effects of phosphate binders in moderate ckd€¦ · enrollment, baseline characteristics, and...

CLINICAL RESEARCH www.jasn.org

Effects of Phosphate Binders in Moderate CKD

Geoffrey A. Block,* David C. Wheeler,† Martha S. Persky,* Bryan Kestenbaum,‡

Markus Ketteler,§ David M. Spiegel,| Matthew A. Allison,¶ John Asplin,** Gerard Smits,*Andrew N. Hoofnagle,‡ Laura Kooienga,* Ravi Thadhani,†† Michael Mannstadt,††

Myles Wolf,‡‡ and Glenn M. Chertow§§

*Denver Nephrology, Denver, Colorado; †Centre for Nephrology, Royal Free Campus, University College LondonMedical School, London, United Kingdom; ‡University of Washington, Seattle, Washington; §Nephrologische Klinik,Klinikum Coburg, Coburg, Germany; |University of Colorado, Denver, Colorado; ¶University of California, San Diego,California; **Litholink Corp, Chicago, Illinois; ††Massachusetts General Hospital and Harvard Medical School, Boston,Massachusetts; ‡‡University of Miami, Miami, Florida; and §§Stanford University School of Medicine, Palo Alto,California

ABSTRACTSome propose using phosphate binders in the CKD population given the association between higherlevels of phosphorus and mortality, but their safety and efficacy in this population are not well under-stood. Here, we aimed to determine the effects of phosphate binders on parameters of mineral metab-olism and vascular calcification among patients with moderate to advanced CKD. We randomly assigned148 patients with estimated GFR=20–45 ml/min per 1.73 m2 to calcium acetate, lanthanum carbonate,sevelamer carbonate, or placebo. The primary endpoint was change in mean serum phosphorus frombaseline to the average of months 3, 6, and 9. Serum phosphorus decreased from a baseline meanof 4.2mg/dl in both active and placebo arms to 3.9mg/dl with active therapy and 4.1mg/dl with placebo(P=0.03). Phosphate binders, but not placebo, decreased mean 24-hour urine phosphorus by 22%.Median serum intact parathyroid hormone remained stable with active therapy and increased with placebo(P=0.002). Active therapy did not significantly affect plasma C-terminal fibroblast growth factor 23 levels.Active therapy did, however, significantly increase calcification of the coronary arteries and abdominal aorta(coronary: median increases of 18.1% versus 0.6%, P=0.05; abdominal aorta: median increases of 15.4%versus 3.4%, P=0.03). In conclusion, phosphate binders significantly lower serum and urinary phosphorusand attenuate progression of secondary hyperparathyroidism among patients with CKDwho have normal ornear-normal levels of serum phosphorus; however, they also promote the progression of vascular calcifica-tion. The safety and efficacy of phosphate binders in CKD remain uncertain.

J Am Soc Nephrol 23: 1407–1415, 2012. doi: 10.1681/ASN.2012030223

CKD is a significant public health concern; roughly13% of the US population has an estimated GFR(eGFR) below 60 ml/min per 1.73 m2 or albumin-uria.1 The risks of death and cardiovascular diseasein CKD are not fully explained by associated dia-betes, hypertension, and other conventional riskfactors.2 With declining kidney function, serumphosphorus concentration increases but generallyremains within the normal range until late in stage4 or 5 CKD. A normal or near-normal serum phos-phorus concentration is maintained at the expenseof elevated levels of the phosphaturic hormonesparathyroid hormone (PTH) and fibroblast growth

factor 23 (FGF23).3,4 Serum concentrations ofphosphorus and hormones responsible for its regu-lation have been implicated as putative cardiovascu-lar risk factors.5–9

Received March 2, 2012. Accepted May 10, 2012.

Published online ahead of print. Publication date available atwww.jasn.org.

Correspondence: Dr. Geoffrey A. Block, Denver Nephrology,130 Rampart Way, Suite 300b, Denver, CO 80230. Email: [email protected]

Copyright © 2012 by the American Society of Nephrology

J Am Soc Nephrol 23: 1407–1415, 2012 ISSN : 1046-6673/2308-1407 1407

CLINICALRESE

ARCH

http://www.jasn.org

mailto:[email protected]

mailto:[email protected]

Three intestinal phosphate binders are approved for thetreatment of hyperphosphatemia in patients with ESRD in theUnited States—calcium acetate (Phoslo; Fresenius), lanthanumcarbonate (Fosrenol; Shire), and sevelamer carbonate (Renvela;Genzyme)—but none are approved for use inpatientswithCKDnot on dialysis.10 Because higher serum phosphorus concentra-tions—even within the population reference range—are associ-ated with mortality, some have proposed the use of phosphatebinders to lower serum phosphorus in this population.11 Weundertook this pilot clinical trial to determine the safety andefficacy of phosphate binders in patients with moderate CKDand normal or near-normal serum phosphorus concentrations.

RESULTS

Enrollment, Baseline Characteristics, and StudyConductFigure 1 shows patient disposition. Baseline characteristics(Table 1) were similar across treatment groups. Mean dosesof studymedicationwere 5.9 g/d calcium acetate (1.5 g elementalcalcium), 2.7 g/d lanthanum carbonate, and 6.3 g/d sevelamercarbonate. Average adherence was .85% in each of the treat-ment arms. The median duration of follow-up was 249 days.Changes in selected on-study biochemical parameters areshown in Supplemental Table 1. Table 2 shows adverse ef-fects. Gastrointestinal side effects were more common withlanthanum but were generally mild and of limited duration.Only one related serious adverse event (hypothyroidism inplacebo group) was observed during the study.

Primary EndpointChange in mean serum phosphorus was significantly differentbetween active- and placebo-treated groups (P=0.03). Meanserum phosphorus declined from 4.2 to 3.9 mg/dl in patientstreated with phosphate binders and was unchanged in placebo-treated patients (Figure 2A). Results in the individual activetreatment arms show statistically significant reductions versusplacebo only in the lanthanum carbonate arm (P=0.04), andthey are shown in Supplemental Figure 1A.

Secondary EndpointsUrine PhosphorusChange in mean urine phosphorus excretion was significantlydifferent between active- andplacebo-treated groups (P=0.002)(Figure 2B). Daily urine phosphorus excretion was reduced by22% with active therapy and unchanged with placebo. Similarresults were observed for themean fractional excretion of phos-phorus, which decreased (32%–26%) with active treatmentand was unchanged (31%–32%) with placebo (P,0.0001).All active treatment arms showed similar reductions in phos-phorus excretion (Supplemental Figure 1B).

PTH and 1,25 Dihydroxy Vitamin DChange in PTHwas significantly different between active- andplacebo-treated groups (P=0.002). PTH concentrations re-mained stable with active therapy and increased by 21% inpatients on placebo (P=0.002) (Figure 2C). 1,25 dihydroxyvitamin D concentration was significantly reduced with activetherapy (P=0.004) (Supplemental Table 1). Effects on PTHand 1,25 dihydroxy vitamin D seemed to differ by binder

type; patients treated with calcium acetateexperienced a larger relative decline,whereas the effects of lanthanum and sev-elamer relative to placebo were less pro-nounced (Supplemental Figure 1C andSupplemental Table 1).

FGF23Plasma C-terminal FGF23 concentrationswere markedly elevated in both treatmentarms at baseline relative to populationnorms. There was no significant differencein the change between active and placebo inplasmaC-terminal FGF23 (P=0.67) (Figure2D). Findings using the intact FGF23 assaymeasured at baseline and end of study sim-ilarly showed no difference between all ac-tive and all placebo (P=0.42). The effectson intact FGF23 seemed to differ by bindertype; patients treated with sevelamer car-bonate experienced a significant decrease inintact FGF23 (median decrease=24 pg/ml,P=0.002 versus placebo), whereas those pa-tients treated with calcium acetate had a sig-nificant increase (median increase=28 pg/ml,Figure 1. Patient flow chart.

1408 Journal of the American Society of Nephrology J Am Soc Nephrol 23: 1407–1415, 2012


http://jasn.asnjournals.org/lookup/suppl/doi:10.1681/ASN.2012030223/-/DCSupplemental






P=0.03 versus placebo) and those patients treated with lantha-num carbonate were no different compared with placebo(P=0.30) (Supplemental Figure 2, D and E).

Vascular Calcification and Bone Mineral DensityAssessment of change in vascular calcification was available ina subset of 96 patients (60 active and 36 placebo). No patient

with a zero calcium score at baseline developed a positivescore during follow-up. Among patients with nonzero calciumscores at baseline (n=81), active therapy resulted in significantincreases in median annualized percent change of coronaryartery (P=0.05) (Figure 3A) and abdominal aorta (P=0.03)(Figure 3B) calcium volume scores. Using the Hokansonsquare-root follow-up criterion, 20 of 52 (38%) phosphate

Table 1. Baseline characteristics of study participants

All Active(n=88)

All Placebo(n=57)

Lanthanum(n=28)

Sevelamer(n=30)

Calcium(n=30)

Siga

Demographicsmale 50% 49% 54% 50% 47% 1.00Caucasian 81% 79% 82% 80% 80% 1.00African-American 10% 11% 7% 7% 17%age (years 6 SD) 68611 65612 70610 66612 68612 0.72body mass index (kg/m2 6 SD) 3169 3268 3067 3368 3065 0.08

Comorbiditydiabetes 56% 58% 57% 53% 57% 0.86CHF 27% 11% 25% 23% 33% 0.02CAD 26% 14% 21% 27% 30% 0.10HTN 98% 100% 100% 97% 97% 0.52MI 12% 5% 4% 23% 10% 0.25PVD 24% 19% 21% 20% 30% 0.55CVA 8% 4% 14% 3% 7% 0.48secondaryhyperparathyroidism

72% 72% 79% 60% 77% 1.00

hyperlipidemia 89% 93% 86% 97% 83% 0.57nontraumatic fracture 17% 18% 7% 23% 20% 1.00

Laboratory characteristicsserum phosphorus(mg/dl 6 SD)b

4.260.4 4.260.4 4.260.4 4.260.4 4.260.4 0.47

eGFR (ml/min 6 SD)b 3268.1 3068.5 3367.9 3269.1 3067.1 0.32intact PTH (pg/ml 6 SD)c 78656 91652 87654 70656 76658 0.05C-terminal FGF23(RU/ml 6 IQR)c

227 (135–322) 211 (140–278) 230 (126–368) 230 (143–362) 219 (157–287) 0.59

intact FGF23(pg/mL 6 IQR)c

120 (88–176) 119 (95–169) 111 (65–194) 147 (95–202) 114 (90–152) 0.79

1,25 dihydroxy vitamin Db 25.769.4 27.2610.3 26.9610.0 24.7610.3 25.568.1 0.69coronary artery calciumvolume scorec

262 (87–863) 225 (18–737) 216.5 (102–1694) 362.5 (113–940.5) 130 (13–645) 0.33

thoracic aorta calciumvolume scorec

583 (165–2282) 496 (58–1539) 1609 (98–5030) 536 (351–2147) 511 (50–2145) 0.52

abdominal aorta calciumvolume scorec

1551 (326–5377) 1693 (212–6861) 4035 (200–8578) 1367 (519–3234) 1468 (16–4328) 0.70

L2 to L4 bone mineraldensity (g/cm2)b

111634 108645 99622 111629 120642 0.45

total urinary excretion ofphosphorus (mg 6 SD)b,d

7576444 8056359 7476606 7696350 7276354 0.20

fractional excretion ofphosphorus (% 6 SD)b

0.3260.12 0.3160.10 0.3060.13 0.3360.12 0.3360.11 0.48

total urinary excretion ofcalcium (mg 6 SD)b,d

55642 61647 53640 56643 58643 0.69

CAD, coronary artery disease; CHF, congestive heart failure; CVA, cerebro-vascular disease; HTN, hypertension; MI, myocardial infarction; PVD, peripheral vasculardisease.aSignificance. Fisher exact test for all active versus all placebo for binary measures and Wilcoxon rank sum tests for continuous measures.bMean.cMedian (IQR).dTwenty-four hour urine measurements were adjusted for adequacy of collection using the mean of each individual urinary creatinine excretion.

J Am Soc Nephrol 23: 1407–1415, 2012 Phosphate Binders in Moderate CKD 1409

www.jasn.org CLINICAL RESEARCH


binder-treated patients experienced progression of coronaryartery calcification compared with 5 of 29 (17%) placebo-treated patients (P=0.03). There was no significant differencein the thoracic aorta calcium score by treatment status(Figure 3C). Phosphate binder-treated patients showed sig-nificant improvement (P=0.03) in annualized change inbone mineral density (BMD) (Figure 3D).

The apparent adverse effects on vascular calcification andsalutary effects on BMD were most pronounced among patientsrandomized to calcium acetate (Supplemental Figures 1 and 2).

DISCUSSION

This randomized placebo-controlled pilot clinical trial showsthat the use of phosphate binders in patients with nondialysis-requiring CKD reduces urinary phosphorus excretion (asurrogate of intestinal phosphate absorption) and attenuatesprogressive secondary hyperparathyroidism. The effect onserum phosphorus, although statistically significant, wasmodest, despite relatively high-dose therapy and a significanteffect on urinary phosphorus excretion. Active therapy re-sulted in progression of vascular calcification, particularlyamong patients randomized to calcium acetate.

Higher serum phosphorus concentrations within the nor-mal range are associated with cardiovascular events andmortality in persons with CKD as well as persons with normalor near-normal kidney function.12–14 For example, among3368 participants in the Framingham Offspring Study whowere free of cardiovascular or kidney disease followed for

16.1 years, a serum phosphorus.3.5 mg/dlwas associated with a 55% higher risk for afirst cardiovascular event.12 In the Ath-erosclerosis Risk in Communities Study(n=15,732), the corresponding multivariable-adjusted relative risk of death was increasedby 14% (P,0.001).15 More recently, areanalysis of the Ramipril in NondiabeticRenal Disease study showed that patientswith serum phosphorus.3.5 mg/dl weremore than two times as likely to experienceprogression to ESRD or a doubling of se-rum creatinine. Moreover, higher serumphosphorus attenuated the renoprotectivebenefit of ramipril.16

The lack of a larger change in measuredserum phosphorus is surprising given therelatively large reduction in urine phos-phorus excretion, although similar findingshave been reported in the work by Oliveiraet al.17 A single determination of serumphosphorus may not accurately reflect thetime-averaged value. In normal, healthymen given phosphorus-supplementeddiets, single, fasting morning serum phos-

phorus concentrations grossly underestimate 24-hour expo-sure to phosphorus and fail to distinguish between low andhigh dietary phosphorus intake.18 Our results suggest that, inmoderate to advanced CKD, the serum phosphorus concen-tration may be a relatively poor marker of phosphate binderefficacy or adherence.

Intestinal absorption of phosphate is highly dependent onthe nature and absolute amount of ingested phosphorus andthe expression of the intestinal sodium–phosphorus cotrans-porter type 2b (NPT2b). Heterozygous NPT2b knockoutmiceshow a 50% reduction in receptor expression but have iden-tical serum phosphorus concentrations compared with wild-type mice.19 Mice with an inducible conditional knockout ofNPT2b also maintain serum phosphate concentrations, pos-sibly the result of compensatory increases in paracellular in-testinal phosphate absorption and renal expression of NPT2aand NPT2c.4,20 If phosphate binders were to upregulate ex-pression of intestinal NPT2b and enhance paracellular trans-port, the efficacy of binders could be limited, particularly aftermissed doses. Clinical trials of niacin and niacin derivatives,which are known to inhibit the expression of intestinal NPT2band kidney NPT2a, have shown significant reductions in se-rum phosphorus in patients with mild to moderate CKD.21

FGF23 seems to be a sensitive marker of impaired kidneyfunction, with elevations preceding detectable increases inserum creatinine.22 Circulating FGF23 concentrations arestrongly associated with progression to ESRD and mortality,and higher concentrations of FGF23 have recently been shownto cause left ventricular hypertrophy in rats.9 It had been as-sumed that reduction in intestinal phosphate absorption

Table 2. Related adverse events on study

Related AdverseEvent

AllActive(n=88)

AllPlacebo(n=57)

Lanthanum(n=28)

Sevelamer(n=30)

Calcium(n=30)

One or more adverseevents (n; %)

31 12 8 (29) 10 (33) 13 (43)

Constipation (n; %) 11 0 2 (7) 5 (17) 4 (13)Diarrhea (n; %) 5 1 5 (18)Dyspepsia (n; %) 4 1 1 (4) 1 (3) 2 (7)Hypercalcemia (n; %) 5 0 5 (17)Vomiting (n; %) 4 1 4 (14)Flatulence (n; %) 2 2 1 (4) 1 (3)Nausea (n; %) 3 1 3 (11)Hypophosphatemia(n; %)

2 1 1 (4) 1 (3)

Abdominal pain (n; %) 0 2Acidosis (n; %) 2 0 1 (4) 1 (3)Myalgia (n; %) 1 1 1 (3)AKI (n; %) 1 0 1 (3)Dysgeusia (n; %) 0 1GERD (n; %) 1 0 1 (3)Hyperparathyroidism(n; %)

0 1

Hypothyroidism (n; %) 0 1Fisher exact test for all active versus all placebo. GERD, gastro-esophageal reflux disease.




would decrease FGF23, although shorter-term phosphatebinder intervention failed to show any reduction in FGF23,despite yielding significant reductions in urinary phosphorusexcretion.23 Whether our disparate findings with C-terminaland intact FGF23 reflect better precision with the intact assayor true differential effects on both molecules is unknown. Asimilarly discrepant result between the C-terminal and intactFGF23 assays was described in 66 subjects with normal kidneyfunction, in whom dietary phosphate restriction reduced in-tact FGF23 but had no effect on C-terminal FGF2.24 Our re-sults confirm the results in the works by Olivieri et al.17 and

Yilmaz et al.25 Despite what seems to represent similar efficacyin terms of intestinal phosphate binding, intestinal phosphatebinders seem to exert disparate effects on serum concentrationsof FGF23.17,25 Treatment with phosphate binders resulted insignificant progression of vascular calcification in coronary ar-teries and the abdominal aorta. Assessment of coronary arterycalcification (CAC) progression using the square-root follow-up method in the work by Hokanson et al.26 has been shownto be the best method to predict mortality in a large cohort ofasymptomatic patients.26,27 These results support the conceptualrisk of inducing positive calcium balance in patients with CKD,

Figure 2. Biochemical changes on study by treatment arm. Change in (A) median (interquartile range [IQR]) serum phosphorus, (B) 24-hoururine phosphorus, (C) PTH, and (D) C-terminal FGF23 over the study period among all active- and placebo-treated patients.



where disordered mineral metabolism results in dystrophic cal-cification. Short-term calcium balance studies suggest that theprovision of 2000 mg/d elemental calcium in patients withCKD stage 3 results in substantial net positive calcium bal-ance.28 However, progression of vascular calcification wasevident not only in patients receiving calcium acetate, sug-gesting that use of other phosphate binders in CKD mightalso have unexpected and potentially, adverse consequences.

It is possible that noncalcium-containing phosphate bindersalso enhance the availability of free intestinal calcium andresult in a positive calcium balance. Although our primaryanalyses compared all actively treated groups with placebo,subgroup analyses suggested that calcium- and noncalcium-containing phosphate binders exert differential effects onPTH, FGF23, and mineralization of the bone and vascula-ture. In contrast, the work by Russo et al.29 studied 2-year

Figure 3. Vascular calcification and bone mineral density changes on study by treatment arm. Percent change in (A) median (IQR)annualized total coronary artery, (B) abdominal aorta, and (C) thoracic aorta calcium volume scores among active- and placebo-treatedpatients. (D) Median (IQR) percent change in annualized BMD among active- and placebo-treated patients. A trained reader examinedeach region of interest with exclusion of vertebral abnormalities.



progression of calcification in 90 patients with CKD stage 3/4randomized to a low phosphate diet, sevelamer carbonate, orcalcium carbonate. Patients receiving calcium carbonate ex-periencedmarked progression of CAC; however, in their anal-ysis, subjects given a low phosphate diet alone had rates ofCAC progression similar those subjects receiving calcium,whereas our placebo-treated subjects experienced little orno progression in CAC.

This study has several strengths, including the relativelylong study duration, blinding of participants and investigators,and excellent overall adherence. This study also has severalimportant limitations. This study was a single-center study,which served to diminish any impact of regional or culturaldifferences of diet on serum phosphorus concentrationsbut may have limited the generalizability of our findings toother geographic regions. We may have observed larger or lessvariable effects of phosphate binder therapy on serum phos-phorus had we standardized the time of day at which serumphosphorus wasmeasured.Wemade no attempts to standardizeintake of phosphorus, calcium, or other dietary components,and we made no attempt to standardize sun exposure or limitor match enrollment by season.

In summary, treatment with phosphate binders in patientswith moderate to advanced CKD and normal or near-normalserum phosphorus concentrations significantly loweredserum and urinary phosphorus and attenuated progressionof secondary hyperparathyroidism, but it resulted in pro-gression of coronary artery and abdominal aortic calcification.The safety and efficacy of phosphate binders in patients withCKD and normal serum phosphorus remain uncertain.

CONCISE METHODS

Study SettingThis trial was a randomized, double blind, placebo-controlled trial

of phosphate binders in patients with moderate to advanced CKD.

Recruitment at a single study center (Denver Nephrologists, Denver,

CO) began in February of 2009 and continued until September of

2010. Active drug and matching placebo were supplied by the three

manufacturers. The protocol was developed by an independent steer-

ing committee with full authority over study design and conduct. A

four-member independent data and safety monitoring board reviewed

safety data. The study was approved by the Schulman Institutional

Review Board (Cincinnati, OH) and registered at ClinicalTrials.gov

(NCT 00785629).

Study PopulationWe obtained written informed consent from all patients. Patients

were identified using an electronic medical record to capture all

consecutive patients seen in the study center clinic offices who met

preliminary study criteria. Patients were required to have an eGFR

using the Modification of Diet in Renal Disease study equation

between 20 and 45 ml/min per 1.73 m2, a serum phosphorus$3.5

and,6.0 mg/dl, and willingness to avoid any intentional change in

diet. Key exclusion criteria included use of any medication for

the purpose of binding intestinal phosphate, use of any active vita-

min D or the calcimimetic cinacalcet, intact PTH$500 pg/ml, or

uncontrolled hyperlipidemia.

Study DesignAn independent statistician performed randomization using SAS

version 9.1. A random seed was set, and data were generated using a

Data Step with a sequence of uniform random deviates. Patients were

block-randomized into one of three treatment arms, calcium acetate,

lanthanum carbonate, and sevelamer carbonate, in a 1:1:1 fashion.

We then randomly assigned patients within each treatment arm to

receive either active medication or matching placebo (3:2). Statistical

comparisonsweremadebetween the two treatment groups—all active

versus all placebo. We stratified randomization by diabetic status.

Sealed envelopes were opened at the study center by a staff member

not involved in the conduct of the trial. All study medication was

released by a single unblinded staff member in bottles identified by a

unique identification number. All clinical personnel, data analysts,

and participants remained blinded to study treatment assignment.

We administered 1 or 2 units per meal of study medication (calcium

acetate=667 mg, lanthanum carbonate=500 mg, and sevelamer car-

bonate=800 mg) based on screening serum phosphorus below or

above 4.5 mg/dl. We increased study medication at each visit if the

serum phosphorus remained .3.5 mg/dl to a maximum dose of 4

calcium acetate per meal (2668mg), 3 lanthanum carbonate per meal

(1500 mg), or 4 sevelamer carbonate per meal (3200 mg). We gave all

patients cholecalciferol (1000 IU daily). Patients underwent labora-

tory testing and event assessment at week 2 andmonths 1, 2, 3, 6, and

9. We assessed adherence with pill counts at each study visit.

Study AssessmentsLaboratory AssessmentsAll clinical chemistry analyses were performed by Quest Diagnos-

tics (Denver, CO). We obtained samples in a nonfasting state but

generally at the same time of day. Timed and spot urine chemistry

measurements were performed by Litholink Corp. (Chicago, IL).

Protein intake was estimated using the equation 6.253(urine urea

nitrogen in grams+30 mg/kg)+1 g/g proteinuria.5 g.29 Phosphorus

intake was estimated using the equation 11.83protein intake in

grams+78.30

1,25 Dihydroxy vitamin D assays were performed using a commer-

cially available radioimmunosassay (Diasorin Inc., Stillwater, MN).

We measured FGF23 in plasma using the second generation C-

terminal ELISA assay (Immutopics, San Clemente, CA); we also

measured intact FGF23 at baseline and month 9 using a sandwich

immunoassay (Kainos, Japan).

Imaging AssessmentsWeused the GE-Imatron C150 scanner at baseline andmonth 9 using

a standard protocol as previously described.31 We defined athero-

sclerotic calcium as a plaque area$1 mm2 with a density of $130

Hounsfield units. Total calcium volume score was derived by the sum

of all lesion volumes in cubic millimeters.32 The thoracic aorta was de-

fined as the segment from the aortic root to the diaphragm, whereas



the abdominal aorta was the segment from the diaphragm to the iliac

bifurcation. A single experienced investigator, blinded to treatment

assignment, performed all image assessments.

Lumbar BMD was determined using abdominal computed

tomography scans with a calibrated phantom of known density

(Image Analysis QCT 3D PLUS, Columbia, KY). Measurements of

BMD were performed in a 5-mm-thick slice of trabecular bone from

each vertebra (L2 to L4) at baseline and month 9.

EndpointsThe primary endpoint was change in serum phosphorus from base-

line (the average of the last two measures in screening) to the mean of

months 3, 6, and 9 among all active- versus all placebo-treated patients.

Secondary laboratory endpoints included changes in serum PTH,

FGF23, 1,25 dihydroxy vitamin D, urine phosphorus, and fractional

excretion of phosphorus. Secondary clinical endpoints included

change in coronary artery, thoracic and abdominal aorta calcium

volume scores, and change in lumbar BMD. Comparisons of indi-

vidual treatment arms versus placebo were considered exploratory.

Sample Size DeterminationWe estimated that a sample size of 150 (90 active versus 60 placebo)

would provide .80% power to detect a change in serum phos-

phorus$0.4 mg/dl between groups, with a two-sided a=0.05 and

an anticipated dropout rate of 25%.

Statistical AnalysesWe analyzed all data in accordance with the intent-to-treat principle

comparing all actively treated patients with all placebo-treated

patients. We imputed missing data using last observation carried

forward. We used analysis of covariance to examine change in

endpoints incorporating age, sex, race/ethnicity, diabetes, eGFR,

and baseline values as covariates. Because of the association of race

with PTH concentration, race/ethnicity was self-reported by subjects

from a prespecified list including Caucasian, African-American,

Hispanic or Latino, Asian, and American Indian, Alaskan Native,

or Native Hawaiian/Pacific Islander, or Other. We used a Dunnett

correction when comparing individual active arms with placebo. For

the analysis of change in calcification, we included only patients with

nonzero calcification at baseline. In companion analyses, we used the

Hokanson criterion to gauge progression of vascular calcification.26

For the analysis of change in BMD, we additionally considered

Quetélet’s (body mass) index and history of nontraumatic fracture

as covariates. We performed statistical analyses using SAS software

version 9.2 (SAS Institute Inc., Cary, NC).

ACKNOWLEDGMENTS

The authors would like to acknowledge Stephanie Brillhart, MSCI,

and Linda Loftin, NP, for their assistance in the conduct of the trial as

the designated unblinded study representatives for study drug stor-

age and assignment. The authors would also like to acknowledge

the important contribution of our Data Safety and Monitoring

Committee, led by the Chairman, Colin Baigent, FRCP. Members of

this committee donated their time and expertise in providing over-

sight of the trial conduct. In addition to Dr. Baigent, committee

members included Grahame Elder, Alistair Hutchison, FRCP, Jona-

than Emberson, PhD, and Edmung Ng.

G.A.B., D.C.W., M.S.P., B.K., M.K., D.M.S., J.A., R.T., M.W., and

G.M.C. participated in the design, analysis, interpretation, and manu-

script preparation. G.S., M.A.A., A.N.H., andM.M. participated in data

acquisition, analysis, interpretation, and manuscript preparation. L.K.

participated in the design and data acquisition. G.A.B. had full access to

all of the data in the study and takes responsibility for the integrity of the

data and the accuracy of the data analysis.

Funding for this investigator-initiated study was provided by

Shire, Inc., Fresenius NA, Genzyme, Inc., Denver Nephrologists, PC,

Novartis, Inc., and Davita, Inc.

Funding entities had no role in the design, conduct, analysis, in-

terpretation, or preparation of the manuscript. Each funding entity

was permitted to review the manuscript for verification that no

proprietary or confidential information was included.

DISCLOSURESG.A.B. has received research grants from Amgen, Roche, Cytochroma, and

CMD; received honoraria from Genzyme; has been on advisory boards or a

consultant for Amgen, Genzyme, KAI, Ardelyx, and Mitsubishi; and was a

Medical Director for Davita. D.C.W. received research grants fromGenzyme

and Abbott and honoraria from Amgen, Fresenius, and Shire; and has been

on the advisory board for KAI. M.S.P. has consulted for KAI. B.K. has

received a research grant from Amgen. M.K. has received research grants

from Amgen and Abbott; honoraria from Amgen, Abbott, Genzyme, FMC,

Medice, and Shire; and payment for development of educational presenta-

tions fromAmgen andAbbott; and has been a consultant for Amgen, Abbott,

Genzyme, FMC,Medice, and Shire. D.M.S. has received research grants from

Amgen and Keryx and honoraria from Amgen and Sanofi; has been on

advisory boards for Affymax, Amgen, and Sanofi; and has been a consultant

for Amgen. J.A. has been an employee of LabCorp and a consultant for

Oxthera Corp. .N.H. has received research grants fromWaters and has been

a consultant for Onconome. R.T. has received a research grant from Abbott;

has been a consultant for Fresenius North America and Mitsubishi Tanabe;

and has received honoraria from Shire. M.W. has received research grants

from National Institutes of Health, Shire, Amgen, and Genzyme; has received

honoraria from Abbott, Genzyme, and Shire; and has been a consultant for

Abbott, Genzyme, Luitpold, Mitsubishi, Cytochroma, Biotrends, Ardelyx, and

Astellas. G.M.C. has received a research grant from National Institute of

Diabetes and Digestive and Kidney Diseases, Amgen, and Reata; and has

been an advisor for Ardelyx. M.A.A., G.S., L.K., and M.M. have nothing to

disclose.

REFERENCES

1. US Renal Data System: USRDS 2008 Annual Data Report: Atlas ofChronic Kidney Disease and End-Stage Renal Disease in the UnitedStates, Bethesda, MD, National Institutes of Health, National Instituteof Diabetes and Digestive and Kidney Diseases, 2008

2. Go AS, Chertow GM, Fan D, McCulloch CE, Hsu CY: Chronic kidneydisease and the risks of death, cardiovascular events, and hospitaliza-tion. N Engl J Med 351: 1296–1305, 2004



3. Hasegawa H, Nagano N, Urakawa I, Yamazaki Y, Iijima K, Fujita T,Yamashita T, Fukumoto S, Shimada T: Direct evidence for a causativerole of FGF23 in the abnormal renal phosphate handling and vitamin Dmetabolism in rats with early-stage chronic kidney disease. Kidney Int78: 975–980, 2010

4. Stubbs JR, He N, Idiculla A, Gillihan R, Liu S, David V, Hong Y, QuarlesLD: Longitudinal evaluation of FGF23 changes andmineral metabolismabnormalities in a mouse model of chronic kidney disease. J BoneMiner Res , 201110.1002/jbmr.516

5. Isakova T, Xie H, Yang W, Xie D, Anderson AH, Scialla J, Wahl P,Gutiérrez OM, Steigerwalt S, He J, Schwartz S, Lo J, Ojo A, SondheimerJ, Hsu CY, Lash J, Leonard M, Kusek JW, Feldman HI, Wolf M; ChronicRenal Insufficiency Cohort (CRIC) StudyGroup: Fibroblast growth factor23 and risks of mortality and end-stage renal disease in patients withchronic kidney disease. JAMA 305: 2432–2439, 2011

6. TengM,WolfM,OfsthunMN, Lazarus JM, HernánMA, CamargoCA Jr,Thadhani R: Activated injectable vitamin D and hemodialysis survival: Ahistorical cohort study. J Am Soc Nephrol 16: 1115–1125, 2005

7. Danese MD, Belozeroff V, Smirnakis K, Rothman KJ: Consistent controlof mineral and bone disorder in incident hemodialysis patients. Clin JAm Soc Nephrol 3: 1423–1429, 2008

8. ChueCD, EdwardsNC,WilliamsME, Steeds RP, Townend JN, FerroCJ:Serumphosphate is associatedwith left ventricularmass in patients withchronic kidney disease: A cardiac magnetic resonance study. Heart 98:219–224, 2012

9. Faul C, Amaral AP, Oskouei B, Hu MC, Sloan A, Isakova T, GutiérrezOM, Aguillon-Prada R, Lincoln J, Hare JM,Mundel P, Morales A, SciallaJ, Fischer M, Soliman EZ, Chen J, Go AS, Rosas SE, Nessel L, TownsendRR, Feldman HI, St John Sutton M, Ojo A, Gadegbeku C, Di Marco GS,Reuter S, Kentrup D, Tiemann K, BrandM, Hill JA, MoeOW, Kuro-OM,Kusek JW, Keane MG, Wolf M: FGF23 induces left ventricular hyper-trophy. J Clin Invest 121: 4393–4408, 2011

10. Kovesdy CP, Kuchmak O, Lu JL, Kalantar-Zadeh K: Outcomes associ-ated with phosphorus binders in men with non-dialysis-dependentCKD. Am J Kidney Dis 56: 842–851, 2010

11. National Kidney Foundation: K/DOQI clinical practice guidelines forbone metabolism and disease in chronic kidney disease. Am J KidneyDis 42[Suppl 3]: S1–S201, 2003

12. Dhingra R, Sullivan LM, Fox CS, Wang TJ, D’Agostino RB Sr, GazianoJM, Vasan RS: Relations of serum phosphorus and calcium levels to theincidence of cardiovascular disease in the community. Arch InternMed167: 879–885, 2007

13. Kestenbaum B, Sampson JN, Rudser KD, Patterson DJ, Seliger SL,Young B, Sherrard DJ, Andress DL: Serum phosphate levels and mor-tality risk amongpeoplewith chronic kidney disease. J AmSocNephrol16: 520–528, 2005

14. Tonelli M, Sacks F, Pfeffer M, Gao Z, Curhan G; Cholesterol And Re-current Events Trial Investigators: Relation between serum phosphatelevel and cardiovascular event rate in people with coronary disease.Circulation 112: 2627–2633, 2005

15. Foley RN, Collins AJ, Ishani A, Kalra PA: Calcium-phosphate levels andcardiovascular disease in community-dwelling adults: The Atheroscle-rosis Risk in Communities (ARIC) Study.AmHeart J 156: 556–563, 2008

16. Zoccali C, Ruggenenti P, Perna A, Leonardis D, Tripepi R, Tripepi G,Mallamaci F, Remuzzi G; REIN Study Group: Phosphate may promoteCKD progression and attenuate renoprotective effect of ACE in-hibition. J Am Soc Nephrol 22: 1923–1930, 2011

17. Oliveira RB, Cancela AL, Graciolli FG, Dos Reis LM, Draibe SA, CuppariL, Carvalho AB, Jorgetti V, Canziani ME, Moysés RM: Early control ofPTH and FGF23 in normophosphatemic CKD patients: A new target inCKD-MBD therapy? Clin J Am Soc Nephrol 5: 286–291, 2010

18. Portale AA, Halloran BP, Morris RC Jr: Dietary intake of phosphorusmodulates the circadian rhythm in serum concentration of phosphorus.Implications for the renal production of 1,25-dihydroxyvitamin D. J ClinInvest 80: 1147–1154, 1987

19. Ohi A, Hanabusa E, Ueda O, Segawa H, Horiba N, Kaneko I, KuwaharaS, Mukai T, Sasaki S, Tominaga R, Furutani J, Aranami F, Ohtomo S,Oikawa Y, Kawase Y, Wada NA, Tachibe T, Kakefuda M, Tateishi H,Matsumoto K, Tatsumi S, Kido S, Fukushima N, Jishage K, Miyamoto K:Inorganic phosphate homeostasis in sodium-dependent phosphatecotransporter Npt2b⁺/⁻ mice. Am J Physiol Renal Physiol 301: F1105–F1113, 2011

20. Sabbagh Y, O’Brien SP, Song W, Boulanger JH, Stockmann A,Arbeeny C, Schiavi SC: Intestinal npt2b plays a major role in phos-phate absorption and homeostasis. J Am Soc Nephrol 20: 2348–2358, 2009

21. Ix JH, Ganjoo P, Tipping D, Tershakovec AM, Bostom AG: Sustainedhypophosphatemic effect of once-daily niacin/laropiprant in dyslipi-demic CKD stage 3 patients. Am J Kidney Dis 57: 963–965, 2011

22. Isakova T, Wahl P, Vargas GS, Gutiérrez OM, Scialla J, Xie H, ApplebyD, Nessel L, Bellovich K, Chen J, Hamm L, Gadegbeku C, Horwitz E,Townsend RR, Anderson CA, Lash JP, Hsu CY, Leonard MB, Wolf M: Fi-broblast growth factor 23 is elevated before parathyroid hormone andphosphate in chronic kidney disease. Kidney Int 79: 1370–1378, 2011

23. Isakova T, Gutierrez OM, Smith K, Epstein M, Keating LK, Jappner H,Wolf M: Pilot study of dietary phosphorus restriction and phosphorusbinders to target fibroblast growth factor 23 in patients with chronickidney disease.Nephrol Dial Transplant 26: 584–591, 2011

24. Burnett SM, Gunawardene SC, Bringhurst FR, Jüppner H, Lee H,Finkelstein JS: Regulation of C-terminal and intact FGF-23 by dietaryphosphate in men and women. J BoneMiner Res 21: 1187–1196, 2006

25. Yilmaz MI, Sonmez A, SaglamM, Yaman H, Kilic S, Eyileten T, Caglar K,Oguz Y, Vural A, Yenicesu M, Mallamaci F, Zoccali C: Comparison ofcalcium acetate and sevelamer on vascular function and fibroblastgrowth factor 23 in CKD patients: A randomized clinical trial. Am JKidney Dis 59: 177–185, 2012

26. Hokanson JE, MacKenzie T, Kinney G, Snell-Bergeon JK, Dabelea D,Ehrlich J, Eckel RH, Rewers M: Evaluating changes in coronary arterycalcium: An analytic method that accounts for interscan variability. AJRAm J Roentgenol 182: 1327–1332, 2004

27. Budoff M, Hokanson JE, Nasir K, Shaw LJ, Kinney GL, ChowD, DemossDNuguri V, Nabavi V, Ratakonda R, BermanDS, Raggi P: Progression ofcoronary artery calcium predicts all-cause mortality. JACC CardiovascImaging 3: 1229–1236, 2010

28. Spiegel DM, Brady K: Calcium balance in normal individuals and inpatients with chronic kidney disease on low- and high-calcium intake.Kidney Int 81: 1116–1122, 2012

29. Russo D, Miranda I, Ruocco C, Battaglia Y, Buonanno E, Manzi S, RussoL, Scafarto A, Andreucci YE: The progression of coronary artery calci-fication in pre-dialysis patients on calcium carbonate or sevelamer.Kidney Int 72: 1255–1261, 2007

30. Kalantar-Zadeh K, Gutekunst L, Mehrotra R, Kovesdy CP, Bross R,Shinaberger CS, Noori N, Hirschberg R, Benner D, Nissenson AR,Kopple JD: Understanding sources of dietary phosphorus in thetreatment of patients with chronic kidney disease. Clin J Am SocNephrol 5: 519–530, 2010

31. Spiegel DM, Raggi P, Mehta R, Lindberg JS, Chonchol M, Ehrlich J,James G, Chertow GM, Block GA: Coronary and aortic calcifications inpatients new to dialysis. Hemodial Int 8: 265–272, 2004

32. Callister TQ, Raggi P, Cooil B, Lippolis NJ, Russo DJ: Effect of HMG-CoAreductase inhibitors on coronary artery disease as assessed by electron-beam computed tomography. N Engl J Med 339: 1972–1978, 1998

See related editorial, “Phosphate Binders in CKD: Bad News or Good News?,”on pages 1277–1280.

This article contains supplemental material online at http://jasn.asnjournals.org/lookup/suppl/doi:10.1681/ASN.2012030223/-/DCSupplemental.





effects of phosphate binders in moderate ckd€¦ · enrollment, baseline characteristics, and...

Documents