393)
>2 86.7% (372/429) 88.2% (157/178)
FIB-4
3.25 92.6% (2,407/2,599) <0.001 94.4% (234/248)
>3.25 87.6% (779/889) 90.1% (291/323)
HCV RNA (log IU/mL)
<6,000,000 IU/mL 91.4% (2,657/2,908) 91.9% (440/479)
6,000,000 IU/mL 91.0% (534/587) 92.6% (87/94)
HCV subtype
1a
†
90.9% (2,402/2,642) 91.6% (414/452)
1b 92.5% (789/853) 93.4% (113/121)
IL28B polymorphism n 5441 n 565
CC 92.2% (107/116) 90.0% (9/10)
CT 91.8% (202/220) 94.4% (34/36)
TT 93.3% (98/105) 89.5% (17/19)
4-week HCV RNA n 53,036 n 5532
Undetectable 95.3% (2,204/2,312) NS 95.3% (384/403) NS
Detectable <15 IU/mL 93.2% (260/279) <0.001 88.6% (39/44) <0.001
Detectable 15 IU/mL 84.8% (393/445) 88.2% (75/85)
*Pvalues only listed if <0.01 between categories of each baseline variable for the same regimen.
†
Subtype 1a includes 1a, mixed 1a/1b, and 1 with subtype unspecified.
Abbreviations: HIV, human immunodeficiency virus; NS, not significant.
]
]
]
]
]]
]
HEPATOLOGY, Vol. 64, No. 2, 2016 BACKUS ET AL.
409
with undetectable 4-week HCV RNA (Supporting
Table S1).
In multivariate analysis, significant independent pre-
dictors of SVR were African American race compared
to Caucasian (odds ratio [OR] 0.70, 95% confidence
interval [CI] 0.54-0.90, P50.004) and FIB-4 >3.25
compared to FIB-4 3.25 (OR 0.56, 95% CI 0.43-
0.71, P<0.001) (Table 3). Age, sex, BMI, diabetes,
history of decompensated liver disease, genotype 1 sub-
type, and regimen did not predict SVR. When
restricted to those patients who completed 12 weeks of
treatment, only FIB-4 >3.25 remained a significant
predictor of decreased odds of SVR (OR 0.35, 95% CI
0.24-0.50, P<0.001). In both models the substitution
of APRI with a cut point of 2 in place of FIB-4 pro-
duced similar ORs (data not shown).
Four-week on-treatment HCV RNA was an inde-
pendent predictor of SVR when included as a variable
in the models (Supporting Table S2). Having a detect-
able HCV RNA 15 IU/mL was associated with
reduced odds of SVR compared to undetectable in the
full model (OR 0.40, 95% CI 0.29-0.56, P<0.001)
and in the model limited to those who completed 12
weeks of treatment (OR 0.38, 95% CI 0.25-0.58, P<
0.001).
With regard to treatment duration, as noted above,
among patients who received LDV/SOF, 37.0% (n 5
1392) received 8 weeks and 57.7% (n 52172) received
12 weeks. Of the 1,392 receiving 8 weeks, 86.9% (n 5
1209) met FDA criteria of a baseline HCV RNA
<6,000,000 IU/mL when “without cirrhosis” was
defined as FIB-4 3.25 and 92.7% (n 51291) met
FDA criteria when “without cirrhosis” was defined as
APRI 2. Of the 2,172 patients who received 12
weeks of LDV/SOF, 45.0% (n 5977) also met the
FDA criteria for 8 weeks of treatment using the FIB-4
definition of “without cirrhosis” and 58.4% (n 51268)
met the FDA criteria using the APRI definition of
“without cirrhosis.”
The overall SVR rate in those who completed 8
weeks of LDV/SOF treatment was 92.1% compared
to 94.3% in those who completed 12 weeks of treat-
ment (Table 4). Among patients in the subgroups who
met the FDA criteria, SVR rates were statistically
lower for those who received 8 weeks of LDV/SOF
than for those who received 12 weeks of LDV/SOF,
regardless of whether “without cirrhosis” was defined
by APRI 2 or FIB-4 3.25. The numeric difference
in SVR between those receiving 8 or 12 weeks was
2.8% and 3.4% when “without cirrhosis” was defined
by APRI 2 or FIB-4 3.25, respectively. SVR rates
for other baseline characteristics in patients who com-
pleted 8 or 12 weeks of LDV/SOF treatment are avail-
able in Supporting Table S1.
In multivariate analysis of patients who completed 8
or 12 weeks of LDV/SOF, significant independent
predictors of SVR were African American race (OR
0.66, 95% CI 0.48-0.90, P50.008), FIB-4 >3.25
(OR 0.45, 95% CI 0.32-0.62, P<0.001), and 8-week
treatment duration compared to 12-week treatment
duration (OR 0.56, 95% CI 0.41-0.76, P<0.001)
(Table 5). In models limited to patients with baseline
HCV RNA <6,000,000 IU/mL and FIB-4 3.25,
only 8-week treatment duration remained a significant
TABLE 3. ORs for SVR in Multivariable Model for Treatment-Naive
Genotype 1 Patients Treated With LDV/SOF-Based Regimens
OR (95% CI)
(n 54,059)
Completed 12 weeks
OR (95% CI)
(n 52,533)
Age <55 years (ref. 55-64) 1.43 (0.95-2.24) 1.31 (0.69-2.75)
Age 65 years (ref. 55-64) 1.39 (1.07-1.82) 1.31 (0.89-1.96)
Female (ref. male) 1.83 (0.94-4.12) 2.72 (0.83-16.8)
African American (ref. Caucasian) 0.70 (0.54-0.90)* 0.85 (0.57-1.26)
Hispanic (ref. Caucasian) 0.62 (0.39-1.03) 0.77 (0.38-1.71)
Other/multiple (ref. Caucasian) 0.73 (0.48-1.16) 0.50 (0.29-0.91)
BMI <25 kg/m
2
(ref. BMI 25-29 kg/m
2
) 0.71 (0.53-0.96) 0.86 (0.54-1.38)
BMI 30 kg/m
2
(ref. BMI 25-29 kg/m
2
) 0.73 (0.55-0.95) 0.68 (0.45-1.00)
Diabetes 0.97 (0.76-1.25) 0.93 (0.65-1.35)
FIB-4 >3.25 (ref. FIB-4 3.25) 0.56 (0.43-0.71)
†
0.35 (0.24-0.50)
†
History of decompensated liver disease (ref. no) 0.88 (0.51-1.59) 0.89 (0.47-1.84)
Subtype 1b 1.26 (0.96-1.68) 1.13 (0.75-1.77)
LDV/SOF1RBV (ref. LDV/SOF) 1.34 (0.96-1.91) 1.31 (0.86-2.06)
*P<0.01.
†
P<0.001.
Abbreviation: ref., reference.
BACKUS ET AL. HEPATOLOGY, August 2016
410
predictor of reduced odds of SVR (OR 0.47, 95% CI
0.30-0.72, P<0.001). Similar results were obtained in
models limited to patients with baseline HCV RNA
<6,000,000 IU/mL and APRI 2 (data not shown).
Discussion
In this real-world cohort of treatment-naive, geno-
type 1 HCV-infected veterans treated with LDV/
SOF-based therapy at any VA facility nationwide, we
observed SVR rates over 90%, nearly matching the
rates reported in clinical trials.
(8,9)
The response rates
were consistently high across all subgroups evaluated,
regardless of regimen or duration, and ranged from
86% to 97%. Use of RBV did not appear to offer gen-
eral benefit and was associated with higher early dis-
continuation rates, although overall more veterans were
able to successfully complete treatment with LDV/
SOF-based regimens compared to regimens previously
evaluated.
(1,5)
Shorter, 8-week, regimens of LDV/SOF
were widely used in patients with baseline HCV RNA
below 6,000,000 IU/mL without cirrhosis; however,
SVR rates were significantly lower in those treated
with 8 weeks compared to 12 weeks.
The large differences between real-world effective-
ness and clinical trial efficacy previously observed have
now been almost eliminated with the use of potent all-
oral regimens, which are well-tolerated and easily
administered for shorter durations. The remaining
small decrement in effectiveness in this real-world
cohort may be explained by differences in patient pop-
ulations as our cohort consisted of mostly older male
patients with generally higher BMI, a greater propor-
tion of African Americans, and more prevalent evi-
dence of advanced liver disease. Furthermore, there are
persistent variations in practice patterns, patient moti-
vation, provider knowledge, provider resources, and
ancillary services in routine medical practice compared
to highly structured, highly resourced clinical trials.
Reported virologic response rates to other direct-
acting antiviral and prior interferon-based regimens
showed wide variability in clinical practice, particularly
TABLE 4. SVR Rates by Regimen for Genotype 1 Treatment-Naive Patients Who Completed 8 or 12 Weeks of LDV/SOF
LDV/SOF 8 weeks LDV/SOF 12 weeks
P
*
Overall SVR 92.1% (1,169/1,269) 94.3% (1,924/2,040)
HCV RNA <6,000,000 IU/mL 92.7% (1,138/1,227) 94.1% (1,431/1,520)
APRI 2 92.1% (1,124/1,220) 95.5% (1,603/1,679) <0.001
FIB-4 3.25 92.5% (1,043/1,128) 96.3% (1,285/1,335) <0.001
HCV RNA <6,000,000 IU/mL and APRI 2 92.8% (1,095/1,180) 95.6% (1,139/1,191) 0.004
HCV RNA <6,000,000 IU/mL and FIB-4 3.25 93.2% (1,020/1,094) 96.6% (875/906) 0.001
*Pvalues listed if <0.01 between 8 weeks and 12 weeks.
TABLE 5. ORs for SVR in Multivariable Model for Genotype 1 Treatment-Naive Patients
Who Completed 8 or 12 Weeks of LDV/SOF
OR (95% CI)
(n 53,303)
Baseline FIB-4 £3.25 and
HCV <6,000,000 IU/mL
OR (95% CI) (n 51,989)
Age <55 years (ref. 55-64) 1.27 (0.77-2.19) 1.38 (0.71-2.92)
Age 65 years (ref. 55-64) 1.25 (0.90-1.75) 1.22 (0.76-2.00)
Female (ref. male) 3.22 (1.19-13.3)
‡
African American (ref. Caucasian) 0.66 (0.48-0.90)* 0.75 (0.48-1.17)
Hispanic (ref. Caucasian) 0.67 (0.36-1.37) 0.60 (0.26-1.63)
Other/multiple (ref. Caucasian) 0.59 (0.36-1.03) 1.12 (0.50-3.00)
BMI <25 kg/m
2
(ref. BMI 25-29 kg/m
2
) 0.79 (0.55-1.14) 0.79 (0.48-1.34)
BMI 30 kg/m
2
(ref. BMI 25-29 kg/m
2
) 0.73 (0.52-1.01) 0.78 (0.49-1.26)
Diabetes 1.01 (0.74-1.40) 0.74 (0.47-1.18)
FIB-4>3.25 (ref. FIB-43.25) 0.45 (0.32-0.62)
†§
History of decompensated liver disease (ref. no) 0.91 (0.40-2.42)
§
Subtype 1b 1.32 (0.94-1.90) 1.28 (0.80-2.10)
8 weeks (ref. 12 weeks) 0.56 (0.41-0.76)
†
0.47 (0.30-0.72)
†
*P<0.01.
†
P<0.001.
‡
Female not included as a variable because of small numbers of females.
§
FIB-4 and history of decompensated liver disease not included in the model as patients had to have FIB-4 3.25 to be included.
HEPATOLOGY, Vol. 64, No. 2, 2016 BACKUS ET AL.
411
in patients with specific baseline characteristics includ-
ing African American race, genotype subtype 1a,
interleukin-28B (IL28B) non-CC subtype, and high
baseline HCV RNA.
(16-20)
A strength of the current
study is the large number of patients evaluated, which
allows for the detection of differences in subgroups that
might not otherwise be identified in smaller studies or
clinical trials. Nearly 1,500 African Americans were
evaluated in this cohort study, and our results indicate
that response rates may still be reduced in African
Americans compared to Caucasians. African American
race was a predictor of reduced likelihood of SVR in
multivariate models of the overall cohort but not in
models of those patients completing a full treatment
course, suggesting that perhaps tolerability or adherence
may be a contributing factor, though we did not specifi-
cally evaluate this. While in general RBV did not offer
any benefit overall in achieving SVR, there is an indica-
tion that its use in African Americans may improve
SVR rates as evidenced by the 89.8% SVR rate in those
who did not receive RBV and the 93.2% SVR in those
that did. Furthermore, treating for 12 weeks may also
overcome this difference as SVR rates in African Amer-
icans completing 12 weeks of LDV/SOF therapy were
higher than in those completing 8 weeks (94.0% versus
89.8%) and were similar to SVR rates achieved in Cau-
casians (12 weeks 95.2%, 8 weeks 93.9%). This finding
of lower SVR rates in African Americans receiving an
8-week treatment course of LDV/SOF is consistent
with results from the ION clinical trials, which showed
higher relapse rates among black patients receiving 8
weeks of treatment compared with those receiving 12
and 24 weeks of treatment.
(16)
The largest difference
between blacks and nonblacks occurred in those who
received 8 weeks of LDV/SOF therapy with baseline
HCV RNA 6,000,000 (83% versus 92%). This aligns
with our findings of those completing 8 or 12 weeks of
LDV/SOF treatment in which African American race
predicted decreased odds of achieving SVR overall but
no longer predicted decreased odds of SVR when lim-
ited to those who did not have cirrhosis with baseline
HCV RNA <6,000,000.
Clinical trials suggested that an 8-week treatment
course of LDV/SOF could be used in patients with
specific baseline characteristics, namely treatment-
naive patients without cirrhosis and with a pretreat-
ment HCV RNA <6,000,000 IU/mL; and these
criteria are included in FDA labeling.
(8,9,21)
The use of
an 8-week treatment course, however, is not recom-
mended in the American Association for the Study of
Liver Diseases/Infectious Diseases Society of America
hepatitis C guidance due to the higher relapse rate.
(22)
Our analysis highlights the complexity of the treatment
duration decision. SVR rates were statistically lower
with 8 weeks of treatment compared to 12 weeks of
treatment, and patients had approximately 53%
reduced odds of achieving SVR in multivariable mod-
els with 8 weeks of treatment compared to 12 weeks.
The absolute numeric SVR difference, however, was
small at only 3.4%; and SVR rates were debatably high
with either duration. From a patient perspective, the
benefit of even marginally increased SVR rates may
outweigh the inconvenience of an additional 4 weeks
of treatment. From a payer perspective, however, the
calculus is likely more complex and depends not only
on the initial drug cost but also on variables regarding
the cost and success rate of retreatment. In our cohort,
for example, we identified 977 patients out of a total of
2,303 patients with low baseline HCV RNA and FIB-
43.25 who met criteria for 8 weeks of treatment but
nevertheless completed 12 weeks of treatment. Had
these patients who qualified for an 8-week regimen
actually received 8 weeks rather than 12 weeks, over
$11 million in initial drug costs could have been
avoided; but this would be offset to an unknown
degree by the approximately 3%-4% of patients failing
to achieve SVR who received the shorter course. These
patients would require retreatment at a higher cost to
the patients and the health care system in terms of not
only drug cost but also treatment duration, potential
for toxicity and drug resistance, and overall health care
resources. Given current resource constraints, the high
cost of LDV/SOF, and emerging data regarding
retreatment options, additional study will be needed to
assess whether from a health care system perspective it
may be more cost-effective to treat all patients who
qualify for an 8-week course with an 8-week course
and retreat the small proportion who may fail.
In clinical trials, the stage of liver disease is often
determined by biopsy or, more recently, transient elas-
tography. In clinical practice, however, practitioners
may conclude that a patient does not have cirrhosis
based on biopsy, transient elastography, FIB-4, APRI,
imaging, or platelet count depending on the available
resources and information. This variability means that
guidance recommending specific treatment for those
“without cirrhosis” may be applied in heterogeneous
ways. In the VA, few patients undergo liver biopsy and
transient elastography is not yet widely available, nor
are such results captured in standardized data fields in
the electronic medical record. Thus, in our analysis
advanced fibrosis or cirrhosis was assessed using
BACKUS ET AL. HEPATOLOGY, August 2016
412
laboratory markers which are commonly used in clini-
cal practice.
(9,10,22)
In multivariate models, the pres-
ence of advanced liver disease as assessed by these
simple laboratory tests independently predicted
reduced odds of achieving SVR by nearly half. In the
resource-restrained environment surrounding HCV,
our study suggests that simple calculations of FIB-4 or
APRI from inexpensive laboratory tests can be useful
in discussions with patients regarding the likelihood of
treatment success. These same inexpensive, readily
available laboratory markers can be used in a wide vari-
ety of settings to determine whether a patient might
meet criteria for shortened 8-week treatment.
In this analysis, 4-week viral kinetics predicted SVR
with a greater effect in those who received LDV/SOF.
Significant reductions of 10.5% and 7.1% in SVR rates
were observed in patients receiving LDV/SOF and
LDV/SOF1RBV, respectively, who had a detectable
4-week HCV RNA 15 IU/mL compared to those
with undetectable 4-week HCV RNA. In patients
who completed 12-week courses of LDV/SOV there
remained a significant 6.4% reduction in SVR rates
between those with detectable 4-week HCV RNA
15 IU/mL and those with undetectable 4-week
HCV. For those who completed 12 weeks of LDV/
SOF1RBV the difference in SVR rates was no longer
statistically significant. In multivariable analysis, hav-
ing a HCV RNA 15 IU/mL after 4 weeks of treat-
ment was associated with a 60% reduced odds of
achieving SVR for the cohort and a 62% reduced odds
of achieving SVR for those who completed 12 weeks
of treatment. The VA guidance recommends obtaining
4-week HCV RNA testing, thus providing a large
sample size to evaluate this variable which has not gen-
erally been assessed in prior LDV/SOF studies. The
clinical implications of this finding on treatment deci-
sions, such as potentially adding RBV or extending
treatment duration based on 4-week on-treatment
HCV RNA, warrant further study.
While this study includes a large cohort of diverse
patients treated in clinical practice, there are limitations.
Due to the nature of the electronic data, we could not
determine the original intended duration of treatment.
It is possible that some patients who received 12 weeks
of treatment were originally planned to receive 8 weeks
and vice versa. In addition, specific reasons for early dis-
continuation (i.e., adverse events or poor tolerability)
could not be determined. Treatment duration and early
treatment discontinuation rates were determined based
on the cumulative dispensed days’ supply. This may
overestimate the treatment duration as patients may
have discontinued treatment even with medication in
their possession. In the VA, many prescriptions are
filled for small quantities (e.g., 7-14 days at a time),
which would limit the extent of the overestimation. Few
patients underwent IL28B testing, and no patients
underwent resistant associated variant testing; thus, we
were unable to assess the impact of these factors. From
the univariate analysis, IL28B did not appear to have an
impact on SVR rates. Approximately 6.8% (297/4,365)
of the cohort lacked definitive laboratory data to deter-
mine SVR status and may represent a potential source
of bias, although all were undetectable on their most
recent HCV RNA occurring while still on treatment or
less than 10 weeks after the EOT. If all of these patients
in fact had SVR, the maximum overall SVR rate for
LDV/SOF would be 91.9% and that for LDV/
SOF1RBV would be 92.4%, similar to the reported
SVR rates of 91.3% and 92.0%, respectively. With this
scenario significant predictors of SVR do not change
and the ORs in multivariate models are not substantially
altered (data not shown).
In conclusion, in this large real-world cohort,
treatment-naive, genotype 1 HCV-infected veterans
treated with LDV/SOF-based therapy achieved high
SVR rates, comparable to clinical trials, which were
consistently high across all subgroups evaluated. SVR
rates in patients without cirrhosis and with low base-
line HCV RNA treated for 8 weeks were significantly
lower than those achieved with 12 weeks of treatment,
although the numeric difference was small. The real-
world experience from large cohorts is essential to pro-
vide practical information to better inform and refine
HCV management strategies.
REFERENCES
1) Backus LI, Belperio PS, Shahoumian TA, Cheung R, Mole LA.
Comparative effectiveness of the hepatitis C virus protease inhib-
itors boceprevir and telaprevir in a large U.S. cohort. Aliment
Pharmacol Ther 2014;39:93-103.
2) Hezode C, Fontaine H, Dorival C, Larrey D, Zoulim F, Canva V,
et al. Triple therapy in treatment-experienced patients with HCV-
cirrhosis in a multicenter cohort of the French Early Access Pro-
gramme (ANRS CO20-CUPIC)-NCT01514890. J Hepatol 2013;
59:434-441.
3) Price JC, Murphy RC, Shvachko VA, Pauly MP, Manos MM.
Effectiveness of telaprevir and boceprevir triple therapy for
patients with hepatitis C virus infection in a large integrated care
setting. Dig Dis Sci 2014;59:3043-3052.
4) Gordon SC, Muir AJ, Lim JK, Pearlman B, Argo CK, Ramani
A, et al. Safety profile of boceprevir and telaprevir in chronic
hepatitis C: real world experience from HCVTARGET.
J Hepatol 2015;62:286-293.
HEPATOLOGY, Vol. 64, No. 2, 2016 BACKUS ET AL.
413
5) Backus LI, Belperio PS, Shahoumian TA, Loomis TP, Mole
LA. Effectiveness of sofosbuvir-based regimens in genotype 1
and 2 hepatitis C virus infection in 4026 U.S. veterans. Aliment
Pharmacol Ther 2015;42:559-573.
6) Sulkowski MS, Vargas HE, Di Bisceglie AM, Kuo PA, Reddy
KR, Lim JK, et al. Effectiveness of simeprevir plus sofosbuvir,
with or without ribavirin, in real-world patients with HCV geno-
type 1 infection. Gastroenterology 2016;150:419-429.
7) Shiffman ML, James AM, Long AG, Alexander PC. Treatment
of chronic HCV with sofosbuvir and simeprevir in patients with
cirrhosis and contraindications to interferon and/or ribavirin. Am
J Gastroenterol 2015;110:1179-1185.
8) Afdhal N, Zeuzem S, Kwo P, Chojkier M, Gitlin N, Puoti M,
et al. Ledipasvir and sofosbuvir for untreated HCV genotype 1
infection. N Engl J Med 2014;370:1889-1898.
9) Kowdley KV, Gordon SC, Reddy KR, Rossaro L, Bernstein DE,
Lawitz E, et al. Ledipasvir and sofosbuvir for 8 or 12 weeks for
chronic HCV without cirrhosis. N Engl J Med 2014;370:1879-1888.
10) Backus LI, Belperio PS, Loomis TP, Yip GH, Mole LA. Hepa-
titis C virus screening and prevalence among US veterans in
Department of Veterans Affairs care. JAMA Intern Med 2013;
173:1549-1552.
11) Belperio PS, Backus LI, Ross D, Neuhauser MM, Mole LA. A
population approach to disease management: hepatitis C direct-
acting antiviral use in a large health care system. J Manag Care
Pharm 2014;20:533-540.
12) Backus LI, Gavrilov S, Loomis TP, Halloran JP, Phillips BR,
Belperio PS, et al. Clinical case registries: simultaneous local and
national disease registries for population quality management.
J Am Med Inform Assoc 2009;16:775-783.
13) Sterling RK, Lissen E, Clumeck N, Sola R, Correra MC,
Montaner J, et al. Development of a simple non-invasive index
to predict significant fibrosis in patients with HIV/HCV coinfec-
tion. HEPATOLOGY 2007;43:1317-1325.
14) Vallet-Pichard A, Mallet V, Nalpas B, Verkarre V, Nalpas A,
Dhalluin-Venier V, et al. FIB-4: an inexpensive and accurate
marker of fibrosis in HCV infection. Comparison with liver
biopsy and FibroTest. HEPATOLOGY 2007;46:32-36.
15) Chou R, Wasson N. Blood tests to diagnose fibrosis or cirrhosis
in patients with chronic hepatitis C virus infection: a systematic
review. Ann Intern Med 2013;158:807-820.
16) Wilder JM, Jeffers LJ, Ravendhran N, Shiffman ML, Poulos J,
Sulkowski MS, et al. Safety and efficacy of ledipasvir-sofosbuvir
in black patients with hepatitis C virus infection: a retrospective
analysis of phase 3 data. HEPATOLOGY 2016;63:437-444.
17) Ioannou GN, Beste LA, Green PK. Similar effectiveness of
boceprevir and telaprevir treatment regimens for hepatitis C virus
infection, based on a nationwide study of veterans. Clin Gastro-
enterol Hepatol 2014;12:1371-1380.
18) Ghany MG, Strader DB, Thomas DL, Seeff LB; American
Association for the Study of Liver Diseases. Diagnosis, manage-
ment, and treatment of hepatitis C: an update. HEPATOLOGY
2009;49:1335-1374.
19) Sterling RK, Kuo A, Rustgi VK, Sulkowski MS, Stewart TG,
Fenkel JM, et al. Virological outcomes and treatment algorithms
utilisation in observational study of patients with chronic hepati-
tis C treated with boceprevir or telaprevir. Aliment Pharmacol
Ther 2015;41:671-685.
20) Thomas DL, Thio CL, Martin MP, Qi Y, Ge D, O’Huigin C,
et al. Genetic variation in IL28B and spontaneous clearance of
hepatitis C virus. Nature 2009;461:798-801.
21) Harvoni (ledipasvir and sofosbuvir) tablet product information.
Foster City, CA: Gilead Sciences, Inc.; 2015.
22) AASLD/IDSA HCV Guidance Panel. Hepatitis C guidance:
AASLD-IDSA recommendations for testing, managing, and
treating adults infected with hepatitis C virus. HEPATOLOGY
2015;62:932-954. http://www.hcvguidelines.org/full-report-view.
Supporting Information
Additional Supporting Information may be found at
onlinelibrary.wiley.com/doi/10.1002/hep.28625/suppinfo.
BACKUS ET AL. HEPATOLOGY, August 2016
414
