Scholarly article on topic 'Treatment with patiromer decreases aldosterone in patients with chronic kidney disease and hyperkalemia on renin-angiotensin system inhibitors'

Treatment with patiromer decreases aldosterone in patients with chronic kidney disease and hyperkalemia on renin-angiotensin system inhibitors Academic research paper on "Health sciences"

CC BY-NC-ND
0
0
Share paper
Academic journal
Kidney Int
OECD Field of science
Keywords
{""}

Academic research paper on topic "Treatment with patiromer decreases aldosterone in patients with chronic kidney disease and hyperkalemia on renin-angiotensin system inhibitors"

Treatment with patiromer decreases aldosterone in patients with chronic kidney disease and hyperkalemia on renin-angiotensin system inhibitors

Matthew R. Weir1, George L. Bakris2, Coleman Gross3, Martha R. Mayo3, Dahlia Garza3, Yuri Stasiv3, Jinwei Yuan3, Lance Berman3 and Gordon H. Williams4

'Division of Nephrology, Department of Medicine, University of Maryland, Baltimore, Maryland, USA; 2University of Chicago, Chicago, Illinois, USA; 3Relypsa, Redwood City, California, USA; and 4Brigham and Women's Hospital/Harvard Medical School, Boston, Massachusetts, USA

Elevated serum aldosterone can be vasculotoxic and facilitate cardiorenal damage. Renin-angiotensin system inhibitors reduce serum aldosterone levels and/or block its effects but can cause hyperkalemia. Patiromer, a nonabsorbed potassium binder, decreases serum potassium in patients with chronic kidney disease on renin-angiotensin system inhibitors. Here we examined the effect of patiromer treatment on serum aldosterone, blood pressure, and albuminuria in patients with chronic kidney disease on renin-angiotensin system inhibitors with hyperkalemia (serum potassium 5.1-6.5 mEq/l). We analyzed data from the phase 3 OPAL-HK study (4-week initial treatment phase of 243 patients; 8-week randomized withdrawal phase of 107 patients). In the treatment phase, the (mean ± standard error) serum potassium was decreased concordantly with the serum aldosterone ( —1.99 ± 0.51 ng/dl), systolic/diastolic blood pressure (-5.64 ± 1.04 mm Hg/ —3.84 ± 0.69 mm Hg), and albumin-to-creatinine ratio (— 203.7 ± 54.7 mg/g), all in a statistically significant manner. The change in the plasma renin activity ( — 0.44 ± 0.63 mg/l/hr) was not significant. In the withdrawal phase, mean aldosterone levels were sustained with patiromer ( + 0.23 ± 1.07 ng/dl) and significantly increased with placebo ( + 2.78 ± 1.25 ng/dl). Patients on patiromer had significant reductions in mean systolic/

diastolic blood pressure (-6.70 ± 1.59/-2.15 ± 1.06 mm Hg), whereas those on placebo did not (-1.21 ± 1.89 mm Hg/+1.72 ± 1.26 mm Hg). Significant changes in plasma renin activity were found only in the placebo group (-3.90 ± 1.41 mg/l/hr). Thus, patiromer reduced serum potassium and aldosterone levels independent of plasma renin activity in patients with chronic kidney disease and hyperkalemia on renin-angiotensin system inhibitors.

Kidney International (2016) ■, http://dx.doi.org/10.1016/ j.kint.2016.04.019

Correspondence: M.R. Weir, N3W143 Nephrology, University of Maryland Medical Center, 22 S. Greene Street, Baltimore, Maryland 21201, USA. E-mail: mweir@medicine.umaryland.edu

Received 17 December 2015; revised 29 March 2016; accepted 4 April 2016

KEYWORDS: ACE inhibitors; aldosterone; cardiovascular disease; chronic kidney disease; renin angiotensin system

Copyright © 2016, International Society of Nephrology. Published by Elsevier Inc. All rights reserved.

Because of the increased prevalence of hyperkalemia (HK) in patients with renal dysfunction, novel agents have recently been developed to reduce serum potassium (sK) levels by binding potassium in the gastrointestinal tract.1 Although these agents have produced the intended beneficial effect of reducing potassium levels,2 there may be added benefits from this form of therapy—a reduction in aldosterone production.

Higher levels of serum aldosterone are associated with increased mineralocorticoid receptor stimulation; this ligand-receptor interaction has been well demonstrated to promote cardiovascular and renal disease progression in experimental and clinical studies.3-9 By blocking this interaction, miner-alocorticoid receptor antagonists have demonstrated benefits in patients with heart failure and post-acute myocardial infarction.7 Moreover, blocking the mineralocorticoid receptor is associated with reductions in blood pressure (BP) and albuminuria,8 suggesting that this type of therapeutic approach may reduce the rate of renal disease progression. Recent studies have also suggested that mineralocorticoid receptor activation may lead to metabolic dysregulation and susceptibility to type 2 diabetes mellitus and atheroscle-rosis.10-13

Consequently, there is great interest in reducing the adverse effects of aldosterone, which can be accomplished with mineralocorticoid receptor antagonists.7'14 Yet, the use of mineralocorticoid receptor antagonists in patients with renal dysfunction is associated with a substantial incidence of HK, which limits their use either alone or combined with angiotensin-converting enzyme inhibitors or angiotensin receptor blockers.15-19 For example, in 2009, Khosla et al.20 reported a rate of HK (potassium >5.5 mEq/l) of 34.7% in patients treated with a mineralocorticoid receptor antagonist who had a baseline serum potassium level >4.5 mEq/l and a glomerular filtration rate #45 ml/min per 1.73 m2. This limitation is a concern, especially in patients with congestive heart failure, with or without renal dysfunction, in whom the

effects of aldosterone reduction provide the greatest benefit.5-8

Aldosterone production is not only regulated by the renin-angiotensin-aldosterone system (RAAS), but also by potassium,5'21'22 which is equally potent in modulating aldosterone secretion.23-28 Thus, the possibility exists that agents that lower potassium levels may also reduce aldosterone secretion and may provide an added benefit beyond potassium reduction in patients with renal dysfunction.

To test this hypothesis, we performed a prespecified exploratory analysis of the OPAL-HK clinical trial in which patiromer, a novel potassium-binding polymer that uses calcium rather than sodium as the cation for exchange with potassium, was shown to decrease potassium levels in patients with chronic kidney disease (CKD) and HK who were being treated with a RAAS inhibitor (RAASi).2 Based on the pivotal OPAL-HK study,2 patiromer was recently approved by the US Food and Drug Administration for the treatment of HK. The unique design of the OPAL-HK study allowed us to evaluate changes in aldosterone, BP, and albuminuria according to baseline potassium level and in a placebo-controlled fashion to determine whether there are additional beneficial effects of patiromer beyond sK reduction.

RESULTS

Patient disposition and demographics

Overall, 243 patients were enrolled in the initial treatment phase, and 107 patients participated in the randomized withdrawal phase of the study (55 continued patiromer and 52 switched to placebo). The majority of patients had hypertension, diabetes mellitus, and advanced CKD. The baseline characteristics of patients in these analyses are shown in Table 1.

Initial treatment phase

Over the 4-week initial treatment phase, mean ± SE serum aldosterone levels decreased (-1.99 ± 0.51 ng/dl, P = 0.0001) in parallel with decreases in sK (Figure 1). When examined by baseline sK levels, only patients with moderate to severe HK had significant changes from baseline in mean ± SE serum aldosterone levels at week 4 (mild HK: -0.67 ± 0.80, P = not significant [NS]; moderate to severe HK: -3.32 ± 0.63, P <0.0001) (Table 2). As previously reported,2 patients with moderate to severe HK (5.5 to <6.5 mEq/l) at baseline had greater changes from baseline in sK compared with patients with mild HK (5.1 to <5.5 mEq/l) at baseline. Patients with mild HK had a mean decrease in sK from baseline to week 4 of -0.65 (0.05) mEq/l compared with -1.23 (0.04) mEq/l in patients with moderate to severe HK at baseline; both were statistically significant.

BP decreased over the initial 4-week treatment period in patients receiving patiromer (Figure 2). The mean ± SE systolic BP (SBP) change from baseline in patients with mild HK was -5.7 (1.6) mm Hg (P = 0.0005) and -5.5 (1.3) mm Hg (P <0.0001) in patients with moderate to severe HK (Table 2). Mean diastolic BP (DBP) changed from baseline by -3.2 (1.1)

Table 11 Baseline demographic and clinical characteristics

Characteristic

Initial treatment phase Overall (N = 243)

Randomized withdrawal phase

Placebo (N = 52)

Patiromer (N = 55)

Male sex, n (%) Age, yr, mean ± SD White race, n (%)a Type 2 diabetes, n (%) Heart failure, n (%) NYHA functional class, n (%)

Myocardial infarction, n (%) Hypertension, n (%) Serum potassium, mEq/l,

mean ± SDb Estimated GFR, ml/min per 1.73 m2, n (%)c

140 (58) 64.2 ± 10.5 239 (98) 139 (57) 102 (42)

19 (19) 66 (65) 17 (17) 60 (25) 236 (97) 5.6 0.5

30 (58) 65.0 ± 9.1 52 (100) 33 (63) 22 (42)

4 (18) 14 (64) 4 (18) 14 (27) 50 (96) 5.9 0.4

28 (51) 65.5 ± 9.4 55 (100) 34 (62) 27 (49)

5 (19) 18 (67) 4 (15) 18 (33) 54 (98) 5.9 0.6

Stage 2: 60 to <90 22 (9) 4 (8) 8 (15)

Stage 3A: 45 to <60 49 (20) 11 (21) 11 (20)

Stage 3B: 30 to <45 63 (26) 14 (27) 15 (27)

Stage 4/5: <30 109 (45) 23 (44) 21 (38)

RAASi use, n (%)c 243 (100) 52 (100) 55 (100)

ACE inhibitor 170 (70) 38 (73) 37 (67)

Angiotensin II receptor 92 (38) 16 (31) 24 (44)

blocker

MRA 22 (9) 4 (8) 4 (7)

Renin inhibitor 2 (1) 0 0

Dual RAAS blockaded 41 (17) 6 (12) 10 (18)

Receiving maximal dosee 106 (44) 21 (40) 21 (38)

Non-RAASi diuretic use, 132 (54) 27 (52) 28 (51)

n (%)c

Thiazide 70 (29) 11 (21) 16 (29)

Loop 77 (32) 20 (38) 16 (29)

Serum aldosterone, ng/dl, mean SDb

11.86 11.31 11.83 12.40 13.36 12.14

ACE, angiotensin-converting enzyme; GFR, glomerular filtration rate; MRA, mineral-ocorticoid receptor antagonist; NYHA, New York Heart Association; RAAS, renin-angiotensin-aldosterone system; RAASi, renin-angiotensin-aldosterone system inhibitor.

aRace was determined by the investigators.

bThe values in this row refer to baseline values at the start of the study.

cThe values for patiromer and placebo in this row refer to baseline values at the start

of the randomized withdrawal phase.

dDual RAAS blockade refers to any combination of $2 of the following: ACE inhibitor, angiotensin II receptor blocker, aldosterone antagonist, or renin inhibitor. eThe maximal dose was determined according to the judgment of the investigator in accordance with the local standard of care.

mm Hg in patients with mild HK (P = 0.003) and by -4.5 (0.9) mm Hg in patients with moderate to severe HK (P < 0.0001) (Table 2).

In the extended multivariate mixed model for repeated measures (MMRM) model, decreases in mean ± SE aldo-sterone level were associated with baseline serum aldosterone levels (P < 0.0001), age 65 years or older (P = 0.04), the presence of type 2 diabetes mellitus (P = 0.02), and the absence of heart failure (P = 0.04). There was no significant effect of sex and CKD stage on aldosterone reduction (Supplementary Table S1). After adjustment for baseline aldosterone and the previously cited covariates, reduction in

Figure 11 Observed mean values of serum aldosterone and potassium over time during the initial treatment phase. The observed mean values were measured in a central laboratory. Missing central laboratory serum potassium values were imputed from local laboratory data.

serum aldosterone in the moderate to severe HK group was still significant (P < 0.0001).

In the sensitivity analysis, the observed reduction in aldosterone concentrations remained significant after the data were adjusted for modification of antihypertensive agents (including RAASi and diuretics) and any use of mineralo-corticoid receptor antagonist at baseline or throughout the initial treatment phase (Table 3). There was a low incidence of change in antihypertensive medications in the initial treatment phase (Supplementary Table S2). Overall, 94% of patients (228/243) had no change in RAASi use; 92% (224/243) had no change in non-RAASi, nondiuretic use; and 97% (236/ 243) had no change in diuretic use. The reason for RAASi modification included heart failure, hypertension, hypotension, and HK.

At baseline, patients with mild HK had a mean ± SE serum plasma renin activity (PRA) of 11.15 (1.38) mg/l/hr and patients with moderate to severe HK had a mean ± SE serum PRA of 8.08 (0.92) mg/l/hr) (P = NS). At week 4 of the initial treatment phase, there were no statistically significant changes in mean ± SE PRA in the total population (—0.44 ± 0.63 mg/l/hr) or in patients with mild HK (—1.10 ± 0.99 mg/l/hr) or moderate to severe HK (0.22 ± 0.78 mg/l/hr) (Table 2).

There were decreases in mean ± SE albumin-to-creatinine ratio (ACR) from baseline to week 4 for the overall group (—203.7 ± 54.7 mg/g, P = 0.0003). The mean ACR decreased significantly in both the mild HK group (—174.3 ± 87.4 mg/g, P # 0.05) and the moderate to severe HK group (—222.1 ± 69.2 mg/g, P = 0.0015) (Table 2).

In order to explore whether serum aldosterone changes were associated with increased or decreased potassium, PRA, SBP, and DBP, we conducted 1-way analysis of variance analyses by aldosterone tertiles at baseline and by tertiles of aldosterone change from baseline at day 3 (early effect) and at week 4 (end of the initial treatment phase). There was no correlation between serum aldosterone tertiles and SBP and DBP at baseline (P = NS for both groups). SBP decreases were observed across all serum aldosterone changes from baseline tertiles at day 3 (P = NS) and week 4 (P = 0.034).

DBP decreases were observed across all tertiles of serum aldosterone change from baseline at day 3 (P = 0.067) and week 4 (P = NS) (Tables 4-6). Higher serum aldosterone levels (tertiles) at baseline were significantly associated with higher mean sK levels (P = 0.025), and greater decreases in serum aldosterone were significantly associated with greater decreases in sK at day 3 (P = 0.004) but not at week 4 (P = NS). Higher levels of aldosterone were correlated with higher mean PRA (P = 0.04) at baseline, but the tertiles of change in aldosterone levels were not associated with changes in PRA at day 3 and week 4.

Randomized withdrawal phase

At the start of the randomized withdrawal phase, differences in aldosterone levels between patients randomized to patiromer or placebo were not statistically significant. In patients continuing on patiromer, the mean ± SE serum aldosterone concentrations did not change substantially from the start (week 4 of the initial treatment phase) to week 8 of the randomized withdrawal phase (0.23 ± 1.07), whereas serum aldosterone concentrations increased in patients switched to placebo (2.78 ± 1.25, P # 0.03) (Table 7).

From the start of the randomized withdrawal phase to week 8, patients continuing on patiromer and who completed the withdrawal phase (N = 45) demonstrated statistically significant reductions in mean ± SE SBP (—6.70 ± 1.59 mm Hg, P <0.0001) and DBP (—2.15 ± 1.06 mm Hg, P # 0.05). Those who switched to placebo and completed the initial treatment and withdrawal phases (N = 29) did not demonstrate statistically significant changes in SBP (—1.21 ± 1.89 mm Hg, P = NS) or DBP (+1.72 ± 1.26 mm Hg, P = NS) (Table 7). There was a statistically significant decrease in PRA from the start of the randomized withdrawal phase to week 8 (—3.90 ± 1.41 mg/l/hr, P = 0.0067) in placebo patients; there were no statistically significant changes for those patients continuing on patiromer.

By the end of the randomized withdrawal phase, 94% of patients in the patiromer group were still receiving RAASi medication compared with 44% of patients in the placebo

Kidney International (2016)

<u en ^ +1 ae

§E O m

"sV LU

J? vÔ

S V rao

■c m

(u m ^ +1 ae

LU ® C

2 v Mio

v m ae

I I I I I

I I I I I

m o o oo

-H-H-H-H-H -H-H-H-H^

^r o am o

LO 00 LO -sT

^ O 00 ^ o

OJ to eu

2 15 -

O O O 00 o

o o o ^r o

o o o o o

-H -H -H -H ^

^ ^r^r ^r o

lû W ^ IN

^ LO ro O ro

^o m m ^r o 00 IN fN oo i-1 ^r 00 ci "¡r

m fN fN m

CN CN CN CN CN

I I I I I

I I I I I

ClC0fN(N

^ m ^ oi o ^ o o

OOhvC^O 03 CO N r^ ON LO-sT 00 on

^coco^m

(NfNfNfNfN

OOOhvO

o o o rv o o o o o o

ooolooo<— vo m oo hv^

-H -H -H -H +1

fO LO fN^ C0 LO -sT O CN

hv -sf C^

o ^ o o rv

-H-H-H-H-H -H-H-H-H^

OfNCOO vo 00 Ohv ^ 00 00 ^ t N <N

I I I I I

I I I I I

<N00<—0

O^C^lO Nvp

00 ".O "sT 00 on

mrv -sT

<— LO

OOLO -sT-sT ■sTomLOhv

OOOvo<sT ■sT O O <N O

O o o o o

ooohvam oo vo o ^ |<

O^r^OCO

-H -H -H -H +1

N (N O LO on

OfNm o o m m o

-H-H-H-H-H -H-H-H-H^

lolooo -sr mrv o ON

2 25 -

"Ss^Sj

.¡^ ra u

■qô:<

a c ai

u „ > « ■ •

ct -a .E

group. Fifty-two percent of patients in the placebo group discontinued RAASi therapy due to their insufficient HK control while on RAASi therapy. Modification of non-RAASi, nondiuretic, antihypertensive medications occurred in #2% of patients on patiromer and #8% of those switched to placebo. Through week 8 of the randomized withdrawal phase, modifications in diuretics occurred in #3% of patients switched to placebo and in no patients continuing patiromer.

Safety

Patiromer adverse events (AEs) have been previously described.2 Briefly, patiromer was well tolerated with a low rate of discontinuations due to AEs (6% in the initial treatment phase and 2% in the randomized withdrawal phase). Overall, 47% of patients in the initial treatment phase had at least 1 AE, with similar percentages reported in the randomized withdrawal phase (47% and 50% in the patiromer and placebo groups, respectively). Serious AEs occurred in 3 patients in the initial treatment phase, 1 patient on placebo, and no patients on patiromer in the randomized withdrawal phase. As determined by the investigator, none of the serious AEs were related to patiromer. The mean serum magnesium level remained decreased slightly during both phases; no episodes of serum magnesium <1.2 mg/dl occurred. Hypokalemia (sK <3.5 mEq/l) was infrequent, occurring in 3% of patients in the initial treatment phase; in the randomized withdrawal phase, 5% of patients on patiromer and 2% on placebo met prespecified withdrawal criteria for sK <3.8 mEq/l.

DISCUSSION

The current analysis notes that patiromer treatment resulted in reduced aldosterone levels, decreases in BP, and reductions in urine albumin excretion. The effect of patiromer on lowering potassium without delivering sodium through counterion exchange,2 as well as observing decreases in serum aldosterone, provides an opportunity to examine the biological effects of reducing sK and serum aldosterone levels. It is unlikely that the observed reduction in serum aldosterone occurred by chance and was unrelated to patiromer's effect on sK. Decreases in the magnitude that we observed are unlikely to occur spontaneously in patients with kidney disease,29'30 suggesting that the observed effect on aldosterone was driven by patiromer treatment. The effect of patiromer on aldosterone is further supported by the findings that subjects switched to placebo in the randomized withdrawal phase had increases in aldosterone, whereas those who continued patiromer therapy maintained the decreased aldosterone levels achieved in the initial treatment phase. The effect of patiromer on aldosterone was maintained after adjusting for age, diabetes, and heart failure, factors that may alter aldosterone levels and could have affected the results. Although patiromer and other agents are documented to reduce potassium levels in individuals with CKD,^31-34 there are no previously published reports that demonstrate a reduction in aldosterone, blood pressure, and albuminuria with these agents.

Figure 21 Effect of patiromer on observed mean values of systolic (a) and diastolic (b) blood pressure over time during the initial treatment phase.

Aldosterone secretion is primarily controlled by angio-tensin II, potassium, and adrenocorticotrophic hormone, and increased levels of any of these will increase aldosterone production, primarily due to effects on an early step in aldosterone biosynthesis.21"23'35"37 Changes in potassium levels modify aldosterone levels, as reported in the current study, due to the high sensitivity of zona glomerulosa cells to sK levels.21 For example, administering potassium to normal subjects in doses so small that sK levels are not measurably changed can modify aldosterone levels by as much as 25%.25

Furthermore, in anephric individuals, there were parallel changes in potassium and aldosterone in between dialysis sessions.38 Similarly, in the current study, when patiromer was stopped, potassium and aldosterone levels rose in parallel; conversely, aldosterone and potassium levels remained suppressed in patients in whom patiromer was continued. Although the overall duration of the study was longer than required for aldosterone to escape from the suppressing effects of agents that interrupt the RAAS,30 it is possible that aldosterone levels may not continue to be reduced with

Table 31 Serum aldosterone and blood pressure adjusted for concomitant selected antihypertensive use in the initial treatment phase

Mild hyperkalemia (5.1 to <5.5 mEq/l) Moderate/severe hyperkalemia (5.5 to <6.5 mEq/l) Overall (5.1 to <6.5 mEq/l)

Measure LS mean LS mean change from baseline ± SE P value change from baseline ± SE P value LS mean change from baseline ± SE P value

Serum aldosterone (ng/dl) Systolic blood pressure (mm Hg) Diastolic blood pressure (mm Hg) -1.81 ± 0.74 -5.40 ± 1.84 -3.34 ± 1.21 0.0156 0.0037 0.0064 -3.19 ± 0.58 <0.0001 -5.57 ± 1.44 0.0001 -4.67 ± 0.95 <0.0001 -2.50 ± 0.47 -5.48 ± 1.17 -4.00 ± 0.77 <0.0001 <0.0001 <0.0001

LS, least squares.

The same mixed-effects, repeated-measures model in Table 2 was used by excluding any aldosterone antagonist use and any modifications of renin-angiotensin-aldosterone system inhibitor, non-renin-angiotensin-aldosterone system inhibitor, nondiuretic antihypertensive medications, or of diuretics. P value for mean change compared with baseline.

Table 4| Baseline (mean ± SD) clinical and laboratory parameters by the tertile of baseline (mean ± SD) serum aldosterone

First tertile of aldosterone at Second tertile of aldosterone at Third tertile of aldosterone at Baseline baseline (3.49 ± 1.2 ng/dl) baseline (8.49 ± 1.74 ng/dl) baseline (23.77 ± 12.65 ng/dl) P value

N 81 82 80

Potassium (mEq/l) 5.48 ± 0.40 5.54 ± 0.46 5.67 ± 0.47 0.0246

Systolic blood pressure (mm Hg) 142.03 ± 17.35 141.00 ± 17.17 140.88 ± 16.90 0.8962

Diastolic blood pressure (mm Hg) 78.37 ± 11.59 79.91 ± 10.45 77.90 ± 10.54 0.4691

Plasma renin activity (mg/l/hr) 8.21 ± 11.27 7.57 ± 10.55 12.00 ± 14.01 0.0423

ACR (mg/g) 1308.91 ± 2048.70 888.36 ± 1494.20 783.19 ± 1987.20 0.1672

ACR, albumin-to-creatinine ratio; P value for comparisons among tertiles was based on 1-way ANOVA.

Table 51 Change from baseline to day 3 (mean ± SD) in clinical and laboratory parameters by the tertile of change from baseline to day 3 (mean ± SD) in serum aldosterone3

First tertile of aldosterone Second tertile of aldosterone Third tertile of aldosterone change to day 3 change to day 3 change to day 3

Change from baseline to day 3 (-6.74 ± 6.89 ng/dl) (-0.20 ± 0.55 ng/dl) (4.45 ± 4.36 ng/dl) P value

N 71 73 71

Potassium (mEq/l) -0.56 ± 0.56 -0.37 ± 0.41 -0.31 ± 0.43 0.0044

Systolic blood pressure (mm Hg) -4.61 ± 14.76 -3.35 ± 15.35 -4.24 ± 12.95 0.8651

Diastolic blood pressure (mm Hg) -3.64 ± 8.39 -2.87 ± 8.88 -0.55 ± 7.20 0.0671

Plasma renin activity (mg/l/hr) 0.26 ± 10.03 -2.26 ± 10.44 2.33 ± 7.56 0.0162

aACR was not available on day 3. P value for comparisons among tertiles was based on 1-way ANOVA.

long-term patiromer treatment. Patiromer is unlikely to exert direct, systemic effects on aldosterone production or metabolism, BP, or albuminuria because patiromer itself is not systemically absorbed.39

The combination of lower levels of aldosterone, BP, and sK with patiromer would be expected to have induced a reactive increase in PRA23'25; there are at least 3 potential explanations why this did not occur. First, decrease in aldosterone did not result in a decrease in extracellular volume. Because assessments of volume status were not conducted in this study, we have no way of directly assessing this possibility. Second, the usual reciprocal relationship between RAAS and intravascular volume was abnormal in the basal state (before patiromer treatment). Thus, even though extracellular volume might have decreased during patiromer treatment, it did not modify PRA levels significantly, similar to what is seen in low renin hypertension.40,41 We have no data to support or refute this possibility. Third, the regulation of renin release/secretion was altered in these patients, similar to the hyporenemic-hypoaldosterone syndrome found in some patients with chronic renal disease.40,41 This seems the least likely

explanation because renin levels were not decreased at baseline in this study. In sum, the reason that PRA did not increase is unexplained and requires further study.

Based on the current study, there is no direct evidence to determine whether reductions in BP and albuminuria were secondary to reductions in aldosterone levels. Aldosterone levels were not outside the normal range before patiromer was given and were still within the normal range after drug administration. Other studies, however, analyzed animal models that were placed on an unrestricted sodium diet and had an increased cardiovascular and renal risk and discovered that blocking the mineralocorticoid receptor or performing an adrenalectomy prevented cardiovascular and renal dam-age.42,43 Moreover, restoring aldosterone levels in the adre-nalectomized rodents restored the damage.42,43 Thus, aldosterone levels do not need to be clinically elevated to induce damage. The potential clinical implications of these findings suggest that patiromer may have secondary effects on aldosterone levels that may result in cardiovascular and renal benefits. However, these effects of patiromer need to be formally tested.

Table 6| Change from baseline to week 4 (mean ± SD) in clinical and laboratory parameters by the tertile of change from baseline to week 4 (mean ± SD) in serum aldosterone

Change from baseline to week 4 First tertile of aldosterone change to week 4 (-11.72 ± 8.83 ng/dl) Second tertile of aldosterone change to week 4 (-1.39 ± 1.26 ng/dl) Third tertile of aldosterone change to week 4 (6.05 ± 5.95 ng/dl) P value

N 71 73 71

Potassium (mEq/l) -1.07 ± 0.70 -1.01 ± 0.59 -0.93 ± 0.72 0.4413

Systolic blood pressure (mm Hg) -7.26 ± 16.66 -1.56 ± 17.57 -8.41 ± 16.39 0.0342

Diastolic blood pressure (mm Hg) -5.08 ± 11.34 -2.94 ± 10.98 -3.63 ± 11.40 0.5026

Plasma renin activity (mg/l/hr) -0.91 ± 11.51 -0.52 ± 7.50 0.51 ± 11.19 0.6865

ACR (mg/g) -100.96 ± 648.21 -169.97 ± 817.36 -345.41 ± 914.91 0.1733

ACR, albumin-to-creatinine ratio. P value for comparisons among tertiles was based on 1-way ANOVA.

Table 71 Mean ± SE change in clinical and laboratory parameters from start of the randomized withdrawal phase to weeks 4 and 8 of the randomized withdrawal phase

Measure Patiromer, N = 55 Placebo, N = 52

n Mean ± SE LS mean change from start ± SE P value n Mean ± SE LS mean change from start ± SE P value

Serum aldosterone (ng/dl) 55 10.52 ± 1.25 — — 52 8.61 ± 1.38 — —

Potassium (mEq/l) 55 4.50 ± 0.06 — — 52 4.45 ± 0.05 — —

Systolic blood pressure (mm Hg) 55 134.80 ± 2.19 — — 52 135 ± 2.51 — —

Diastolic blood pressure (mm Hg) 55 76.48 ± 1.60 — — 52 73.65 ± 1.54 — —

Plasma renin activity (mg/l/hr) 55 8.71 ± 1.59 — — 52 7.90 ± 1.53 — —

ACRa (mg/g) 55 441.70 ± 105.50 — — 50 509.40 ± 124.20 — —

Week 4

Serum aldosterone (ng/dl) 50 11.17 ± 1.62 0.94 ± 1.18 0.4262 45 12.28 ± 2.02 3.37 ± 1.24 0.0078

Potassium (mEq/l) 50 4.55 (0.06) 0.07 (0.06) 0.2621 45 4.95 (0.07) 0.51 (0.07) <0.0001

Systolic blood pressure (mm Hg) 50 131.30 ± 2.25 -3.91 ± 2.04 0.0582 45 135.90 ± 2.53 0.76 ± 2.14 0.7216

Diastolic blood pressure (mm Hg) 50 73.88 ± 1.38 -2.59 ± 1.20 0.0335 45 76.47 ± 1.62 0.78 ± 1.26 0.5359

Plasma renin activity (mg/l/hr) 50 7.37 ± 1.41 -0.61 ± 1.15 0.5956 44 6.34 ± 1.45 -1.66 ± 1.21 0.1718

ACRa (mg/g) 49 401.80 ± 115.50 50.38 ± 64.17 0.4345 42 613.50 ± 214.70 70.93 ± 69.33 0.3091

Week 8

Serum aldosterone (ng/dl) 45 10.63 ± 1.14 0.23 ± 1.07 0.8296 29 13.08 ± 2.93 2.78 ± 1.25 0.0278

Potassium (mEq/l) 45 4.52 ± 0.06 0.04 ± 0.07 0.5050 29 4.85 ± 0.08 0.55 ± 0.08 <0.0001

Systolic blood pressure (mm Hg) 45 128.50 ± 1.88 -6.70 ± 1.59 <0.0001 29 133.50 ± 2.23 -1.21 ± 1.89 0.5234

Diastolic blood pressure (mm Hg) 45 74.98 ± 1.13 -2.15 ± 1.06 0.0458 29 77.63 ± 1.98 1.72 ± 1.26 0.1734

Plasma renin activity (mg/l/hr) 44 9.64 ± 1.71 1.71 ± 1.21 0.1624 29 4.99 ± 1.37 -3.90 ± 1.41 0.0067

ACRa (mg/g) 45 417.10 ± 115.80 18.85 ± 46.59 0.6870 29 339.80 ± 134.90 -88.49 ± 58.04 0.1318

ACR, albumin-to-creatinine ratio; LS, least squares.

The mean changes from the start were estimated using a mixed-effects, repeated-measures model, which included the fixed continuous covariate of start values as well as fixed categorical covariates of the analysis visit and treatment group and used an unstructured covariance structure. P value for mean change compared with start of randomized withdrawal phase.

aACR was only collected at monthly visits; the mean change from start was based on an analysis of covariance by adjusting for the start value.

There are several limitations to our analyses. The study was not designed to establish a causal relationship between aldosterone decrease and decreases in BP or ACR. The initial treatment phase was not placebo controlled, and the sample size of the placebo-controlled randomized withdrawal phase of the study was relatively small (N = 107). Although infrequent, changes in non-RAASi antihypertensive medications may also have confounded the observed effects on BP and ACR despite our attempts to control for these factors. In the randomized withdrawal phase, discontinuation of RAASi medications occurred in more than half of the placebo group, preventing direct between-group statistical comparisons of BP, ACR, and aldosterone levels with the patiromer group in whom discontinuations were uncommon (6%). We chose to examine within-group comparisons for evaluating changes beyond the initial 4 weeks of treatment (continued patiromer treatment) as well as the effects of patiromer withdrawal (placebo group) rather than between-group comparisons, which would have been subject to greater confounding due to differences in RAASi adherence. Because the study population was almost entirely white, the results may not be generalizable to other groups that may have differences in RAAS compared with whites.44 The findings of this study should be interpreted with caution. Even considering the previously cited factors, the reduction in aldosterone concentration during the initial treatment phase is striking as is the persistence of the effect over the 12 weeks of treatment and the loss of this effect in

patients switched to placebo in the randomized withdrawal phase. Finally, it is not possible to determine whether the results observed in this study are equally applicable to all potassium-lowering agents.

In summary, patiromer treatment in individuals with HK and CKD lowers potassium and aldosterone levels and is associated with reductions in BP and albuminuria. Treatment with patiromer may improve cardiovascular risk beyond that associated with reduction in potassium levels. This hypothesis will need formal testing.

METHODS

Study design and participants

The study design of the OPAL-HK study was previously described2 as a sequential, 2-part, placebo-controlled, 12-week phase 3 study evaluating patiromer treatment in 243 patients with CKD and HK on RAASi medications. Briefly, participants were enrolled in an initial 4-week treatment phase followed by an 8-week randomized withdrawal phase. Eligible patients were adults with CKD stage 3 or 4 (estimated glomerular filtration rate 15 to <60 ml/min per 1.73 m2) and sK levels of 5.1 to <6.5 mEq/l at screening based on local laboratory measurement and who were on a stable dose of RAASi medication for at least 28 days before screening. Patiromer starting doses at the beginning of the initial 4-week treatment phase were based on screening sK levels (4.2 g twice daily for mild HK [5.1 to <5.5 mEq/l] and 8.4 g twice daily for moderate to severe HK [5.5 to <6.5 mEq/l]). The dose was adjusted according to a prespecified algorithm2 to maintain sK within the range of 3.8 to <5.1 mEq/l. During this phase, RAASi therapy was not allowed to be adjusted unless medically necessary, but could be

discontinued if potassium was $6.5 mEq/l (or $5.1 mEq/l while on maximum doses of patiromer).

Patients eligible for the 8-week randomized withdrawal phase of the study had moderate to severe HK ($5.5 mEq/l based on central laboratory values) at baseline of the initial 4-week treatment phase and sK between 3.8 and <5.1 mEq/l (normokalemic) at the end of that phase. Additional eligibility requirements included receiving both patiromer treatment and RAASi therapy at the end of the initial 4-week treatment phase. Eligible patients were randomized (1:1 ratio) to either continue patiromer at the daily dose that they were receiving at week 4 of the initial treatment phase or to switch to placebo. Recurrence of HK was managed according to a prespecified algorithm as previously described.2 During the first 4 weeks of the 8-week withdrawal phase, while the primary endpoint was under evaluation, recurrent HK (serum K+ $5.5 mEq/l) was managed by increasing the dose in the patiromer group at an increment of 8.4 g/day or by decreasing the RAASi dose by 50% in the placebo group. During the second 4 weeks, after the primary endpoint had been evaluated, these measures were implemented if the serum potassium increased to 5.1 mEq/l. At any time, if these measures were insufficient to control recurrent HK, RAASi medication in either group had to be withdrawn. Investigators could add or modify potassium-neutral non-RAASi antihypertensive medications to control BP during the initial 4-week treatment phase or the 8-week randomized withdrawal phase.

Laboratory assessments conducted at baseline (day 1), on day 3 of both treatment phases, and weekly thereafter until the end of the study included sK, serum chemistry (including serum creatinine and estimated glomerular filtration rate), aldosterone, and PRA. Sitting BP was measured in triplicate at baseline, day 3 of each phase, and weekly thereafter. Urine samples were collected for ACR at baseline and every 4 weeks from the start of the study until the end of the study. Aldosterone and PRA were measured by chemiluminescent immunoassay.45,46 Intra-assay coefficient of variation was 2.5% to 5.4% and 4.6% to 10%, and the interassay coefficient of variation was 3.8% to 15.7% and 5.6% to 7.6% for aldosterone and PRA assays, respectively.

Statistical methods

Initial treatment phase. To evaluate the mean changes in serum aldosterone, PRA, SBP, and DBP in the initial treatment phase, data from all 243 enrolled patients were analyzed. An MMRM using restricted likelihood estimation was respectively fitted to weekly change from baseline measurements based on each central laboratory parameter (or clinical assessment) as the response variables by adjustment for the corresponding baseline value, visit, baseline HK severity, and interaction of visit by baseline HK severity. The dependency in repeated observations was modeled using an unstructured covariance fitted to within-subject correlations. The mean ± SE for all measures at specified time points (i.e., baseline and weeks 1-4) were based on observed cases and plotted.

Because ACR was only collected at monthly visits, the mean change in ACR from baseline to week 4 by baseline HK severity was based on analysis of covariance by adjusting for the baseline ACR value.

To exclude other possible drug effects on serum aldosterone, a sensitivity analysis was performed on serum aldosterone change by using the same model described previously, excluding any aldoste-rone antagonist use and any modifications of RAASi medications, non-RAASi, nondiuretic antihypertensive medications, or diuretics.

In addition, to further confirm the serum aldosterone change pattern and estimate the other covariate effects, an extended

multivariate MMRM was conducted. Along with adjustments in the original MMRM, this extended model also included pooled age group (65 years of age or older vs. younger than 65 years of age), sex, CKD stages by estimated glomerular filtration rate, presence or absence of type 2 diabetes mellitus, and history of heart failure.

In order to explore whether serum aldosterone changes are associated with patterned increased or decreased potassium, PRA, SBP, and DBP at baseline, day 3 (early effect), and week 4 (end of the initial treatment phase), a 1-way analysis of variance was used for baseline or change from baseline comparisons of central laboratory measures and BPs when the patients were subdivided into aldosterone tertiles; tertiles were defined bybaseline values and the change in aldosterone values at day 3 and week 4. Values are expressed as mean ± SD.

Randomized withdrawal phase. To investigate the mean changes in serum aldosterone, PRA, SBP, and DBP in the 8-week randomized withdrawal treatment phase, data from the 107 randomized patients were analyzed. An MMRM was fitted to weekly change from the start of the randomized withdrawal phase measurements based on the central laboratory as the response variables by adjustment for the start value, visit, treatment (patiromer vs. placebo), and interaction of visit by treatment. The dependency in repeated observations was modeled using an unstructured covari-ance fitted to within-patient errors. The mean and SE for all measures at the specified time points (e.g., start value, weeks 4 and 8 during the randomized withdrawal phase) were based on observed cases. The ACR was analyzed by using the same methods described in the initial treatment phase.

Descriptive statistics for baseline demographic and characteristics were summarized as mean ± SD for continuous variables or proportions for categorical variables. All analyses were performed using SAS version 9.4 (SAS Institute, Cary, North Carolina), with statistical significance set at P < 0.05.

DISCLOSURE

MRW reports consulting fees from Akebia Therapeutics, Amgen, AstraZeneca, Boston Scientific, Janssen Pharmaceutica, Lexicon, Merck Sharp & Dohme, Relypsa, Inc, Sanofi, and Sandoz . GLB reports consulting fees from AbbVie Inc., AstraZeneca, Bayer, CVRx, Janssen Pharmaceutica, Medtronic, Relypsa, Inc., and Takeda Pharmaceutical Company. GHW reports consulting fees from Daiichi Sankyo, Mitsubishi Tanabe Pharma, Pfizer Japan Inc, and Relypsa, Inc. CG, MRM, DG, JY, and LB are employees of Relypsa, Inc. YS was an employee of Relypsa, Inc. when the study was conducted and a paid consultant of Relypsa Inc. when this manuscript was submitted.

ACKNOWLEDGMENTS

This study was funded by Relypsa, Inc . Editorial support was provided by Colville Brown, MD, of AlphaBioCom, LLC, and Julie Ann Obeid of Relypsa, Inc., funded by Relypsa, Inc.

SUPPLEMENTARY MATERIAL

Table S1. Assessment of covariate effects on change from baseline to week 4 in serum aldosterone in the initial treatment phase: MMRM. Table S2. RAASi and diuretic use in the initial treatment phase . Supplementary material is linked to the online version of the paper at www .kidney-international .org.

REFERENCES

1. McCullough PA, Costanzo MR, Silver M, et al . Novel agents for the prevention and management of hyperkalemia . Rev Cardiovasc Med. 2015;16:140-155.

2. Weir MR, Bakris GL, Bushinsky DA, et al. Patiromer in patients with kidney 26. disease and hyperkalemia receiving RAAS inhibitors. N Engl J Med. 2015;372:211-221.

3. Blasi ER, Rocha R, Rudolph AE, et al. Aldosterone/salt induces renal 27. inflammation and fibrosis in hypertensive rats. Kidney Int. 2003;63: 1791-1800.

4. Brown JM, Underwood PC, Ferri C, et al. Aldosterone dysregulation with 28. aging predicts renal vascular function and cardiovascular risk. Hypertension. 2014;63:1205-1211.

5. Williams GH. Cardiovascular benefits of aldosterone receptor 29. antagonists: what about potassium? Hypertension. 2005;46:265-266.

6. Pitt B, Reichek N, Willenbrock R, et al. Effects of eplerenone, enalapril, and eplerenone/enalapril in patients with essential hypertension and left ventricular hypertrophy: the 4E-left ventricular hypertrophy study. 30. Circulation. 2003;108:1831-1838.

7. Pitt B, Remme W, Zannad F, et al. Eplerenone, a selective aldosterone blocker, in patients with left ventricular dysfunction after myocardial 31. infarction. N Engl J Med. 2003;348:1309-1321.

8. Epstein M, Williams GH, Weinberger M, et al. Selective aldosterone blockade with eplerenone reduces albuminuria in patients with type 2 diabetes. Clin J Am Soc Nephrol. 2006;1:940-951. 32.

9. Lazich I, Bakris GL. Prediction and management of hyperkalemia across the spectrum of chronic kidney disease. Semin Nephrol. 2014;34:333-339.

10. Garg R, Hurwitz S, Williams GH, et al. Aldosterone production and insulin 33. resistance in healthy adults. J Clin Endocrinol Metab. 2010;95:1986-1990.

11. McGraw AP, Bagley J, Chen WS, et al. Aldosterone increases early atherosclerosis and promotes plaque inflammation through a placental growth factor-dependent mechanism. J Am Heart Assoc. 2013;2:e000018. 34.

12. Pruthi D, McCurley A, Aronovitz M, et al. Aldosterone promotes vascular remodeling by direct effects on smooth muscle cell mineralocorticoid receptors. Arterioscler Thromb Vasc Biol. 2014;34:355-364. 35.

13. Catena C, Lapenna R, Baroselli S, et al. Insulin sensitivity in patients with primary aldosteronism: a follow-up study. J Clin Endocrinol Metab. 2006;91:3457-3463. 36.

14. Pitt B, Zannad F, Remme WJ, et al. The effect of spironolactone on morbidity and mortality in patients with severe heart failure.

Randomized Aldactone Evaluation Study Investigators. N Engl J Med. 37.

1999;341:709-717.

15. Ito Y, Mizuno M, Suzuki Y, et al. Long-term effects of spironolactone in 38. peritoneal dialysis patients. J Am Soc Nephrol. 2014;25:1094-1102.

16. Sengul E, Sahin T, Sevin E, et al. Effect of spironolactone on urinary

protein excretion in patients with chronic kidney disease. Ren Fail. 39.

2009;31:928-932.

17. Schepkens H, Vanholder R, Billiouw JM, et al. Life-threatening hyperkalemia during combined therapy with angiotensin-converting enzyme inhibitors and spironolactone: an analysis of 25 cases. Am J Med. 2001;110:438-441. 40.

18. Volk MJ, Bomback AS, Klemmer PJ. Mineralocorticoid receptor blockade

in chronic kidney disease. Curr Hypertens Rep. 2011;13:282-288. 41.

19. Schwenk MH, Hirsch JS, Bomback AS. Aldosterone blockade in CKD: emphasis on pharmacology. Adv Chronic Kidney Dis. 2015;22:123-132.

20. Khosla N, Kalaitzidis R, Bakris GL. Predictors of hyperkalemia risk 42. following hypertension control with aldosterone blockade. Am J Nephrol. 2009;30:418-424. 43.

21. Williams GH, Dluhy RG. Aldosterone biosynthesis: interrelationship of regulatory factors. Am J Med. 1972;53:595-605.

22. Williams JS, Williams GH. 50th anniversary of aldosterone. J Clin 44. Endocrinol Metab. 2003;88:2364-2372.

23. Dluhy RG, Underwood RH, Williams GH. Influence of dietary potassium

on plasma renin activity in normal man. J Appl Physiol. 1970;28:299-302. 45.

24. Dluhy RG, Axelrod L, Underwood RH, et al. Studies of the control of plasma aldosterone concentration in normal man: II. Effect of dietary potassium and acute potassium infusion. J Clin Invest. 1972;51:1950-1957. 46.

25. Himathongkam T, Dluhy RG, Williams GH. Potassim-aldosterone-renin interrelationships. J Clin Endocrinol Metab. 1975;41:153-159.

Dluhy RG, Greenfield M, Williams GH. Effect of simultaneous potassium and saline loading on plasma aldosterone levels. J Clin Endocrinol Metab. 1977;45:141-146.

Williams GH, Braley LM. Effects of dietary sodium and potassium intake and acute stimulation on aldosterone output by isolated human adrenal cells. J Clin Endocrinol Metab. 1977;45:55-64. Adler GK, Moore TJ, Hollenberg NK, et al. Changes in adrenal responsiveness and potassium balance with shifts in sodium intake. Endocr Res. 1987;13:419-445.

Persson F, Lews JB, Lewis EJ, et al. Impact of aliskiren treatment on urinary aldosterone levels in patients with type 2 diabetes and nephropathy: an AVOID substudy. J Renin Angiotensin Aldosterone Syst. 2011;13:118-121.

Bomback AS, Rekhtman Y, Klemmer PJ, et al. Aldosterone breakthrough

during aliskiren, valsartan and combination (aliskiren+valsartan)

therapy. J Am Soc Hypertens. 2012;6:338-345.

Bakris GL, Pitt B, Weir MR, et al. Effect of patiromer on serum

potassium level in patients with hyperkalemia and diabetic kidney

disease: the AMETHYST-DN randomized clinical trial. JAMA. 2015;314:

151-161.

Bushinsky DA, Williams GH, Pitt B, et al. Patiromer induces rapid and sustained potassium lowering in patients with chronic kidney disease and hyperkalemia. Kidney Int. 2015;88:1427-1433. Ash SR, Singh B, Lavin PT, et al. A phase 2 study on the treatment of hyperkalemia in patients with chronic kidney disease suggests that the selective potassium trap, ZS-9, is safe and efficient. Kidney Int. 2015;88: 404-411.

Lepage L, Dufour A-C, Doiron J, et al. Randomized clinical trial of sodium polystyrene sulfonate for the treatment of mild hyperkalemia in CKD. Clin J Am Soc Nephrol. 2015;10:2136-2142.

Gallo-Payet N, Grazzini E, Côté M, et al. Role of Ca2+ in the action of adrenocorticotropin in cultured human adrenal glomerulosa cells. J Clin Invest. 1996;98:460-466.

Kopf PG, Gauthier KM, Zhang DX, et al. Angiotensin II regulates adrenal vascular tone through zona glomerulosa cell-derived EETs and DHETs. Hypertension. 2011;57:323-329.

Williams GH. Aldosterone biosynthesis, regulation, and classical

mechanism of action. Heart Fail Rev. 2005;10:7-13.

Williams GH, Bailey GL, Hampers CL, et al. Studies on the metabolism of

aldosterone in chronic renal failure and anephric man. Kidney Int. 1973;4:

280-288.

Li L, Harrison SD, Cope MJ, et al. Mechanism of action and pharmacology of patiromer, a nonabsorbed cross-linked polymer that lowers serum potassium concentration in patients with hyperkalemia [e-pub ahead of print]. J Cardiovasc Pharmacol Ther. 2016;pii:1074248416629549. Accessed June 3, 2016.

Schambelan M, Stockigt JR, Biglieri EG. Isolated hypoaldosteronism in adults. A renin-deficiency syndrome. N Engl J Med. 1972;287:573-578. Weidmann P, Maxwell MH, Rowe P, et al. Role of the renin-angiotensin-aldosterone system in the regulation of plasma potassium in chronic renal disease. Nephron. 1975;15:35-49.

Rocha R, Stier CT Jr, Kifor I, et al. Aldosterone: a mediator of myocardial necrosis and renal arteriopathy. Endocrinology. 2000;141:3871 -3878. Guo C, Martinez-Vasquez D, Mendez GP, et al. Mineralocorticoid receptor antagonist reduces renal injury in rodent models of types 1 and 2 diabetes mellitus. Endocrinology. 2006;147:5363-5373. Rifkin DE, Khaki AR, Jenny NS, et al. Association of renin and aldosterone with ethnicity and blood pressure: the Multi-Ethnic Study of Atherosclerosis. Am J Hypertens. 2014;27:801-810.

Underwood RH, Williams GH. The simultaneous measurement of aldosterone, cortisol, and corticosterone in human peripheral plasma by displacement analysis. J Lab Clin Med. 1972;79:848-862. Vaidya A, Sun B, Forman JP, et al. The Fok1 vitamin D receptor gene polymorphism is associated with plasma renin activity in Caucasians. Clin Endocrinol (Oxf). 2011;4:783-790.