Nephrol Dial Transplant (2017) 32: 2043–2051
doi: 10.1093/ndt/gfw321
Advance Access publication 17 September 2016
Randomized multicentre pilot study of sacubitril/valsartan
versus irbesartan in patients with chronic kidney disease:
United Kingdom Heart and Renal Protection
(HARP)- III—rationale, trial design and baseline data
UK HARP-III Collaborative Group
Correspondence and offprint requests to: Richard Haynes, Medical Research Council Population Health Research
Unit, Nuffield Department of Population Health, Richard Doll Building, Old Road Campus, Roosevelt Drive, Oxford
OX3 7LF, UK; E-mail: harp3@ndph.ox.ac.uk
ABSTRACT
Background. Patients with chronic kidney dise ase (CKD) are
at risk of progression to end-stage renal disease and cardio-
vascular disease. Data from other populations and animal
experiments sugge st that neprilysin inhibition (which aug-
ments the natriuretic peptide system) may reduce these risks,
but clinical trials among patients with CKD are required to
test this hypothesis.
Methods. UK Heart and Renal Protection III (HARP-III) is a
multicentre, double-blind, randomized controlled trial compar-
ing sacubitril/valsartan 97/103 mg two times daily (an angioten-
sin receptor–neprilysin inhibitor) with irbesartan 300 mg one
time daily among 414 patients with CKD. Patients 18 years of
age with an estimated glomerular filtration rate (eGFR) of 45
but <60 mL/min/1.73 m
2
and urine albumin:creatinine ratio
(uACR) >20 mg/mmol or eGFR 20 but <45 mL/min/1.73 m
2
(regardless of uACR) were invited to be screened. Following a 4-
to 7-week pre-randomization single-blind placebo run-in phase
(during which any current renin–angiotensin system inhibitors
were stopped), willing and eligible participants were randomly
assigned either sacubitril/valsartan or irbesartan and followed-
up for 12 months. The primary aim was to compare the effects
of sacubitril/valsartan and irbesartan on measured GFR after 12
months of therapy. Important secondary outcomes include
effects on albuminuria, change in eGFR over time and the safety
and tolerability of sacubitril/valsartan in CKD.
Results. Between November 2014 and January 2016, 620
patients attended a screening visit and 566 (91%) entered the
pre-randomization run-in phase. Of these, 414 (73%) partici-
pants were randomized (mean age 63 years; 72% male). The
mean eGFR was 34.0 mL/min/1.73 m
2
and the median uACR
was 58.5 mg/mmol.
Conclusions. UK HARP-III will provide important information
on the short-term effects of sacubitril/valsartan on renal func-
tion, tolerability and safety among patients with CKD.
Keywords: cardiovascular disease, chronic kidney disease,
neprilysin, progression
INTRODUCTION
Chronic kidney disease (CKD) affects between 2 and 17% of the
general population (depending on the country) [
1, 2]andis
associated with increased risks of progression to end-stage renal
disease (ESRD) and morbidity and mortality from cardiovascu-
lar disease (CVD) [
3, 4]. Renin–angiotensin system (RAS)
inhibitors [angiotensin-converting enzyme inhibitors (ACEis)
and angiotensin receptor blockers (ARBs)] have been shown to
reduce the risk of ESRD in patients with proteinuric CKD [
5–
8], but despite such treatments, patients remain at significant
risk of progression to ESRD and CVD.
The natriuretic peptide (NP) system is a neurohormonal sys-
tem that has a variety of potentially beneficial functions, includ-
ing natriuresis, diuresis, vasodilatation and counterregulation of
RAS [
9, 10]. The NP system can be augmented by inhibiting the
main enzyme responsible for degrading NPs, namely neprilysin
[or neutral endopeptidase (NEP)] [
10]. NEP is a membrane-
bound zinc-containing metalloproteinase [
11] that also degrades
other peptides, including angiotensin II, bradykinin, endothelin
and substance P [
12]. However, isolated NEP inhibition (NEPi)
leads to reflex RAS activation, and inhibits angiotensin II break-
down (counteracting any potentially beneficial effects) and there-
fore NEPi must be combined with RAS inhibition.
As NEPi and ACEi both inhibit bradykinin degradation, their
combination is associated with substantially elevated bradykinin
V
C
The Author 2016. Published by Oxford University Press on behalf of ERA-
EDTA.
This is an Open Access article distributed under the terms of the Creative
Commons Attribution License (http://creativecommons.org/licenses/by/4.0/),
which permits unrestricted reuse, distribution, and reproduction in any medium,
provided the original work is properly cited.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2043
Downloaded from https://academic.oup.com/ndt/article/32/12/2043/3059462 by guest on 21 August 2022
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
levels that cause unacceptable rates of angioedema [13 ]. ARBs do
not inhibit bradykinin degradation and can be safely combined
with NEPi [creating a new class of drugs called angiotensin
receptor–neprilys in inhibitors (ARNis)]. Sacubitril/valsartan
(previously known as LCZ696) is the first drug in this new class,
combining valsartan with sacubitril [(AHU377) a prodrug that is
metabolized via esterases to the active NEPi sacubitrilat
(LBQ657)]. Sacubitril/valsartan 97/103 mg provides equivalent
plasma concentrations of valsartan as oral valsartan 160 mg [
14].
In a 5/6 nephrectomy model, treatment with combined
NEP/RAS inhibition was associated with greater reductions in
proteinuria and glomerulosclerosis compared with RAS inhibi-
tion alone [
15, 16]. Micropuncture studies also demonstrated
NEPi led to greater reductions in capillary glomerular pressure
[
15]. Among patients with heart failure, trials comparing sacu-
bitril/valsartan with either ACEi or ARB have suggested that the
estimated glomerular filtration rate (eGFR) of patients allocated
sacubitril/valsartan declined less than those assigned ACEi or
ARB [
17, 18]. Sacubitril/valsartan also reduced blood pressure
more than equivalent doses of valsartan in trials among patients
with elevated blood pressure [
19]. Trials in heart failure popula-
tions suggest NEPi might increase albuminuria [
18, 20], but this
effect was not observed in patients with hypertension [
19]and
baseline albuminuria was very low in all these trials. Overall,
these data raise the hypothesis that treatment with an ARNi
may be superior to either ACEi or ARB alone in slowing the
progression of CKD.
The United Kingdom (UK) Heart and Renal Protection
III(HARP-III) trial (ISRCTN11958993) was designed to pro-
vide information on the short-term efficacy (in terms of effect
on renal function), tolerability and safety of sacubitril/valsartan
among patients with CKD. The trial will also assess the effects
of sacubitril/valsartan on albuminuria, blood pressure and bio-
markers of kidney and cardiac damage.
MATERIALS AND METHODS
Study design
UK HARP-III is a double-blind, multicentre, randomized
controlled trial comparing sacubitril/valsartan 97/103 mg two
times daily versus irbesartan 300 mg one time daily among at
least 400 participants 18 years of age with stages 3 and 4 CKD.
Irbesartan 300 mg was selected as the comparator, as it has been
shown to reduce the risk of ESRD among patients with diabetic
kidney disease and is licensed for the treatment of proteinuric
CKD [
6, 21]. Participants were randomly allocated to receive
sacubitril/valsartan or irbesartan and will be followed up for 1
year (Figure
1). The primary aim of UK HARP-III is to assess
the effect of sacubitril/valsartan 97/103 mg two times daily ver-
sus irbesartan 300 mg one time daily on measured glomerular
filtration rate (mGFR) at 12 months. Important secondary out-
comes include the effect on urine albumin:creatinine ratio
(uACR) and eGFR. All the secondary and tertiary assessments
are shown in Figure
2 and further details are available in the
data analysis plan (see
Supplementary data). A summary of sub-
stantial amendments to the protocol is provided in the
Supplementary data.
Eligibility
To fulfil the inclusion criteria, patients need to be 18 years
of age and have either an eGFR 45 but <60 mL/min/1.73 m
2
with a uACR >20 mg/mmol or eGFR 20 but <45 mL/min/
1.73 m
2
(regardless of uACR). The exclusion criteria were
designed to identify patients for whom the safety of sacubitril/
valsartan or irbesartan may have been a concern. The full eligi-
bility criteria are shown in Figure
3.
FIGURE 1: UK HARP-III trial design.
2044 The UK HARP-III Collaborative Group
Downloaded from https://academic.oup.com/ndt/article/32/12/2043/3059462 by guest on 21 August 2022
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Study enrolment and randomization
Identification and invitation.
After relevant ethics
[Nottingham Research Ethics Committee 2 (13/EM/0434)] and
regulatory approvals had been obtained, sites were established
in UK renal units. Site staff identified potentially eligible
patients from hospital electronic databases, mailed these indi-
viduals an invitation letter and a copy of the patient information
sheet and called them 1 week later to discuss the trial in more
detail, answer any questions they might have and to see whether
they were interested in participating. Those individuals inter-
ested in participating were invited to attend a screening visit.
Screening. At the screening visit, eligibility was assessed and
written informed consent was obtained from eligible individu-
als. All data were recorded directly into a bespoke Internet-
based electronic case report form system. Relevant details of
their medical history (including primary renal diagnosis, pres-
ence of diabetes mellitus and prior CVD) were recorded by
trained research nurses and their height, weight and blood pres-
sure were measured. Blood pressure was measured and
recorded three times using an Omron M6 automated digital
sphygmomanometer after sitting for at least 5 minutes. Willing
and eligible patients entered the pre-randomization run-in
phase. Samples of blood and urine were sent to the local hospital
laboratory for confirmation of eligibility. If the results were con-
sidered inaccurate (e.g. haemolysed sample) by the local study
staff the samples could be repeated once, but if the results did
not confirm eligibility the participant was withdrawn from the
run-in phase.
Pre-randomization run-in. The aims of the pre-
randomization run-in phase were (i) to ‘wash out’ any ACEi
prior to introduction of NEPi, (ii) to allow a comparison of the
acute effects of the study treatments on GFR and (iii) to reduce
the rate of post-randomization discontinuation of study treat-
ment and to produce a consequent improvement in the trial’s
statistical sensitivity [
22]. Following the screening visit, any cur-
rent ACEi and/or ARB that the participant was taking was
stopped and the participant entered the 4- to 7-week single-
blind pre-randomization run-in phase, during which they were
asked to take one placebo sacubitril/valsartan tablet and one
placebo irbesartan capsule once daily. If elevated blood pressure
became a concern during the run-in phase, local investigators
were advised to titrate up or start additional anti-hypertensive
medications, but to avoid an ACEi, ARB or direct renin inhibi-
tor (DRI). The choice of additional anti-hypertensive therapy
remained at the discretion of the responsible clinician.
Participants could withdraw from the trial for any reason dur-
ing this run-in phase. Participants who did not withdraw
returned 4–7 weeks later and had their GFR measured and
attended a randomization visit. GFR was measured using a
standard
51
Cr-EDTA technique, although if this was not avail-
able at the site, other methods (
99m
Tc-DTPA or iohexol) could
be used with the agreement of the coordinating centre. In will-
ing participants, a 24-hour collection of urine for albumin and
sodium quantification was also obtained.
Randomization visit. Participants were not eligible for ran-
domization if the mean of their second and third measurements
of systolic blood pressure was <110 mmHg (or <130 mmHg
withsymptomsofhypotension)oriftheyreportedanadverse
event they believed to be related to their run-in treatment.
Participants who remained willing and eligible were then ran-
domly allocated in a 1:1 ratio to receive either sacubitril/valsar-
tan or irbesartan. Participants were randomized by an Internet-
based system using a minimization algorithm to ensure balance
of important predictors of renal progression, including age, sex,
systolic blood pressure, eGFR, uACR and the presence or
absence of diabetes mellitus.
At the randomization visit, run-in treatment was collected
and willing and eligible participants were issued two bottles of
study treatments: one containing sacubitril/valsartan 97/103 mg
or placebo tablets and the other containing irbesartan 150 mg
or placebo capsules (therefore a double-dummy technique to
protect blinding). Participants were initially instructed to take
one tablet and one capsule daily in the morning (i.e. either sacu-
bitril/valsartan 97/103 mg plus placebo irbesartan or placebo
sacubitril/valsartan plus irbesartan 150 mg). Blood and urine
samples were collected for the local analysis of creatinine, elec-
trolytes, liver function tests and uACR and others were pre-
pared for central analysis (Table
1).
Post-randomization follow-up
Randomization is now complete and all participants are in
follow-up. In order to check potassium and renal function after
starting study treatment, participants attend their study clinic
or local primary care physician at 2 weeks after randomization
for a blood sample. If these results are satisfactory, study treat-
ments are increased to either sacubitril/valsartan 97/103 mg two
times daily plus two capsules of placebo irbesartan one time
daily or one tablet of placebo sacubitril/valsartan two times daily
plus irbesartan 300 mg one time daily.
Follow-up assessments. Study follow-up visits are scheduled
at 1, 3, 6, 9 and 12 months after randomization. At all visits,
study staff systematically seek the information on all serious
adverse events, any non-serious adverse events considered by
participants to be related to study treatment; and on symptoms
FIGURE 2: UK HARP-III trial outcomes.
The UK HARP-III Collaborative Group 2045
Downloaded from https://academic.oup.com/ndt/article/32/12/2043/3059462 by guest on 21 August 2022
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
of hepatitis. Compliance with study treatment is assessed and
participants unable to tolerate the maximum dose of study
treatments are encouraged to continue on the lower dose of
study drug (i.e. sacubitril/valsartan 97/103 mg or irbesartan 150
mg daily) for the remainder of the trial. If relevant, a reason for
discontinuation or dose reduction is recorded. Participants pre-
scribed contraindicated medications (ACEi, ARB or DRI) have
their randomized treatment stopped. Weight and blood pres-
sure are measured (three times after sitting for at least 5
minutes) at all visits. In both treatment groups, blood pressure
is to be controlled according to the Kidney Disease: Improving
Global Outcomes guidelines [
23], with the initiation and choice
of additional anti-hypertensive treatment being at the discretion
of the responsible clinician. Within the 2 weeks before their 12-
month visit participants have their second GFR measurement
(using the same method as at baseline). Copies of results of both
measurements of GFR are sent to the coordinating centre so the
results entered by site staff can be verified by clinical study staff
blind to the treatment allocation.
Biological samples and safety monitoring. At each follow-
up visit, blood and urine samples are sent to the local hospital
laboratory for creatinine, electrolytes, liver function tests (biliru-
bin, alanine or aspartate transaminase and alkaline phospha-
tase) and uACR. In addition, at the 3-, 6- and 12-month visits,
samples are also taken for central analysis. EDTA samples are
centrifuged and the plasma aliquoted into Cryovials, which are
stored locally (with Cryovials of urine) at or below 20
Cprior
to transfer to the central laboratory in Oxford, UK, where they
are stored at 80
C. The main plasma analytes measured at the
central laboratory are creatinine, cardiac and inflammatory bio-
markers and the urine analytes include albumin and markers of
tubular damage and function [including kidney injury molecule
1, neutrophil gelatinase-associated lipocalin, b2-microglobulin
and retinol binding protein; Table
1]. Participants are asked not
to take their morning dose of study treatment on the day of
their 3-month visit (at this visit only) and the date and time of
the last dose is recorded, as these samples are to be used for
pharmacokinetic analyses.
FIGURE 3: Inclusion and exclusion criteria. ALT, alanine aminotransferase; AST, aspartate aminotransferase; ULN, upper limit of normal.
2046 The UK HARP-III Collaborative Group
Downloaded from https://academic.oup.com/ndt/article/32/12/2043/3059462 by guest on 21 August 2022
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
The results of local samples are entered into the trial data-
base once available and reviewed daily by a trained clinician at
the coordinating centre. If the potassium is >5.5 mmol/L, ala-
nine or aspartate transaminase >2 the upper limit of normal
or if the eGFR has fallen >25% from the previous value, then
the trial protocol provides advice on further tests and study
treatment (see
Supplementary data).
Monitoring
Prior to starting recruitment, study staff received training in
the study procedures and the web-based data collection system
at the coordinating centre. Recruitment rates, adherence to trial
procedures and completeness of follow-up data are monitored
closely by staff at the coordinating centre. All sites have at least
one on-site monitoring visit, with further visits as indicated by
the results of central monitoring of the data. An independent
data monitoring committee (see
Supplementary data)regularly
reviews unblinded interim analyses of all relevant data.
Statistical considerations
Sample size.
The chief aim of this study is to compare mGFR
between the two treatment groups at the final follow-up visit.
Analysis of covariance (ANCOVA) compares mean follow-up
mGFR between treatment groups after adjustment for baseline
mGFR [
24]. Assuming a between-person standard deviation
(SD) in mGFR of 15 mL/min/1.73 m
2
and a correlation between
an individual’s baseline and follow-up mGFR of 0.8, randomiza-
tion of 400 participants will provide at least 80% power (at 2 P
¼ 0.05) to detect a difference in mGFR at the final follow-up
(adjusted for baseline values) of 3 mL/min/1.73 m
2
(the chosen
minimum clinically meaningful difference), even if 15% of par-
ticipants discontinue allocated study treatment [
20].
Statistical analysis. All analyses will involve comparing out-
comes during the scheduled treatment period among all those
participants allocated at randomization to receive sacubitril/
valsartan 97/103 mg two times daily versus all those allocated to
receive irbesartan 300 mg one time daily [i.e. intention-to-treat
(ITT) analyses] [
25, 26]. Comparisons of continuous outcomes
(including the primary outcome) between the allocated treat-
ment arms will be performed using ANCOVA adjusted for each
patient’s value at baseline [
27]. If continuous outcomes are not
normally distributed, then appropriate transformations (e.g. log
transformation) will be made. Multiple imputation techniques
will be used to account for any missing data in the primary and
secondary outcomes [
28]. Further details are provided in the
data analysis plan (see
Supplementary data).
RESULTS
Study sites were established in 24 renal units in the UK.
Between November 2014 and January 2016 a total of 620
patients attended the study screening visits and 566 (91%)
entered the pre-randomization run-in (Figure
4).
Pre-randomization run-in
A total of 138 participants withdrew from the pre-
randomization run-in before attending a randomization visit
(Table
2A). The most common medical reason for withdrawal
from run-in was that the results from blood and urine samples
taken at the screening visit did not confirm the participant’s eligibil-
ity (Table
2A). Adverse events were uncommon and four partici-
pants were withdrawn because of a serious adverse event
(myocardial infarction, septic shock and two cases of pneumonia).
In addition, 14 individuals attended a randomization visit
but were not eligible to be randomized: the most common rea-
son for this was their blood pressure being too low (Table
2B).
Overall, 152 (27%) of the 566 individuals who entered the pre-
randomization single-blind placebo run-in phase were not sub-
sequently randomized.
Baseline characteristics of randomized participants
A tot al of 414 people were randomized (Figure 4). The mean
age was 63 (SD 14) years and 298 (72%) were male (Table
3). The
mean systolic blood pressure was 146 (SD 16) mmHg at random-
ization (i.e. after 4–7 weeks of withdrawal of any prior ACEi or
ARB). Based on results from the local laborato ries, the mean
eGFR was 34.0 (SD 10.6) mL/min/1.73 m
2
and the median uACR
was 58.5 (interquartile range 12.5–156.3) mg/mmol. Central lab-
oratory assays will be conducted at the end of the study. About
half of randomized participants had either glomerular [111
(27%)] or diabetic [83 (20 %)] kidney disease and 165 (40%)
patients reported diabetes mellitus at baseline. The median 5-year
risk of ESRD (calculated using a validated risk calculator [
29]was
16.5%, and 62% of participants had a 5-year risk >10%.
DISCUSSION
The UK HARP-III trial has recruited 414 participants with
CKD and will provide information on the short-term effects of
sacubitril/valsartan on the change in kidney function (using
mGFR) and the tolerability and safety of the drug compared
Table 1. Planned central laboratory blood and urine analyses
Analyte Time point
Randomization 3
months
6
months
12
months
EDTA plasma samples
Creatinine 䊏䊏䊏䊏
Albumin 䊏䊏䊏
Troponin-I 䊏䊏䊏
NT-proBNP 䊏䊏䊏
CRP 䊏䊏䊏
IL-6 䊏䊏䊏
Urine samples
Albumin:creatinine
ratio
䊏䊏䊏䊏
KIM-1 䊏䊏䊏
NGAL 䊏䊏䊏
cGMP 䊏䊏䊏
b-2-microglobulin 䊏䊏䊏
Retinol binding protein 䊏䊏䊏
NT-proBNP, N-terminal prohormone brain natriuretic peptide; CRP, C-reactive protein;
IL-6, interleukin 6; KIM-1, kidney injury molecule 1; NGAL, neutrophil gelatinase-asso-
ciated lipocalin; cGMP, cyc lic guanosine monophosphate.
The UK HARP-III Collaborative Group 2047
Downloaded from https://academic.oup.com/ndt/article/32/12/2043/3059462 by guest on 21 August 2022