- Email: [email protected]
Meta-Analysis of Biochemical and Patient-Level Effects of Calcimimetic Therapy
AJKD
VOL 47, NO 5, MAY 2006
American Journal of Kidney Diseases
REVIEWS
Meta-Analysis of Biochemical and Patient-Level Effects of Calcimimetic Therapy Giovanni F.M. Strippoli, MD, Suetonia Palmer, MBChB, Allison Tong, BMedSc, Grahame Elder, MBBS, PhD, Piergiorgio Messa, MD, and Jonathan C. Craig, MBChB, PhD ● Background: Many randomized trials have now evaluated the effects of calcimimetics in patients with chronic kidney disease and secondary hyperparathyroidism (SHPT) on standard therapy with vitamin D and/or phosphate binders. We conducted a meta-analysis to evaluate outcomes of therapy with these novel agents. Methods: MEDLINE, EMBASE, the Cochrane Controlled Trials Register, and conference proceedings were searched for randomized controlled trials evaluating any calcimimetic against placebo or another agent in predialysis or dialysis patients with chronic kidney disease. Data were extracted for all relevant patient-centered and surrogate outcomes. Analysis was by means of a random-effects model, and results are expressed as relative risk or weighted mean difference (WMD) with 95% confidence intervals (CIs). Results: Eight trials (1,429 patients) were identified that compared a calcimimetic agent plus standard therapy with placebo plus standard therapy. End-of-treatment values for parathyroid hormone (4 trials; 1,278 patients; WMD, ⴚ290.49 pg/mL; 95% CI, ⴚ359.91 to ⴚ221.07), serum calcium (3 trials; 1,201 patients; WMD, ⴚ0.85 mg/dL; 95% CI, ⴚ1.14 to ⴚ0.56), serum phosphorus (3 trials; 1,195 patients; WMD, ⴚ0.29 mg/dL; 95% CI, ⴚ0.50 to ⴚ0.08), and calcium ⴛ phosphorus product (3 trials; 1,194 patients; WMD, ⴚ7.90 mg2/dL2; 95% CI, ⴚ10.25 to ⴚ5.54) were significantly lower with calcimimetic therapy compared with placebo. No significant effects on patient-based end points were shown. Conclusion: Calcimimetic treatment of patients with SHPT leads to significant improvements in biochemical parameters that observational studies have associated with increased mortality, cardiovascular risk, and osteitis fibrosa, but patient-based benefits have not yet been shown. For patients with SHPT, the benefits of calcimimetics over standard therapy remain uncertain until additional randomized trials become available. Am J Kidney Dis 47:715-726. © 2006 by the National Kidney Foundation, Inc. INDEX WORDS : Calcimimetic; secondary hyperparathyroidism; renal bone disease.
A
BNORMALITIES OF CALCIUM and phosphorus metabolism are present in all individuals with chronic kidney disease (CKD). These biochemical changes cause many bone and metabolic disorders, including renal osteodystrophy,
which is characterized by abnormalities of bone turnover (ranging from high-turnover osteitis fibrosa to adynamic bone disease); mineralization defects (osteomalacia); and architectural change. Renal osteodystrophy is associated not only with
From the NHMRC Centre for Clinical Research Excellence in Renal Medicine, Cochrane Renal Group, and School of Public Health, University of Sydney; Centre for Transplant and Renal Research, Westmead Millennium Institute, Sydney, Australia; Department of Nephrology, Christchurch Hospital, Christchurch, NZ; and Department of Nephrology and Urology, Ospedale Maggiore-Policlinico di Milano, Milan, Italy. Received October 19, 2005; accepted in revised form January 26, 2006.
Originally published online as doi:10.1053/j.ajkd.2006.01.015 on March 17, 2006. Support: None. Potential conflicts of interest: None. Address reprint requests to Giovanni F.M. Strippoli, MD, NHMRC Centre for Clinical Research Excellence in Renal Medicine, Cochrane Renal Group, University of Sydney, Australia. E-mail: [email protected] © 2006 by the National Kidney Foundation, Inc. 0272-6386/06/4705-0001$32.00/0 doi:10.1053/j.ajkd.2006.01.015
American Journal of Kidney Diseases, Vol 47, No 5 (May), 2006: pp 715-726
715
716
an increased incidence of fracture, bone and muscular pain, and abnormalities of bone and joint morphological characteristics, but also vascular and soft-tissue calcification. In addition to causing decreased quality of life, these complications and the associated abnormalities of elevated phosphorus, calcium, calcium ⫻ phosphorus product, and parathyroid hormone (PTH) levels have been associated with increased mortality.1-6 Standard management of patients with CKD, particularly those on dialysis therapy, includes treatment to control calcium, phosphorus, and PTH levels to prevent bone and soft-tissue complications. Based on a number of association studies,3-5,7,8 including studies of bone histomorphometric characteristics,9-12 optimal ranges for serum phosphorus, calcium, calcium ⫻ phosphorus product, and PTH levels have been suggested.13,14 However, success in achieving these targets has been limited.15 Specific management of patients with secondary hyperparathyroidism (SHPT) with CKD stages 3 and 4 may be accomplished by restriction of dietary phosphorus, calcium supplementation, and/or the use of vitamin D. When patients have started dialysis therapy, standard therapy for SHPT generally includes calcitriol or vitamin D analogues, calcium-based and/or newer phosphate-binding agents, and parathyroidectomy.16,17 Recently, use of a novel class of drugs, the calcimimetics, was proposed as a strategy to decrease PTH secretion and possibly decrease parathyroid cell proliferation, while decreasing serum calcium, phosphorus, and calcium ⫻ phosphorus product levels.18,19 Results of randomized trials testing the efficacy and safety of calcimimetics in patients undergoing dialysis are becoming available. With the aim of preventing complications associated with SHPT, cinacalcet hydrochloride (HCl) has been incorporated into many treatment algorithms. However, several aspects of calcimimetic therapy require further evaluation. In the dialysis population, for which these drugs are approved, the most important question is the degree to which calcimimetics impact on such clinically relevant end points as parathyroidectomy rates, fracture, renal osteodystrophy, and cardiovascular disease and death, as well as such surrogate markers for these conditions as abnormalities of
STRIPPOLI ET AL
serum calcium and phosphorus levels. Other important questions include the optimal time for the start of calcimimetic therapy, the influence of calcimimetics on standard treatment regimens, and the effectiveness of calcimimetics in patients with different stages of CKD and after transplantation. This systematic review was performed at an early phase of calcimimetic use, but 1 year after licensing by the Food and Drug Administration,20 to assess available efficacy and safety data and highlight research questions that need additional investigation. METHODS
Inclusion Criteria We included all randomized controlled trials of any calcimimetic agent, cinacalcet HCl (AMG-073, Sensipar; Amgen Inc, Thousand Oaks, CA), NPS R-467, or NPS R568 administered to patients with CKD for the treatment of SHPT.
Search Strategy Electronic searches were performed in MEDLINE (1966 to March 2005), EMBASE (1980 to March 2005), and the Cochrane Renal Group Specialized Register by using optimally sensitive search strategies for the identification of randomized trials developed by the Cochrane Collaboration (full details available upon request). Medical subject headings and text words used were relevant to the intervention (all calcimimetic agents) and the condition of interest (SHPT, calcium, phosphorus, calcium ⫻ phosphorus product, levels of PTH and biochemical targets, bone disease, renal osteodystrophy, and cardiovascular disease). Results of these searches were analyzed in title and abstract form by 2 authors according to the inclusion criteria in consultation with a third author. Reference lists from the identified articles also were searched, and information about unpublished or ongoing trials was sought from experts in the field and pharmaceutical companies. Trials were considered without language restriction.
Data Extraction and Trial Quality Assessment Each trial was assessed by 2 independent authors. From all included trials, data were extracted for study sample characteristics; type of agent, dose, and route of administration; and trial methods and outcomes. The following outcome measures were considered: end-of-treatment values for intact PTH (iPTH) and/or PTH(1-84), 30% or greater decrease in PTH levels from baseline, serum calcium level, hypocalcemia, serum phosphorus and calcium ⫻ phosphorus product levels, cardiovascular events, cardiovascular and all-cause mortality, fractures at any site, bone histomorphometry, bone mineral density, bone pain, muscle pain or weakness, paresthesias, nausea, vomiting, dyspnea, hospitalization, and quality of life. Methods and quality of included trials were assessed by using standard criteria (allocation concealment; blinding of participants, investigators, and
CALCIMIMETICS IN CHRONIC KIDNEY DISEASE
outcome assessors; analysis by intention to treat, and completeness of follow-up). Discrepancies in data extraction were resolved by discussion among the authors, and when data were missing or incomplete, the investigators of the trial were contacted for clarification.
Statistical Analysis The estimate of effect of an experimental versus a control intervention on categorical outcomes (eg, fracture rate and all-cause mortality, including sudden death) was analyzed by using the relative risk (RR) measure and its 95% confidence interval (CI) for each trial. For continuous variables, the weighted mean difference (WMD) and its 95% CI were calculated by using end-of-treatment values for variables in the experimental and control groups. Heterogeneity of treatment effects between studies was tested formally by using the Q (heterogeneity chi-square) and I2 statistics.21 When appropriate, summary estimators of treatment effects were calculated by using a random-effects model and expressed as RR or WMD with 95% CIs. Forest plots were used for graphical representation of data. In this type of graphic, vertical lines positioned at 1 for RR and 0 for WMD are consistent with no difference in efficacy between the experimental and control treatments (null hypothesis). Trials shown in the area on the left show a reduction in risk or lower WMD with the experimental intervention (eg, favorable effect on the adverse outcome), whereas those on the right show an increase in risk or greater WMD with the experimental intervention. The numeric scale at the bottom indicates the value of the RR or WMD of the treatment group compared with the control group. A solid square and horizontal line graphically represent the RR or WMD with 95% CIs. The area of the black square indicates how much the trial contributed to the analysis in terms of sample size and number of events recorded (weight). Horizontal lines indicate 95% CIs, with large informative trials having smaller CIs. The diamond-shaped symbol is the summary estimate of effect expressed as either an RR or WMD with 95% CIs, which is a weighted average of the summative treatment effect across all trials.
RESULTS
Our search for randomized trials of calcimimetic interventions to treat patients with SHPT identified 186 articles. Of these, 148 articles were excluded after title and abstract review because they clearly were ineligible (nonrandomized studies, randomized trials of interventions not relevant to SHPT, noncalcimimetic interventions, duplicate articles of the same trial, or review articles; Fig 1). Of the remaining 38 potentially eligible studies (either full-text or abstract publications), 30 studies were excluded because we could not confirm from the full-text analysis or from contacting the investigators that they were randomized trials or not a duplicate publication. Two attempts were made to contact all investigators of the trials to clarify study
717
designs and request supplemental data, but we were not able to obtain some of the data or ascertain whether some reports22-31 (presented in abstract form at the American Society of Nephrology and the European Renal Association– European Dialysis and Transplantation Association meetings of 2003 and 2004) were subsets of other publications that subsequently appeared as full-text articles in scientific journals or were unique unpublished trials. These studies therefore could not be included. Eight trials (8 publications) enrolling 1,429 patients finally were included.18,32-38 Trial Characteristics Characteristics of the populations and interventions of all studies included in this systematic review are listed in Table 1.18,32-38 There also were 2 ongoing trials.39,40 Characteristics of the 11 potentially eligible studies for which we were unable to ascertain eligibility are listed in Table 2. The 8 trials included compared cinacalcet HCl (845 patients) versus placebo (584 patients). Three of these trials reported cinacalcet HCl as AMG073. One trial reported on the first-generation calcimimetic R-568. This drug has been withdrawn from clinical use because of poor bioavailability, variable serum concentrations, and potential drug interactions caused by cytochrome P-450 activity.18,41 Because there were no differences in results of analyses performed with and without inclusion of this trial, combined results are reported. In addition to these randomized interventions, most patients were administered an active vitamin D for suppression of PTH and phosphate binders for management of hyperphosphatemia as cointerventions in all trials in a nonrandomized fashion. There were no significant differences in proportions of patients prescribed calcitriol, vitamin D analogues, and phosphate binders as cointerventions between the calcimimetic and placebo groups of the trials. Entry to some trials was restricted when patients had severe SHPT (eg, iPTH ⱖ 800 pg/mL [ⱖ800 ng/L]), whereas others stratified patients according to severity of hyperparathyroidism. Mean age of patients enrolled in the trials ranged from 47 to 55 years. All patients had SHPT. On average, a greater proportion of males were enrolled in the trials (388
718
STRIPPOLI ET AL
Fig 1. Flow chart indicating the number of citations retrieved by individual searches and the final number of included trials: reasons for exclusion are provided, as well as the number of trials providing outcomes for analysis and number of patients they enrolled.
males compared with 220 females in the 6 trials that reported sex distribution). Follow-up of studies ranged from 3 weeks to 26 weeks. All studies were supported by Amgen Inc, Thousand Oaks, CA, which holds the cinacalcet HCl patent. Of note, the largest report published32 was based on pooled results of 2 separate trials. Method Quality of Included Studies By current standards, the method quality of all except 2 trials was suboptimal (Table 3), although all trials were published after release of the Consolidated Standards for Reporting
Trials (CONSORT) Statement (a checklist designed to improve quality of trial reporting).42 Allocation concealment was unclear in 6 of 8 trials (75%). All studies were reported as double blinded (to indicate that participants and investigators were blinded to the intervention), but no study mentioned the blinding of outcome assessors. Intention-to-treat analysis (ie, analysis of all patients according to randomized allocations, rather than actual intervention) was performed in 3 of 8 trials (38%). The proportion of patients lost to follow-up ranged from 0% to 56%. The proportion of patients who stopped taking the medication was un-
Interventions
Reference Block et al,*32 2004
No. of Patients (treatment/ control) 741 (371/370)
Inclusion Criteria
Exclusion Criteria
Hemodialysis ⱖ 3 mo iPTH ⱖ 300 pg/mL Serum calcium ⱖ 8.4 mg/dL
Infection, malignancy, diseases causing hypercalcemia, tricyclic antidepressant use Malignancy, diseases causing hypercalcemia, myocardial infarction previous 6 mo, tricyclic antidepressant use Malignancy, diseases causing hypercalcemia, myocardial infarction previous 6 mo, tricyclic antidepressant use, transaminase or bilirubin ⬎ 2 ⫻ upper limit of normal Not available
Goodman et al,18 2000
21 (16/5)
Hemodialysis ⱖ 3 mo iPTH, 300-1,200 pg/mL Serum calcium ⬎ 9.0 mg/dL
Goodman et al,33 2002
52 (40/12)
Hemodialysis ⱖ 3 mo iPTH, 250-1,500 pg/mL Serum calcium ⬎ 9.0 mg/dL
Harris et al,34 2004
23 (17/6)
Hemodialysis Serum calcium ⱖ 8.4 mg/dL
Lindberg et al,35 2003
78 (39/39)
Lindberg et al,36 2005
395 (294/101)
Malluche et al,37 2004 (ERA abstract)
48 (32/16)
Hemodialysis ⱖ 3 months iPTH ⱖ 300 pg/mL Serum calcium, 8.8-11.0 mg/dL Calcium ⫻ phosphorus ⬍ 70 mg2/dL2 Hemodialysis and peritoneal dialysis (continuous ambulatory PD, automated PD) ⬎ 1 mo iPTH ⱖ 300 pg/mL Serum calcium ⱖ 8.4 mg/dL Serum phosphorus ⱖ 2.5 mg/dL Calcium ⫻ phosphorus ⱖ 70 mg2/dL2 Stable vitamin D ⬎ 30 d iPTH ⱖ 300 pg/mL
Quarles et al,38 2003
71 (36/35)
Hemodialysis at least 3 mo iPTH ⱖ 300 pg/mL Serum calcium, 8.8-11.0 mg/dL Calcium ⫻ phosporus ⬍ 70 mg2/dL2 Stable vitamin D ⬎ 21 d
Treatment Group (target PTH)
Cointerventions (type, %)
Control
Treatment Group
Control
Duration (wk)
Cinacalcet, 30-180 mg/d (iPTH ⬍ 250 pg/mL)
Placebo
Vitamin D, 66%; phosphate binder, 92%
Vitamin D, 67%; phosphate binder, 93%
26 Titration, 12; maintenance, 14
R-568, 100 mg/d
Placebo
Vitamin D, 38%; phosphate binder, 31%
Vitamin D, 50%; phosphate binder, 50%
3
AMG073,* 5-100 mg/d
Placebo
Not available
Not available
2
Cinacalcet, 25 mg-300 mg/d
Placebo
AMG073,* 10-50 mg/d (iPTH decrease ⱖ 30%)
Placebo
Vitamin D, 60%; phosphate binder, 80% Vitamin D, 67%; phosphate binder, 87%
1
Infection, malignancy, diseases causing hypercalcemia, hepatic transaminase or bilirubin ⬎ 2 ⫻ upper limit of normal
Vitamin D, 71%; phosphate binder, 100% Vitamin D, 62%; phosphate binder, 87%
Parathyroidectomy, previous myocardial infarction (3 mo)
Cinacalcet, 30-180 mg/d (iPTH ⬍ 250 pg/mL)
Placebo
Not available
Not available
26 Titration, 16; maintenance, 10
Not available
Cinacalcet, 30-180 mg/d (iPTH ⱕ 250 pg/mL)
Placebo
Not available
Not available
Infection, malignancy, diseases causing hypercalcemia, hepatic transaminase or bilirubin ⬎ 2 ⫻ upper limit of normal
AMG073,* 25-100 mg/d (iPTH decrease ⱖ 30%)
Placebo
Vitamin D, 61%; phosphate binder, 100%
Vitamin D, 69%; phosphate binder, 94%
52 Titration, 24; maintenance, 28 18 Titration, 12; maintenance, 6
18 Titration, 12; maintenance, 6
719
NOTE. To convert PTH in pg/mL to ng/L, multiply by 1; calcium in mg/dL to mmol/L, multiply by 0.2495; phosphorus in mg/dL to mmol/L, multiply by 0.3229. *AMG073 and cinacalcet are the same compound.
CALCIMIMETICS IN CHRONIC KIDNEY DISEASE
Table 1. Characteristics of Trials of Calcimimetic Agents for SHPT in Patients With CKD
720
STRIPPOLI ET AL Table 2. Characteristics of Ongoing Trials or Trials With Insufficient Data to Assess Eligibility for Inclusion in This Systematic Review Interventions
Study and Reference
No. of Patients (treatment/control)
Inclusion Criteria
Treatment Group (target PTH)
Control
Study Duration (wk)
Major Reason for Inability to Include in This Review
Studies identified as “ongoing trials,” but reported in only abstract form with insufficient data for inclusion TARGET39 424 (NA/NA) Mild/moderate SHPT Cinacalcet (dose Placebo 8-14 Data still unpublished iPTH, 300-800 pg/mL NA) Preliminary data NA CONTROL40 76 (NA/NA) KDOQI iPTH target Cinacalcet, 30-180 Placebo 8 Data still unpublished (150-300 pg/mL) mg/d (iPTH, Preliminary data NA Calcium ⫻ phosphorus 150-300 pg/mL) target outside KDOQI range Studies in abstract form with insufficient data available to assess whether they were randomized trials and could be included in this analysis 54 (19/21) Nondialysis Cinacalcet, 30-180 Placebo 18 Data NA to assess if a randomized Coburn et al,22 2003 iPTH ⱖ 130 pg/mL mg/d (iPTH study decrease ⱖ 30%) 121 (NA/NA) Hemodialysis and Cinacalcet, 30-180 Placebo 24 Analysis of 3 trials, unconfirmed if Goodman et al,23 2003 peritoneal dialysis mg/d randomized data ⬍ 12 mo iPTH ⱖ 300 pg/mL 410 (205/205) Hemodialysis Cinacalcet (dose Placebo 26 Possible (unconfirmed data) Martin et al,24 2003 iPTH ⱖ 300 pg/mL NA) (iPTH duplicate of a previous ⱕ 250 pg/mL) publication 1,136 (665/471) Hemodialysis and Cinacalcet (dose Placebo 24 3 trials, possible (unconfirmed data) Quarles et al,25 2003 peritoneal dialysis NA) duplicate of a previous iPTH ⱖ 300 pg/mL publication 26 210 (99/111) Hemodialysis Cinacalcet (dose Placebo 48 2 trials, possible (unconfirmed data) Sterrett et al, 2004 iPTH ⱖ 300 pg/mL NA) duplicate of a previous publication 297 (182/115) Hemodialysis Cinacalcet ⬎ 180 Placebo 208 Analysis of 5 trials, unconfirmed if Cunningham et al,27 2003 iPTH ⱖ 300 pg/mL mg/d all randomized data Cunningham et al,28 2003 1,184 (697/487) Hemodialysis Cinacalcet (dose Placebo 72 5 trials, possible (unconfirmed data) iPTH ⱖ 300 pg/mL NA) (iPTH, 250 duplicate of a previous pg/mL) publication 29 NA Renal replacement Cinacalcet, 30-180 Placebo Fournier et al, 2004 NA 3 trials, possible (unconfirmed data) therapy mg/d (iPTH duplicate of a previous iPTH ⱖ 300 pg/mL ⱕ 250 pg/mL) publication 311 (139/150) Hemodialysis Cinacalcet, 30-180 Placebo NA Possible (unconfirmed data) de Francisco et al,30 2004 iPTH ⱖ 300 pg/mL mg/d (iPTH duplicate of a previous ⱕ 250 pg/mL) publication 31 1,136 (556/471) Renal replacement Cinacalcet ⬎ 180 Frazao et al, 2004 Placebo NA Analysis of 3 trials, unconfirmed if therapy mg/d (iPTH randomized data, possible iPTH ⱖ 300 pg/mL ⱕ 250 pg/mL) (unconfirmed data) duplicate of a previous publication
NOTE. To convert PTH in pg/mL to ng/L, multiply by 1. Abbreviations: TARGET, treatment strategies to achieve recommended KDOQI goals in ESRD patients on cinacalcet; CONTROL, enhanced achievement of NKF-KDOQI bone metabolism and disease targets using cinacalcet HCl (Sensipar); NA, not available; KDOQI, Kidney Disease Outcomes Quality Initiative.
clear from reports of most trials and either was not reported or reported to be minimal. Trial Results: Patient-Level Outcomes All-cause mortality was not significantly different with calcimimetics compared with placebo (5 trials conducted during 0.5 to 6.0 months; 1,285 patients; RR, 0.75; 95% CI, 0.30 to 1.88), with no significant heterogeneity in this analysis. The effect of calcimimetics on other patientlevel outcomes was extracted from individual trials, and summary estimates of effect are listed in Table 4. Fractures and other clinically impor-
tant measures of renal bone disease were not reported by any trial except for 2 abstracts; in one of these trials, data reported were insufficient to ascertain eligibility,27 and in the other, data were insufficient to estimate observed treatment effects.37 There was no report that rates of hospital admission or study withdrawal differed between patients treated with calcimimetic or placebo. Treatment with calcimimetic or placebo did not result in significant differences in the use of phosphate binders, calcitriol, or vitamin D analogues (reported in 5 trials). Risk for nausea and vomiting increased with calcimimetic
CALCIMIMETICS IN CHRONIC KIDNEY DISEASE
721
Table 3. Assessment of Method Quality of Randomized Controlled Trials of Calcimimetic Agents Included in This Analysis Blinding
Reference
Block et al,32 2004 Goodman et al,18 2000 Goodman et al,33 2002 Harris et al,34 2004 Lindberg et al,35 2003 Lindberg et al,36 2005 Malluche et al,37 2004 Quarles et al,38 2003
Allocation Concealment*
Not stated Not stated Not stated Not stated Not stated Adequate Not stated Adequate
Participants
Investigators
Outcome Assessors
Yes Yes Yes Yes Yes Yes Yes Yes
Yes Yes Yes Yes Yes Yes Yes Yes
No No No No No No No No
Intentionto-Treat Analysis
Lost to FollowUp (%)†
No No Yes No No Yes NA Yes
202/741 (27) 7/20 (56) 0/52 (0) 10/23 (40) 11/78 (14) 101/395 (26) NA 6/71 (8)
Abbreviation: NA, data not available in abstract form or obtainable when contacting investigators. *Defined as adequate when sequentially labeled, sealed, opaque envelopes or a central or pharmacy randomization were used; inadequate when a pseudorandomization method including alternation, date of birth, or chart number was used; not stated if no information was available. †Including patients described as lost to follow-up or those who were discontinued and who could have had outcomes measured.
therapy, but hypotension was less common with calcimimetics. These results should be interpreted with caution because few trials reported these outcomes. Trial Results: Biochemical Outcomes The end-of-treatment value for PTH was significantly less with calcimimetics compared with
placebo (4 trials conducted during 0.5 to 7.0 months; 1,278 patients; WMD, ⫺290.49 pg/mL; 95% CI, ⫺359.91 to ⫺221.07; Fig 2A). Similarly, end-of-treatment values for serum calcium (3 trials; 1,201 patients; WMD, ⫺0.85 mg/dL; 95% CI, ⫺1.14 to ⫺0.56; Fig 2B), serum phosphorus (1,195 patients; WMD, ⫺0.29 mg/dL; 95% CI, ⫺0.50 to ⫺0.08; Fig 2C), and calcium ⫻
Table 4. Summary Estimates of Effect of Treatment With Calcimimetics Compared With Placebo in Populations Enrolled in Trials Included in This Systematic Review Outcome
No. of Trials
No. of Patients
RR
95% CI
I2 (%)
P (heterogeneity)
All-cause mortality ⱖ 30% Decrease in mean PTH level Fractures Hypocalcemia* Nausea* Vomiting Dyspnea Muscle weakness or paresthesia Hypotension Upper respiratory tract infection Parathyroidectomy Headache Asthenia Abdominal pain Diarrhea Mixed uremic osteodystrophy Bone histomorphometry
5 4 — 4 5 2 1 1 2 2 1 1 1 1 1 1 —
1,285 1,284 — 868 1,263 1,136 77 20 1,136 1,136 395 395 396 395 395 32 —
0.75 4.49 — 2.89 1.40 1.89 1.54 3.24 0.53 0.86 0.07 0.86 4.12 0.67 1.28 0.34 —
0.30-1.88 3.04-6.64 — 0.71-11.73 1.00-1.95 1.47-2.43 0.44-5.34 0.21-49.01 0.36-0.79 0.34-2.18 0.00-1.43 0.54-1.37 0.99-17.14 0.40-1.13 0.82-2.02 0.07-1.60 —
0.0 47.3 — 0.0 0.0 0.0 NA NA 0.0 85.0 NA NA NA NA NA NA —
0.63 0.13 — 0.78 0.91 0.91 NA NA 0.66 0.01 NA NA NA NA NA NA —
Abbreviation: NA, not applicable because events occurred in 1 trial only. *For outcomes that may occur more than once, number of patients developing 1 or more episodes of that outcome was considered.
722
STRIPPOLI ET AL
Fig 2. Effect of calcimimetic agents compared with placebo on values for (A) PTH, (B) calcium, (C) phosphorus, and (D) calcium ⴛ phosphorus product. To convert PTH in pg/mL to ng/L, multiply by 1; calcium in mg/dL to mmol/L, multiply by 0.2495; phosphorus in mg/dL to mmol/L, multiply by 0.3229.
phosphorus product (3 trials; 1,204 patients; WMD, ⫺7.90 mg2/dL2; 95% CI, ⫺10.25 to ⫺5.54; Fig 2D) were significantly less with calcimimetics than placebo. DISCUSSION
Trials of calcimimetic compounds in patients with CKD are restricted almost exclusively to hemodialysis patients with SHPT, for whom calcimimetics have been added to standard therapy with outcomes measured up to 6 months. In this population, calcimimetic agents compared with placebo were effective in decreasing levels of PTH, serum calcium, phosphorus, and calcium ⫻ phosphorus product. All-cause mortality was reported in 5 studies and not reported in 3 studies. Analysis of these studies suggests that cinacalcet HCl may be associated with a 25% RR for decrease in all-cause mortality, but the estimate is very imprecise, with wide CIs ranging from a 70% decrease to a 90% increase in risk. No significant differences between calcimimetics and placebo were shown for fractures or other patient-based end points. This may be caused by insufficient power
because these events are relatively infrequent, artifactual lack of power by failure to measure and/or report patient-based outcomes, or a true lack of benefit. The annual mortality of patients undergoing dialysis is high, at approximately 15 deaths/100 patient-years at risk, of which approximately 40% are cardiac and 11% are vascular.43 Cinacalcet HCl therefore is a potentially important intervention to improve survival, but existing randomized trials have not yet confirmed the hypothesized benefits of this drug. In general, trials reported transient and mild to moderate toxicity, with only a few participants withdrawing from therapy because of adverse events. However, rates of withdrawal and valid precise estimates of such adverse effects as hypotension and nausea often were not available. One trial32 reported that in the calcimimetic group, 6 of 17 patients withdrew because of gastrointestinal events (nausea/vomiting). The few trials reporting toxicity individually suggested that calcimimetics increase the risk for nausea and vomiting and decrease the number of hypotensive episodes and upper respiratory tract infec-
CALCIMIMETICS IN CHRONIC KIDNEY DISEASE
tion, but our meta-analysis of data from all trials that reported these events did not confirm these findings (Table 4). Despite their adoption into clinical practice and approval by the Food and Drug Administration in the United States, the efficacy and tolerability of calcimimetics and their role in the treatment of patients with high-turnover bone disease caused by SHPT has not yet been investigated adequately. Animal studies showed improvement in bone histological characteristics and bone strength with calcimimetic therapy,44 but these findings are yet to be shown in humans, and there are no major clinical data showing an effect on bone histomorphometric characteristics, which is the best determinant of renal bone disease. In addition, there are no adequate clinical data showing an effect of calcimimetics on fracture risk and subsequent morbidity in patients with CKD. Given the recent introduction of calcimimetic therapy, many of these questions may be answered in coming years. It could be suggested that a meta-analysis of therapeutic responses to calcimimetic therapy, particularly longer term outcomes, is premature at this early stage of use. However, cinacalcet HCl is registered by many national drug regulatory agencies (http://www.amgen.com.au/public/ docs/Sensipar_CMI.pdf, pharmacos.eudra.org/ F2/register/register.htm),20 is in widespread clinical use, and information based on metaanalysis of available randomized controlled trials should be valuable to clinicians evaluating whether to use this novel, but expensive, agent. The argument in favor of calcimimetic use hinges on the acceptance of improvements in calcium and phosphorus metabolism and levels of PTH as valid surrogates of such clinically important outcomes as mortality. A surrogate is a measurable outcome, such as a laboratory or imaging test, that is responsive to the effect of an intervention and associated causally with a clinically important outcome. A valid surrogate end point captures the full effect of an intervention, but earlier in the causal chain of events.45-47 Surrogate end points often are used in preference to hard end points in randomized controlled trials because cost and sample size can be decreased and feasibility can be increased substantially. Compared with hard end points, surrogates allow for shorter study duration and either occur more
723
commonly or are continuous measures and therefore more sensitive to differences in treatment. In patients with kidney disease, surrogates commonly are used in trials and include dialysis adequacy, hemoglobin level, left ventricular hypertrophy, and episodes of acute rejection, which have been the basis for the regulatory approval and clinical use of various drugs.48-51 However, not all surrogates are valid proxies of clinically important patient-centered outcomes. For a surrogate to be valid, 2 criteria must be met. First, there must be a strong, independent, and consistent association between the surrogate and the clinically important outcome, which comes from observational studies. For calcium, phosphorus, and PTH levels, this criterion has been met from a number of large-scale cohort and cross-sectional studies.7,8,52,53 Second, and more importantly, for a surrogate to be valid, there also must be evidence that using an intervention changes a surrogate (eg, decrease in PTH levels with a calcimimetic) and results in an expected change in the patient-based outcome distal to the surrogate in the same causal pathway for the disease in question (eg, decrease in deaths with a calcimimetic). This criterion requires a randomized controlled trial that measures both the surrogate and the hard end point. Our study shows that the second criterion has not yet been met for calcimimetics. Critics of the second criterion argue that it is too stringent and will mean that potentially lifesaving interventions will be withheld. Proponents invoke the usual arguments for superiority of randomized controlled trials compared with observational studies (selection bias and unmeasured confounders) to estimate the true effects of interventions. In addition, some interventions that positively affect a surrogate may have a counterintuitive lack of effect on patient-based outcomes in the CKD population, with the atorvastatin in patients with type 2 diabetes mellitus undergoing hemodialysis (4D) study as a recent example.54 Such results suggest that validation of surrogates in disease-specific populations should be mandatory when adopting novel interventions, and trial results based on unvalidated surrogates should be used cautiously. We identified 5 other problems in the trial basis supporting calcimimetic use. First, despite the widespread adoption of CONSORT, a stan-
724
dardized checklist to improve the quality of reporting of randomized controlled trials, the majority of trials of calcimimetics either did not state or did not use (or both) secure methods of allocation, blinded outcomes assessors, or decreased loss to follow-up to a minimum. These design and reporting problems tend to bias results in favor of the intervention.55 Second, to illustrate the importance of prospective registration of trials recently adopted by all major biomedical journals and kidney journals, to ensure that all trials evaluating an intervention are known and linked with publications to avoid publication bias56 and duplication bias,57 we found it very difficult to link publications with studies. We identified 8 unequivocally separate studies with unique data, but also 7 other studies (published in abstract form) that could be duplicates of data already available in full-text published reports. Duplicate reporting is associated with an overestimation of true treatment effects and spurious precision if the trials are incorporated in meta-analysis.57 Prospective registration with a unique identification number for each trial would avoid this. Third, based on the published report, investigators of the largest trial32 combined results of 2 separate, but similar, studies, a method used by the celecoxib long-term arthritis safety study (CLASS) investigators that has been criticized widely.58 When the cinacalcet group of 1 trial is compared with both the placebo group of the same trial (random allocation) and the placebo group of the other trial (nonrandom allocation), outcome differences between the cinacalcet and placebo groups may be caused by differences in study populations or cointerventions, which are unknown and therefore cannot be adjusted for. Having identical trial designs does not prevent these effects. Such trial results would be better reported separately. Data then could be combined by using meta-analysis to provide a summary weighted estimate of the effects shown in the individual trials.32,58 Fourth, CKD is a long-term problem, but except for 1 study, published trials reported treatment outcomes to 6 months only. Finally, reporting of PTH outcomes may have been informative if the proportion of patients achieving a certain end-of-treatment threshold had been reported, and any differences in base-
STRIPPOLI ET AL
line PTH levels may have been better accounted for by reporting change values, rather than endof-treatment levels alone. Outcome reporting was suboptimal, and although our analysis includes all available data reported in the trials, both the benefits and the harms, there were no data on some relevant outcomes, including bone-specific alkaline phosphatase data, PTH target levels achieved, and others. It remains an important, but unanswered, question whether calcimimetic effects on such surrogate end points as biochemical targets and changes in vascular compliance, indices of vascular calcification, and bone mineral density will translate into improved patient-based outcomes. Given reservations relating to current published trials, calcimimetic therapy for patients with SHPT can be considered of potential, but currently unproven, benefit to patient-based outcomes, including cardiovascular mortality, renal osteodystrophy, and fracture. Randomized controlled trials with adequate power and longer treatment duration are required to determine the most appropriate use of this important new class of drugs. ACKNOWLEDGMENT The authors acknowledge the editorial and administrative support of Narelle Willis and Sandra Puckeridge. Ruth Mitchell, Linda Heslop, and Gail Higgins, trial search coordinators of the Cochrane Renal Group, provided search strategies for this review.
REFERENCES 1. Martin KJ, Olgaard K, Coburn JW, et al: Diagnosis, assessment, and treatment of bone turnover abnormalities in renal osteodystrophy. Am J Kidney Dis 43:558-565, 2004 2. Stehman-Breen C: Osteoporosis and chronic kidney disease. Semin Nephrol 24:78-81, 2004 3. Marco MP, Craver L, Betriu A, Belart M, Fibla J, Fernandez E: Higher impact of mineral metabolism on cardiovascular mortality in a European hemodialysis population. Kidney Int Suppl 85:S111-S114, 2003 4. Block GA, Klassen PS, Lazarus JM, Ofsthun N, Lowrie EG, Chertow GM: Mineral metabolism, mortality, and morbidity in maintenance hemodialysis. J Am Soc Nephrol 15:22082210, 2004 5. Ganesh SK, Stack AG, Levin NW, Hulbert-Shearon T, Port FK: Association of elevated serum PO4, Ca ⫻ PO4 product, and parathyroid hormone with cardiac mortality risk in chronic hemodialysis patients. J Am Soc Nephrol 12:2131-2138, 2001 6. Malluche HH, Mawad H, Monier-Faugere MC: The importance of bone health in end-stage renal disease: Out of the frying pan, into the fire? Nephrol Dial Transplant 19:i9i13, 2004
CALCIMIMETICS IN CHRONIC KIDNEY DISEASE
7. Kestenbaum B, Sampson JN, Rudser KD, et al: Serum phosphate levels and mortality risk among people with chronic kidney disease. J Am Soc Nephrol 16:520-528, 2005 8. Stevens LA, Djurdjev O, Cardew S, Cameron EC, Levin A: Calcium, phosphate, and parathyroid hormone levels in combination and as a function of dialysis duration predict mortality: Evidence for the complexity of the association between mineral metabolism and outcomes. J Am Soc Nephrol 15:770-779, 2004 9. Hutchison AJ, Whitehouse RW, Boulton HF, et al: Correlation of bone histology with parathyroid hormone, vitamin D3, and radiology in end-stage renal disease. Kidney Int 44:1071-1077, 1993 10. Qi Q, Moniere-Faugere MC, Geng Z, Malluche HH: Predictive value of serum parathyroid hormone levels for bone turnover in patients on chronic maintenance hemodialysis. Am J Kidney Dis 26:622-631, 1995 11. Wang M, Hercz G, Sherrard DJ, Maloney NA, Segre GV, Pei Y: Relationship between intact 1-84 parathyroid hormone and bone histomorphometric parameters in dialysis patients without aluminum toxicity. Am J Kidney Dis 26:836844, 1995 12. Ziolkowska H, Panczyk-Tomaszewska M, Debinski A, Polowiec Z, Sawicki A, Sieniawska M: Bone biopsy results and serum bone turnover parameters in uremic children. Acta Paediatr 89:666-671, 2000 13. National Kidney Foundation: K/DOQI Clinical Practice Guidelines for Bone Metabolism and Disease in Chronic Kidney Disease. Am J Kidney Dis 42:S1-S201, 2003 (suppl 3) 14. CARI Guidelines. Available at: http://www.kidney. org.au/cari/guidelines.php. Accessed July 1, 2005 15. Young EW, Akiba T, Albert JM, et al: Magnitude and impact of abnormal mineral metabolism in hemodialysis patients in the Dialysis Outcomes and Practice Patterns Study (DOPPS). Am J Kidney Dis 44:S34-S38, 2004 (suppl 3) 16. Albaaj F, Hutchison A: Hyperphosphatemia in renal failure: Causes, consequences and current management. Drugs 63:577-596, 2003 17. Courant O, Letessier E, Moutel MG, Hamy A, Paineau J, Visset J: [Surgical treatment of secondary hyperparathyroidism in chronic kidney failure. Results of total parathyroidectomy with parathyroid autotransplantation]. J Chir 130:327334, 1993 18. Goodman WG, Frazao JM, Goodkin DA, Turner SA, Liu W, Coburn JW: A calcimimetic agent lowers plasma parathyroid hormone levels in patients with secondary hyperparathyroidism. Kidney Int 58:436-445, 2000 19. Ott SM: Calcimimetics—New drugs with the potential to control hyperparathyroidism. J Clin Endocrinol Metab 83:1080-1082, 1998 20. Amgen Inc: Sensipar (Cinacalcet) HCl Tablets. Available at: http://www.fda.gov/cder/foi/label/2004/21688_ Sensipar_lbl.pdf. Accessed July 1, 2005 21. Higgins JP, Thompson SG, Deeks JJ, Altman DG: Measuring inconsistency in meta-analyses. BMJ 327:557560, 2003 22. Coburn JW, Charytan C, Chonchol M, et al: Cincalcet HCl is an effective treatment for secondary hyperparathyroidism (HPT) in patients with chronic kidney disease (CKD) not yet receiving dialysis. J Am Soc Nephrol 14:460A, 2003 (abstr)
725
23. Goodman WG, Fadda GZ, Finkelstein FO, et al: Cincalcet HCl is an effective primary therapy for the management of secondary hyperparathyroidism. J Am Soc Nephrol 14:460A, 2003 (abstr) 24. Martin KJ, Sherrard DJ, Nassar G, et al: Control of bio-intact parathyroid hormone (PTH) levels with the calcimimetic cinacalcet HCl in hemodialysis patients with secondary hyperparathyroidism (HPT). J Am Soc Nephrol 14: 462A, 2003 (abstr) 25. Quarles LD, Zeig S, Spiegel DM, et al: Cincalcet HCl controls secondary hyperparathyroidism (HPT) in dialysis patients regardless of disease severity. J Am Soc Nephrol 14:463A, 2003 (abstr) 26. Sterrett JR, Strom J, Stummvoll HK, et al: Long-term treatment of secondary hyperparathyroidism (HPT) with cinacalcet HCl (Sensipar) in patients receiving dialysis. J Am Soc Nephrol 15:275A, 2004 (abstr) 27. Cunningham J, Holzer H, Reichel H, et al: Sustained, long-term reductions in parathyroid hormone (iPTH) after 3 years of treatment with the calcimimetic cinacalcet HCl in patients with secondary hyperparathyroidism (SHPT). Nephrol Dial Transplant 18:S150A-S151A, 2003 (suppl 4; abstr) 28. Cunningham J, Chertow GM, Goodman WG, et al: The effect of cincalcinet HCl on parathyroidectomy, fracture, hospitalization, and mortality in dialysis subjects with secondary hyperparathyroidism (HPT). Proceedings of the European Renal Association-European Dialysis and Transplant Association XLI Congress, May 15-18, 2004, Lisbon, Portugal, p 219 (abstr) 29. Fournier A, Pontoiero G, Walker R, et al: Cinacalcet reduces parathyroid hormone (PTH) and calcium-phosphorus product (Ca ⫻ P) regardless of concurrent changes in vitamin D administration. Proceedings of the European Renal Association-European Dialysis and Transplant Association XLI Congress, May 15-18, 2004, Lisbon, Portugal, p. 319 (abstr) 30. De Francisco A, Suranyi MG, Cunningham J, et al: Greater achievement of NKF/DOQI bone metabolism and disease targets with oral cinacalcet HCl in an EU/Australian phase 3 study in hemodialysis subjects. Proceedings of the European Renal Association-European Dialysis and Transplant Association XLI Congress, May 15-18, 2004, Lisbon, Portugal, p 319 (abstr) 31. Frazao JM, Nicolini M, Torregrosa JV, et al: Cinacalcet HCl effectively reduces intact parathyroid hormone (iPTH) and Ca ⫻ P irrespective of the severity of secondary hyperparathyroidism (HPT). Proceedings of the European Renal Association-European Dialysis and Transplant Association XLI Congress, May 15-18, 2004, Lisbon, Portugal, p 219 (abstr) 32. Block GA, Martin KJ, de Francisco AL, et al: Cinacalcet for secondary hyperparathyroidism in patients receiving hemodialysis. N Engl J Med 350:1516-1525, 2004 33. Goodman WG, Hladik GA, Turner SA, et al: The calcimimetic agent AMG 073 lowers plasma parathyroid hormone levels in hemodialysis patients with secondary hyperparathyroidism. J Am Soc Nephrol 13:1017-1024, 2002 34. Harris RZ, Padhi D, Marbury TC, Noveck RJ, Salfi M, Sullivan JT: Pharmacokinetics, pharmacodynamics, and safety of cinacalcet hydrochloride in hemodialysis patients at
726
doses up to 200 mg once daily. Am J Kidney Dis 44:1070-1076, 2004 35. Lindberg JS, Moe SM, Goodman WG, et al: The calcimimetic AMG 073 reduces parathyroid hormone and calcium x phosphorus in secondary hyperparathyroidism. Kidney Int 63:248-254, 2003 36. Lindberg JS, Culleton B, Wong G, et al: Cinacalcet HCl, an oral calcimimetic agent for the treatment of secondary hyperparathyroidism in hemodialysis and peritoneal dialysis: A randomized, double-blind, multicenter study. J Am Soc Nephrol 16:800-807, 2005 37. Malluche HH, Monier-Faugere MC, Wang G, et al: Cinacalcet HCl reduces bone turnover and bone marrow fibrosis in hemodialysis patients with secondary hyperparathyroidism (HPT). Proceedings of the European Renal Association-European Dialysis and Transplant Association XLI Congress, May 15-18, 2004, Lisbon, Portugal, p 218 (abstr) 38. Quarles LD, Sherrard DJ, Adler S, et al: The calcimimetic AMG 073 as a potential treatment for secondary hyperparathyroidism of end-stage renal disease. J Am Soc Nephrol 14:575-583, 2003 39. Chertow GM, Corpier C, Zeig S, et al: Preliminary results from TARGET: Treatment strategies to achieve recommended K/DOQI™ goals in ESRD patients on cinacalcet. J Am Soc Nephrol 15:863A, 2004 (abstr) 40. Reed J, Keightley G, Blumenthal S, et al: The CONTROL Study: Enhanced achievement of NKF-K/ DOQI™ bone metabolism and disease targets using cinalcalcet HCl (Sensipar™). J Am Soc Nephrol 15:280A, 2004 (abstr) 41. Urena P, Frazao JM: Calcimimetic agents: Review and perspectives. Kidney Int Suppl 85:S91-S96, 2003 42. Moher D, Schulz KF, Altman D, for the CONSORT Group (Consolidated Standards of Reporting Trials): The CONSORT statement: Revised recommendations for improving the quality of reports of parallel-group randomized trials. JAMA 285:1987-1991, 2001 43. ANZDATA: 27th Annual Report. Available at: http:// www.anzdata.org.au/anzdata/AnzdataReport/download.htm. Accessed August 8, 2005 44. Sherrard DJ: Calcimimetics in action. Kidney Int 53:510-511, 1998 45. Bucher HC, Guyatt GH, Cook DJ, et al: Users’ guides to the medical literature: XIX. Applying clinical trial results. A. How to use an article measuring the effect of an intervention on surrogate endpoints. Evidence-Based Working Group. JAMA 282:771-778, 1999
STRIPPOLI ET AL
46. Psaty BM, Weiss NS, Furberg CD, et al: Surrogate endpoints, health outcomes, and the drug-approval process for the treatment of risk factors for cardiovascular disease. JAMA 282:786-790, 1999 47. Temple R: Are surrogate markers adequate to assess cardiovascular disease drugs? JAMA 282:790-795, 1999 48. Churchill DM, Thorpe KE, Vonesh EF, Keshaviah PR: Lower probability of patient survival with continuous peritoneal dialysis in the United States compared with Canada. Canada-USA (CANUSA) Peritoneal Dialysis Study Group. J Am Soc Nephrol 8:965-971, 1997 49. Besarab A, Bolton WK, Browne JK, et al: The effects of normal, as compared with low, hematocrit values in patients with cardiac disease who are receiving hemodialysis and epoetin. N Engl J Med 339:584-590, 1998 50. McMahon LP, Roger SD, Levin A, for the SLIMHEART Investigators Group: Development, prevention and potential reversal of left ventricular hypertrophy in chronic kidney disease. J Am Soc Nephrol 15:1640-1647, 2004 51. Borrows R, Loucaidou M, van Tromp J, et al: Steroid sparing with tacrolimus and mycophenolate mofetil in renal transplantation. Am J Transplant 4:1845-1851, 2004 52. Avram MM, Sreedhara R, Avram DK, Muchnick RA, Fein P: Enrollment parathyroid hormone level is a new marker of survival in hemodialysis and peritoneal dialysis therapy for uremia. Am J Kidney Dis 28:924-930, 1996 53. Ganesh SK, Stack AG, Levin NW, Hulbert-Shearon T, Port FK: Association of elevated serum PO4, Ca ⫻ PO4 product, and parathyroid hormone with cardiac mortality risk in chronic hemodialysis patients. J Am Soc Nephrol 12:2131-2138, 2001 54. Wanner C, Krane V, Marz W, et al, for the German Diabetes and Dialysis Study Investigators: Atorvastatin in patients with type 2 diabetes mellitus undergoing haemodialysis. N Engl J Med 353:238-248, 2005 55. Schulz KF, Chalmers I, Hayes RJ, Altman DG: Empirical evidence of bias. Dimensions of methodological quality associated with estimates of treatment effects in controlled trials. JAMA 273:408-412, 1995 56. Simes RJ: Publication bias: The case for an international registry of clinical trials. J Clin Oncol 4:1529-1541, 1986 57. Egger M, Smith GD: Meta-analysis bias in location and selection of studies. BMJ 316:61-66, 1998 58. Juni P, Sterchi R, Dieppe P: Systematic review of celecoxib for osteoarthritis and rheumatoid arthritis. BMJ 326:334, 2003
VOL 47, NO 5, MAY 2006
American Journal of Kidney Diseases
REVIEWS
Meta-Analysis of Biochemical and Patient-Level Effects of Calcimimetic Therapy Giovanni F.M. Strippoli, MD, Suetonia Palmer, MBChB, Allison Tong, BMedSc, Grahame Elder, MBBS, PhD, Piergiorgio Messa, MD, and Jonathan C. Craig, MBChB, PhD ● Background: Many randomized trials have now evaluated the effects of calcimimetics in patients with chronic kidney disease and secondary hyperparathyroidism (SHPT) on standard therapy with vitamin D and/or phosphate binders. We conducted a meta-analysis to evaluate outcomes of therapy with these novel agents. Methods: MEDLINE, EMBASE, the Cochrane Controlled Trials Register, and conference proceedings were searched for randomized controlled trials evaluating any calcimimetic against placebo or another agent in predialysis or dialysis patients with chronic kidney disease. Data were extracted for all relevant patient-centered and surrogate outcomes. Analysis was by means of a random-effects model, and results are expressed as relative risk or weighted mean difference (WMD) with 95% confidence intervals (CIs). Results: Eight trials (1,429 patients) were identified that compared a calcimimetic agent plus standard therapy with placebo plus standard therapy. End-of-treatment values for parathyroid hormone (4 trials; 1,278 patients; WMD, ⴚ290.49 pg/mL; 95% CI, ⴚ359.91 to ⴚ221.07), serum calcium (3 trials; 1,201 patients; WMD, ⴚ0.85 mg/dL; 95% CI, ⴚ1.14 to ⴚ0.56), serum phosphorus (3 trials; 1,195 patients; WMD, ⴚ0.29 mg/dL; 95% CI, ⴚ0.50 to ⴚ0.08), and calcium ⴛ phosphorus product (3 trials; 1,194 patients; WMD, ⴚ7.90 mg2/dL2; 95% CI, ⴚ10.25 to ⴚ5.54) were significantly lower with calcimimetic therapy compared with placebo. No significant effects on patient-based end points were shown. Conclusion: Calcimimetic treatment of patients with SHPT leads to significant improvements in biochemical parameters that observational studies have associated with increased mortality, cardiovascular risk, and osteitis fibrosa, but patient-based benefits have not yet been shown. For patients with SHPT, the benefits of calcimimetics over standard therapy remain uncertain until additional randomized trials become available. Am J Kidney Dis 47:715-726. © 2006 by the National Kidney Foundation, Inc. INDEX WORDS : Calcimimetic; secondary hyperparathyroidism; renal bone disease.
A
BNORMALITIES OF CALCIUM and phosphorus metabolism are present in all individuals with chronic kidney disease (CKD). These biochemical changes cause many bone and metabolic disorders, including renal osteodystrophy,
which is characterized by abnormalities of bone turnover (ranging from high-turnover osteitis fibrosa to adynamic bone disease); mineralization defects (osteomalacia); and architectural change. Renal osteodystrophy is associated not only with
From the NHMRC Centre for Clinical Research Excellence in Renal Medicine, Cochrane Renal Group, and School of Public Health, University of Sydney; Centre for Transplant and Renal Research, Westmead Millennium Institute, Sydney, Australia; Department of Nephrology, Christchurch Hospital, Christchurch, NZ; and Department of Nephrology and Urology, Ospedale Maggiore-Policlinico di Milano, Milan, Italy. Received October 19, 2005; accepted in revised form January 26, 2006.
Originally published online as doi:10.1053/j.ajkd.2006.01.015 on March 17, 2006. Support: None. Potential conflicts of interest: None. Address reprint requests to Giovanni F.M. Strippoli, MD, NHMRC Centre for Clinical Research Excellence in Renal Medicine, Cochrane Renal Group, University of Sydney, Australia. E-mail: [email protected] © 2006 by the National Kidney Foundation, Inc. 0272-6386/06/4705-0001$32.00/0 doi:10.1053/j.ajkd.2006.01.015
American Journal of Kidney Diseases, Vol 47, No 5 (May), 2006: pp 715-726
715
716
an increased incidence of fracture, bone and muscular pain, and abnormalities of bone and joint morphological characteristics, but also vascular and soft-tissue calcification. In addition to causing decreased quality of life, these complications and the associated abnormalities of elevated phosphorus, calcium, calcium ⫻ phosphorus product, and parathyroid hormone (PTH) levels have been associated with increased mortality.1-6 Standard management of patients with CKD, particularly those on dialysis therapy, includes treatment to control calcium, phosphorus, and PTH levels to prevent bone and soft-tissue complications. Based on a number of association studies,3-5,7,8 including studies of bone histomorphometric characteristics,9-12 optimal ranges for serum phosphorus, calcium, calcium ⫻ phosphorus product, and PTH levels have been suggested.13,14 However, success in achieving these targets has been limited.15 Specific management of patients with secondary hyperparathyroidism (SHPT) with CKD stages 3 and 4 may be accomplished by restriction of dietary phosphorus, calcium supplementation, and/or the use of vitamin D. When patients have started dialysis therapy, standard therapy for SHPT generally includes calcitriol or vitamin D analogues, calcium-based and/or newer phosphate-binding agents, and parathyroidectomy.16,17 Recently, use of a novel class of drugs, the calcimimetics, was proposed as a strategy to decrease PTH secretion and possibly decrease parathyroid cell proliferation, while decreasing serum calcium, phosphorus, and calcium ⫻ phosphorus product levels.18,19 Results of randomized trials testing the efficacy and safety of calcimimetics in patients undergoing dialysis are becoming available. With the aim of preventing complications associated with SHPT, cinacalcet hydrochloride (HCl) has been incorporated into many treatment algorithms. However, several aspects of calcimimetic therapy require further evaluation. In the dialysis population, for which these drugs are approved, the most important question is the degree to which calcimimetics impact on such clinically relevant end points as parathyroidectomy rates, fracture, renal osteodystrophy, and cardiovascular disease and death, as well as such surrogate markers for these conditions as abnormalities of
STRIPPOLI ET AL
serum calcium and phosphorus levels. Other important questions include the optimal time for the start of calcimimetic therapy, the influence of calcimimetics on standard treatment regimens, and the effectiveness of calcimimetics in patients with different stages of CKD and after transplantation. This systematic review was performed at an early phase of calcimimetic use, but 1 year after licensing by the Food and Drug Administration,20 to assess available efficacy and safety data and highlight research questions that need additional investigation. METHODS
Inclusion Criteria We included all randomized controlled trials of any calcimimetic agent, cinacalcet HCl (AMG-073, Sensipar; Amgen Inc, Thousand Oaks, CA), NPS R-467, or NPS R568 administered to patients with CKD for the treatment of SHPT.
Search Strategy Electronic searches were performed in MEDLINE (1966 to March 2005), EMBASE (1980 to March 2005), and the Cochrane Renal Group Specialized Register by using optimally sensitive search strategies for the identification of randomized trials developed by the Cochrane Collaboration (full details available upon request). Medical subject headings and text words used were relevant to the intervention (all calcimimetic agents) and the condition of interest (SHPT, calcium, phosphorus, calcium ⫻ phosphorus product, levels of PTH and biochemical targets, bone disease, renal osteodystrophy, and cardiovascular disease). Results of these searches were analyzed in title and abstract form by 2 authors according to the inclusion criteria in consultation with a third author. Reference lists from the identified articles also were searched, and information about unpublished or ongoing trials was sought from experts in the field and pharmaceutical companies. Trials were considered without language restriction.
Data Extraction and Trial Quality Assessment Each trial was assessed by 2 independent authors. From all included trials, data were extracted for study sample characteristics; type of agent, dose, and route of administration; and trial methods and outcomes. The following outcome measures were considered: end-of-treatment values for intact PTH (iPTH) and/or PTH(1-84), 30% or greater decrease in PTH levels from baseline, serum calcium level, hypocalcemia, serum phosphorus and calcium ⫻ phosphorus product levels, cardiovascular events, cardiovascular and all-cause mortality, fractures at any site, bone histomorphometry, bone mineral density, bone pain, muscle pain or weakness, paresthesias, nausea, vomiting, dyspnea, hospitalization, and quality of life. Methods and quality of included trials were assessed by using standard criteria (allocation concealment; blinding of participants, investigators, and
CALCIMIMETICS IN CHRONIC KIDNEY DISEASE
outcome assessors; analysis by intention to treat, and completeness of follow-up). Discrepancies in data extraction were resolved by discussion among the authors, and when data were missing or incomplete, the investigators of the trial were contacted for clarification.
Statistical Analysis The estimate of effect of an experimental versus a control intervention on categorical outcomes (eg, fracture rate and all-cause mortality, including sudden death) was analyzed by using the relative risk (RR) measure and its 95% confidence interval (CI) for each trial. For continuous variables, the weighted mean difference (WMD) and its 95% CI were calculated by using end-of-treatment values for variables in the experimental and control groups. Heterogeneity of treatment effects between studies was tested formally by using the Q (heterogeneity chi-square) and I2 statistics.21 When appropriate, summary estimators of treatment effects were calculated by using a random-effects model and expressed as RR or WMD with 95% CIs. Forest plots were used for graphical representation of data. In this type of graphic, vertical lines positioned at 1 for RR and 0 for WMD are consistent with no difference in efficacy between the experimental and control treatments (null hypothesis). Trials shown in the area on the left show a reduction in risk or lower WMD with the experimental intervention (eg, favorable effect on the adverse outcome), whereas those on the right show an increase in risk or greater WMD with the experimental intervention. The numeric scale at the bottom indicates the value of the RR or WMD of the treatment group compared with the control group. A solid square and horizontal line graphically represent the RR or WMD with 95% CIs. The area of the black square indicates how much the trial contributed to the analysis in terms of sample size and number of events recorded (weight). Horizontal lines indicate 95% CIs, with large informative trials having smaller CIs. The diamond-shaped symbol is the summary estimate of effect expressed as either an RR or WMD with 95% CIs, which is a weighted average of the summative treatment effect across all trials.
RESULTS
Our search for randomized trials of calcimimetic interventions to treat patients with SHPT identified 186 articles. Of these, 148 articles were excluded after title and abstract review because they clearly were ineligible (nonrandomized studies, randomized trials of interventions not relevant to SHPT, noncalcimimetic interventions, duplicate articles of the same trial, or review articles; Fig 1). Of the remaining 38 potentially eligible studies (either full-text or abstract publications), 30 studies were excluded because we could not confirm from the full-text analysis or from contacting the investigators that they were randomized trials or not a duplicate publication. Two attempts were made to contact all investigators of the trials to clarify study
717
designs and request supplemental data, but we were not able to obtain some of the data or ascertain whether some reports22-31 (presented in abstract form at the American Society of Nephrology and the European Renal Association– European Dialysis and Transplantation Association meetings of 2003 and 2004) were subsets of other publications that subsequently appeared as full-text articles in scientific journals or were unique unpublished trials. These studies therefore could not be included. Eight trials (8 publications) enrolling 1,429 patients finally were included.18,32-38 Trial Characteristics Characteristics of the populations and interventions of all studies included in this systematic review are listed in Table 1.18,32-38 There also were 2 ongoing trials.39,40 Characteristics of the 11 potentially eligible studies for which we were unable to ascertain eligibility are listed in Table 2. The 8 trials included compared cinacalcet HCl (845 patients) versus placebo (584 patients). Three of these trials reported cinacalcet HCl as AMG073. One trial reported on the first-generation calcimimetic R-568. This drug has been withdrawn from clinical use because of poor bioavailability, variable serum concentrations, and potential drug interactions caused by cytochrome P-450 activity.18,41 Because there were no differences in results of analyses performed with and without inclusion of this trial, combined results are reported. In addition to these randomized interventions, most patients were administered an active vitamin D for suppression of PTH and phosphate binders for management of hyperphosphatemia as cointerventions in all trials in a nonrandomized fashion. There were no significant differences in proportions of patients prescribed calcitriol, vitamin D analogues, and phosphate binders as cointerventions between the calcimimetic and placebo groups of the trials. Entry to some trials was restricted when patients had severe SHPT (eg, iPTH ⱖ 800 pg/mL [ⱖ800 ng/L]), whereas others stratified patients according to severity of hyperparathyroidism. Mean age of patients enrolled in the trials ranged from 47 to 55 years. All patients had SHPT. On average, a greater proportion of males were enrolled in the trials (388
718
STRIPPOLI ET AL
Fig 1. Flow chart indicating the number of citations retrieved by individual searches and the final number of included trials: reasons for exclusion are provided, as well as the number of trials providing outcomes for analysis and number of patients they enrolled.
males compared with 220 females in the 6 trials that reported sex distribution). Follow-up of studies ranged from 3 weeks to 26 weeks. All studies were supported by Amgen Inc, Thousand Oaks, CA, which holds the cinacalcet HCl patent. Of note, the largest report published32 was based on pooled results of 2 separate trials. Method Quality of Included Studies By current standards, the method quality of all except 2 trials was suboptimal (Table 3), although all trials were published after release of the Consolidated Standards for Reporting
Trials (CONSORT) Statement (a checklist designed to improve quality of trial reporting).42 Allocation concealment was unclear in 6 of 8 trials (75%). All studies were reported as double blinded (to indicate that participants and investigators were blinded to the intervention), but no study mentioned the blinding of outcome assessors. Intention-to-treat analysis (ie, analysis of all patients according to randomized allocations, rather than actual intervention) was performed in 3 of 8 trials (38%). The proportion of patients lost to follow-up ranged from 0% to 56%. The proportion of patients who stopped taking the medication was un-
Interventions
Reference Block et al,*32 2004
No. of Patients (treatment/ control) 741 (371/370)
Inclusion Criteria
Exclusion Criteria
Hemodialysis ⱖ 3 mo iPTH ⱖ 300 pg/mL Serum calcium ⱖ 8.4 mg/dL
Infection, malignancy, diseases causing hypercalcemia, tricyclic antidepressant use Malignancy, diseases causing hypercalcemia, myocardial infarction previous 6 mo, tricyclic antidepressant use Malignancy, diseases causing hypercalcemia, myocardial infarction previous 6 mo, tricyclic antidepressant use, transaminase or bilirubin ⬎ 2 ⫻ upper limit of normal Not available
Goodman et al,18 2000
21 (16/5)
Hemodialysis ⱖ 3 mo iPTH, 300-1,200 pg/mL Serum calcium ⬎ 9.0 mg/dL
Goodman et al,33 2002
52 (40/12)
Hemodialysis ⱖ 3 mo iPTH, 250-1,500 pg/mL Serum calcium ⬎ 9.0 mg/dL
Harris et al,34 2004
23 (17/6)
Hemodialysis Serum calcium ⱖ 8.4 mg/dL
Lindberg et al,35 2003
78 (39/39)
Lindberg et al,36 2005
395 (294/101)
Malluche et al,37 2004 (ERA abstract)
48 (32/16)
Hemodialysis ⱖ 3 months iPTH ⱖ 300 pg/mL Serum calcium, 8.8-11.0 mg/dL Calcium ⫻ phosphorus ⬍ 70 mg2/dL2 Hemodialysis and peritoneal dialysis (continuous ambulatory PD, automated PD) ⬎ 1 mo iPTH ⱖ 300 pg/mL Serum calcium ⱖ 8.4 mg/dL Serum phosphorus ⱖ 2.5 mg/dL Calcium ⫻ phosphorus ⱖ 70 mg2/dL2 Stable vitamin D ⬎ 30 d iPTH ⱖ 300 pg/mL
Quarles et al,38 2003
71 (36/35)
Hemodialysis at least 3 mo iPTH ⱖ 300 pg/mL Serum calcium, 8.8-11.0 mg/dL Calcium ⫻ phosporus ⬍ 70 mg2/dL2 Stable vitamin D ⬎ 21 d
Treatment Group (target PTH)
Cointerventions (type, %)
Control
Treatment Group
Control
Duration (wk)
Cinacalcet, 30-180 mg/d (iPTH ⬍ 250 pg/mL)
Placebo
Vitamin D, 66%; phosphate binder, 92%
Vitamin D, 67%; phosphate binder, 93%
26 Titration, 12; maintenance, 14
R-568, 100 mg/d
Placebo
Vitamin D, 38%; phosphate binder, 31%
Vitamin D, 50%; phosphate binder, 50%
3
AMG073,* 5-100 mg/d
Placebo
Not available
Not available
2
Cinacalcet, 25 mg-300 mg/d
Placebo
AMG073,* 10-50 mg/d (iPTH decrease ⱖ 30%)
Placebo
Vitamin D, 60%; phosphate binder, 80% Vitamin D, 67%; phosphate binder, 87%
1
Infection, malignancy, diseases causing hypercalcemia, hepatic transaminase or bilirubin ⬎ 2 ⫻ upper limit of normal
Vitamin D, 71%; phosphate binder, 100% Vitamin D, 62%; phosphate binder, 87%
Parathyroidectomy, previous myocardial infarction (3 mo)
Cinacalcet, 30-180 mg/d (iPTH ⬍ 250 pg/mL)
Placebo
Not available
Not available
26 Titration, 16; maintenance, 10
Not available
Cinacalcet, 30-180 mg/d (iPTH ⱕ 250 pg/mL)
Placebo
Not available
Not available
Infection, malignancy, diseases causing hypercalcemia, hepatic transaminase or bilirubin ⬎ 2 ⫻ upper limit of normal
AMG073,* 25-100 mg/d (iPTH decrease ⱖ 30%)
Placebo
Vitamin D, 61%; phosphate binder, 100%
Vitamin D, 69%; phosphate binder, 94%
52 Titration, 24; maintenance, 28 18 Titration, 12; maintenance, 6
18 Titration, 12; maintenance, 6
719
NOTE. To convert PTH in pg/mL to ng/L, multiply by 1; calcium in mg/dL to mmol/L, multiply by 0.2495; phosphorus in mg/dL to mmol/L, multiply by 0.3229. *AMG073 and cinacalcet are the same compound.
CALCIMIMETICS IN CHRONIC KIDNEY DISEASE
Table 1. Characteristics of Trials of Calcimimetic Agents for SHPT in Patients With CKD
720
STRIPPOLI ET AL Table 2. Characteristics of Ongoing Trials or Trials With Insufficient Data to Assess Eligibility for Inclusion in This Systematic Review Interventions
Study and Reference
No. of Patients (treatment/control)
Inclusion Criteria
Treatment Group (target PTH)
Control
Study Duration (wk)
Major Reason for Inability to Include in This Review
Studies identified as “ongoing trials,” but reported in only abstract form with insufficient data for inclusion TARGET39 424 (NA/NA) Mild/moderate SHPT Cinacalcet (dose Placebo 8-14 Data still unpublished iPTH, 300-800 pg/mL NA) Preliminary data NA CONTROL40 76 (NA/NA) KDOQI iPTH target Cinacalcet, 30-180 Placebo 8 Data still unpublished (150-300 pg/mL) mg/d (iPTH, Preliminary data NA Calcium ⫻ phosphorus 150-300 pg/mL) target outside KDOQI range Studies in abstract form with insufficient data available to assess whether they were randomized trials and could be included in this analysis 54 (19/21) Nondialysis Cinacalcet, 30-180 Placebo 18 Data NA to assess if a randomized Coburn et al,22 2003 iPTH ⱖ 130 pg/mL mg/d (iPTH study decrease ⱖ 30%) 121 (NA/NA) Hemodialysis and Cinacalcet, 30-180 Placebo 24 Analysis of 3 trials, unconfirmed if Goodman et al,23 2003 peritoneal dialysis mg/d randomized data ⬍ 12 mo iPTH ⱖ 300 pg/mL 410 (205/205) Hemodialysis Cinacalcet (dose Placebo 26 Possible (unconfirmed data) Martin et al,24 2003 iPTH ⱖ 300 pg/mL NA) (iPTH duplicate of a previous ⱕ 250 pg/mL) publication 1,136 (665/471) Hemodialysis and Cinacalcet (dose Placebo 24 3 trials, possible (unconfirmed data) Quarles et al,25 2003 peritoneal dialysis NA) duplicate of a previous iPTH ⱖ 300 pg/mL publication 26 210 (99/111) Hemodialysis Cinacalcet (dose Placebo 48 2 trials, possible (unconfirmed data) Sterrett et al, 2004 iPTH ⱖ 300 pg/mL NA) duplicate of a previous publication 297 (182/115) Hemodialysis Cinacalcet ⬎ 180 Placebo 208 Analysis of 5 trials, unconfirmed if Cunningham et al,27 2003 iPTH ⱖ 300 pg/mL mg/d all randomized data Cunningham et al,28 2003 1,184 (697/487) Hemodialysis Cinacalcet (dose Placebo 72 5 trials, possible (unconfirmed data) iPTH ⱖ 300 pg/mL NA) (iPTH, 250 duplicate of a previous pg/mL) publication 29 NA Renal replacement Cinacalcet, 30-180 Placebo Fournier et al, 2004 NA 3 trials, possible (unconfirmed data) therapy mg/d (iPTH duplicate of a previous iPTH ⱖ 300 pg/mL ⱕ 250 pg/mL) publication 311 (139/150) Hemodialysis Cinacalcet, 30-180 Placebo NA Possible (unconfirmed data) de Francisco et al,30 2004 iPTH ⱖ 300 pg/mL mg/d (iPTH duplicate of a previous ⱕ 250 pg/mL) publication 31 1,136 (556/471) Renal replacement Cinacalcet ⬎ 180 Frazao et al, 2004 Placebo NA Analysis of 3 trials, unconfirmed if therapy mg/d (iPTH randomized data, possible iPTH ⱖ 300 pg/mL ⱕ 250 pg/mL) (unconfirmed data) duplicate of a previous publication
NOTE. To convert PTH in pg/mL to ng/L, multiply by 1. Abbreviations: TARGET, treatment strategies to achieve recommended KDOQI goals in ESRD patients on cinacalcet; CONTROL, enhanced achievement of NKF-KDOQI bone metabolism and disease targets using cinacalcet HCl (Sensipar); NA, not available; KDOQI, Kidney Disease Outcomes Quality Initiative.
clear from reports of most trials and either was not reported or reported to be minimal. Trial Results: Patient-Level Outcomes All-cause mortality was not significantly different with calcimimetics compared with placebo (5 trials conducted during 0.5 to 6.0 months; 1,285 patients; RR, 0.75; 95% CI, 0.30 to 1.88), with no significant heterogeneity in this analysis. The effect of calcimimetics on other patientlevel outcomes was extracted from individual trials, and summary estimates of effect are listed in Table 4. Fractures and other clinically impor-
tant measures of renal bone disease were not reported by any trial except for 2 abstracts; in one of these trials, data reported were insufficient to ascertain eligibility,27 and in the other, data were insufficient to estimate observed treatment effects.37 There was no report that rates of hospital admission or study withdrawal differed between patients treated with calcimimetic or placebo. Treatment with calcimimetic or placebo did not result in significant differences in the use of phosphate binders, calcitriol, or vitamin D analogues (reported in 5 trials). Risk for nausea and vomiting increased with calcimimetic
CALCIMIMETICS IN CHRONIC KIDNEY DISEASE
721
Table 3. Assessment of Method Quality of Randomized Controlled Trials of Calcimimetic Agents Included in This Analysis Blinding
Reference
Block et al,32 2004 Goodman et al,18 2000 Goodman et al,33 2002 Harris et al,34 2004 Lindberg et al,35 2003 Lindberg et al,36 2005 Malluche et al,37 2004 Quarles et al,38 2003
Allocation Concealment*
Not stated Not stated Not stated Not stated Not stated Adequate Not stated Adequate
Participants
Investigators
Outcome Assessors
Yes Yes Yes Yes Yes Yes Yes Yes
Yes Yes Yes Yes Yes Yes Yes Yes
No No No No No No No No
Intentionto-Treat Analysis
Lost to FollowUp (%)†
No No Yes No No Yes NA Yes
202/741 (27) 7/20 (56) 0/52 (0) 10/23 (40) 11/78 (14) 101/395 (26) NA 6/71 (8)
Abbreviation: NA, data not available in abstract form or obtainable when contacting investigators. *Defined as adequate when sequentially labeled, sealed, opaque envelopes or a central or pharmacy randomization were used; inadequate when a pseudorandomization method including alternation, date of birth, or chart number was used; not stated if no information was available. †Including patients described as lost to follow-up or those who were discontinued and who could have had outcomes measured.
therapy, but hypotension was less common with calcimimetics. These results should be interpreted with caution because few trials reported these outcomes. Trial Results: Biochemical Outcomes The end-of-treatment value for PTH was significantly less with calcimimetics compared with
placebo (4 trials conducted during 0.5 to 7.0 months; 1,278 patients; WMD, ⫺290.49 pg/mL; 95% CI, ⫺359.91 to ⫺221.07; Fig 2A). Similarly, end-of-treatment values for serum calcium (3 trials; 1,201 patients; WMD, ⫺0.85 mg/dL; 95% CI, ⫺1.14 to ⫺0.56; Fig 2B), serum phosphorus (1,195 patients; WMD, ⫺0.29 mg/dL; 95% CI, ⫺0.50 to ⫺0.08; Fig 2C), and calcium ⫻
Table 4. Summary Estimates of Effect of Treatment With Calcimimetics Compared With Placebo in Populations Enrolled in Trials Included in This Systematic Review Outcome
No. of Trials
No. of Patients
RR
95% CI
I2 (%)
P (heterogeneity)
All-cause mortality ⱖ 30% Decrease in mean PTH level Fractures Hypocalcemia* Nausea* Vomiting Dyspnea Muscle weakness or paresthesia Hypotension Upper respiratory tract infection Parathyroidectomy Headache Asthenia Abdominal pain Diarrhea Mixed uremic osteodystrophy Bone histomorphometry
5 4 — 4 5 2 1 1 2 2 1 1 1 1 1 1 —
1,285 1,284 — 868 1,263 1,136 77 20 1,136 1,136 395 395 396 395 395 32 —
0.75 4.49 — 2.89 1.40 1.89 1.54 3.24 0.53 0.86 0.07 0.86 4.12 0.67 1.28 0.34 —
0.30-1.88 3.04-6.64 — 0.71-11.73 1.00-1.95 1.47-2.43 0.44-5.34 0.21-49.01 0.36-0.79 0.34-2.18 0.00-1.43 0.54-1.37 0.99-17.14 0.40-1.13 0.82-2.02 0.07-1.60 —
0.0 47.3 — 0.0 0.0 0.0 NA NA 0.0 85.0 NA NA NA NA NA NA —
0.63 0.13 — 0.78 0.91 0.91 NA NA 0.66 0.01 NA NA NA NA NA NA —
Abbreviation: NA, not applicable because events occurred in 1 trial only. *For outcomes that may occur more than once, number of patients developing 1 or more episodes of that outcome was considered.
722
STRIPPOLI ET AL
Fig 2. Effect of calcimimetic agents compared with placebo on values for (A) PTH, (B) calcium, (C) phosphorus, and (D) calcium ⴛ phosphorus product. To convert PTH in pg/mL to ng/L, multiply by 1; calcium in mg/dL to mmol/L, multiply by 0.2495; phosphorus in mg/dL to mmol/L, multiply by 0.3229.
phosphorus product (3 trials; 1,204 patients; WMD, ⫺7.90 mg2/dL2; 95% CI, ⫺10.25 to ⫺5.54; Fig 2D) were significantly less with calcimimetics than placebo. DISCUSSION
Trials of calcimimetic compounds in patients with CKD are restricted almost exclusively to hemodialysis patients with SHPT, for whom calcimimetics have been added to standard therapy with outcomes measured up to 6 months. In this population, calcimimetic agents compared with placebo were effective in decreasing levels of PTH, serum calcium, phosphorus, and calcium ⫻ phosphorus product. All-cause mortality was reported in 5 studies and not reported in 3 studies. Analysis of these studies suggests that cinacalcet HCl may be associated with a 25% RR for decrease in all-cause mortality, but the estimate is very imprecise, with wide CIs ranging from a 70% decrease to a 90% increase in risk. No significant differences between calcimimetics and placebo were shown for fractures or other patient-based end points. This may be caused by insufficient power
because these events are relatively infrequent, artifactual lack of power by failure to measure and/or report patient-based outcomes, or a true lack of benefit. The annual mortality of patients undergoing dialysis is high, at approximately 15 deaths/100 patient-years at risk, of which approximately 40% are cardiac and 11% are vascular.43 Cinacalcet HCl therefore is a potentially important intervention to improve survival, but existing randomized trials have not yet confirmed the hypothesized benefits of this drug. In general, trials reported transient and mild to moderate toxicity, with only a few participants withdrawing from therapy because of adverse events. However, rates of withdrawal and valid precise estimates of such adverse effects as hypotension and nausea often were not available. One trial32 reported that in the calcimimetic group, 6 of 17 patients withdrew because of gastrointestinal events (nausea/vomiting). The few trials reporting toxicity individually suggested that calcimimetics increase the risk for nausea and vomiting and decrease the number of hypotensive episodes and upper respiratory tract infec-
CALCIMIMETICS IN CHRONIC KIDNEY DISEASE
tion, but our meta-analysis of data from all trials that reported these events did not confirm these findings (Table 4). Despite their adoption into clinical practice and approval by the Food and Drug Administration in the United States, the efficacy and tolerability of calcimimetics and their role in the treatment of patients with high-turnover bone disease caused by SHPT has not yet been investigated adequately. Animal studies showed improvement in bone histological characteristics and bone strength with calcimimetic therapy,44 but these findings are yet to be shown in humans, and there are no major clinical data showing an effect on bone histomorphometric characteristics, which is the best determinant of renal bone disease. In addition, there are no adequate clinical data showing an effect of calcimimetics on fracture risk and subsequent morbidity in patients with CKD. Given the recent introduction of calcimimetic therapy, many of these questions may be answered in coming years. It could be suggested that a meta-analysis of therapeutic responses to calcimimetic therapy, particularly longer term outcomes, is premature at this early stage of use. However, cinacalcet HCl is registered by many national drug regulatory agencies (http://www.amgen.com.au/public/ docs/Sensipar_CMI.pdf, pharmacos.eudra.org/ F2/register/register.htm),20 is in widespread clinical use, and information based on metaanalysis of available randomized controlled trials should be valuable to clinicians evaluating whether to use this novel, but expensive, agent. The argument in favor of calcimimetic use hinges on the acceptance of improvements in calcium and phosphorus metabolism and levels of PTH as valid surrogates of such clinically important outcomes as mortality. A surrogate is a measurable outcome, such as a laboratory or imaging test, that is responsive to the effect of an intervention and associated causally with a clinically important outcome. A valid surrogate end point captures the full effect of an intervention, but earlier in the causal chain of events.45-47 Surrogate end points often are used in preference to hard end points in randomized controlled trials because cost and sample size can be decreased and feasibility can be increased substantially. Compared with hard end points, surrogates allow for shorter study duration and either occur more
723
commonly or are continuous measures and therefore more sensitive to differences in treatment. In patients with kidney disease, surrogates commonly are used in trials and include dialysis adequacy, hemoglobin level, left ventricular hypertrophy, and episodes of acute rejection, which have been the basis for the regulatory approval and clinical use of various drugs.48-51 However, not all surrogates are valid proxies of clinically important patient-centered outcomes. For a surrogate to be valid, 2 criteria must be met. First, there must be a strong, independent, and consistent association between the surrogate and the clinically important outcome, which comes from observational studies. For calcium, phosphorus, and PTH levels, this criterion has been met from a number of large-scale cohort and cross-sectional studies.7,8,52,53 Second, and more importantly, for a surrogate to be valid, there also must be evidence that using an intervention changes a surrogate (eg, decrease in PTH levels with a calcimimetic) and results in an expected change in the patient-based outcome distal to the surrogate in the same causal pathway for the disease in question (eg, decrease in deaths with a calcimimetic). This criterion requires a randomized controlled trial that measures both the surrogate and the hard end point. Our study shows that the second criterion has not yet been met for calcimimetics. Critics of the second criterion argue that it is too stringent and will mean that potentially lifesaving interventions will be withheld. Proponents invoke the usual arguments for superiority of randomized controlled trials compared with observational studies (selection bias and unmeasured confounders) to estimate the true effects of interventions. In addition, some interventions that positively affect a surrogate may have a counterintuitive lack of effect on patient-based outcomes in the CKD population, with the atorvastatin in patients with type 2 diabetes mellitus undergoing hemodialysis (4D) study as a recent example.54 Such results suggest that validation of surrogates in disease-specific populations should be mandatory when adopting novel interventions, and trial results based on unvalidated surrogates should be used cautiously. We identified 5 other problems in the trial basis supporting calcimimetic use. First, despite the widespread adoption of CONSORT, a stan-
724
dardized checklist to improve the quality of reporting of randomized controlled trials, the majority of trials of calcimimetics either did not state or did not use (or both) secure methods of allocation, blinded outcomes assessors, or decreased loss to follow-up to a minimum. These design and reporting problems tend to bias results in favor of the intervention.55 Second, to illustrate the importance of prospective registration of trials recently adopted by all major biomedical journals and kidney journals, to ensure that all trials evaluating an intervention are known and linked with publications to avoid publication bias56 and duplication bias,57 we found it very difficult to link publications with studies. We identified 8 unequivocally separate studies with unique data, but also 7 other studies (published in abstract form) that could be duplicates of data already available in full-text published reports. Duplicate reporting is associated with an overestimation of true treatment effects and spurious precision if the trials are incorporated in meta-analysis.57 Prospective registration with a unique identification number for each trial would avoid this. Third, based on the published report, investigators of the largest trial32 combined results of 2 separate, but similar, studies, a method used by the celecoxib long-term arthritis safety study (CLASS) investigators that has been criticized widely.58 When the cinacalcet group of 1 trial is compared with both the placebo group of the same trial (random allocation) and the placebo group of the other trial (nonrandom allocation), outcome differences between the cinacalcet and placebo groups may be caused by differences in study populations or cointerventions, which are unknown and therefore cannot be adjusted for. Having identical trial designs does not prevent these effects. Such trial results would be better reported separately. Data then could be combined by using meta-analysis to provide a summary weighted estimate of the effects shown in the individual trials.32,58 Fourth, CKD is a long-term problem, but except for 1 study, published trials reported treatment outcomes to 6 months only. Finally, reporting of PTH outcomes may have been informative if the proportion of patients achieving a certain end-of-treatment threshold had been reported, and any differences in base-
STRIPPOLI ET AL
line PTH levels may have been better accounted for by reporting change values, rather than endof-treatment levels alone. Outcome reporting was suboptimal, and although our analysis includes all available data reported in the trials, both the benefits and the harms, there were no data on some relevant outcomes, including bone-specific alkaline phosphatase data, PTH target levels achieved, and others. It remains an important, but unanswered, question whether calcimimetic effects on such surrogate end points as biochemical targets and changes in vascular compliance, indices of vascular calcification, and bone mineral density will translate into improved patient-based outcomes. Given reservations relating to current published trials, calcimimetic therapy for patients with SHPT can be considered of potential, but currently unproven, benefit to patient-based outcomes, including cardiovascular mortality, renal osteodystrophy, and fracture. Randomized controlled trials with adequate power and longer treatment duration are required to determine the most appropriate use of this important new class of drugs. ACKNOWLEDGMENT The authors acknowledge the editorial and administrative support of Narelle Willis and Sandra Puckeridge. Ruth Mitchell, Linda Heslop, and Gail Higgins, trial search coordinators of the Cochrane Renal Group, provided search strategies for this review.
REFERENCES 1. Martin KJ, Olgaard K, Coburn JW, et al: Diagnosis, assessment, and treatment of bone turnover abnormalities in renal osteodystrophy. Am J Kidney Dis 43:558-565, 2004 2. Stehman-Breen C: Osteoporosis and chronic kidney disease. Semin Nephrol 24:78-81, 2004 3. Marco MP, Craver L, Betriu A, Belart M, Fibla J, Fernandez E: Higher impact of mineral metabolism on cardiovascular mortality in a European hemodialysis population. Kidney Int Suppl 85:S111-S114, 2003 4. Block GA, Klassen PS, Lazarus JM, Ofsthun N, Lowrie EG, Chertow GM: Mineral metabolism, mortality, and morbidity in maintenance hemodialysis. J Am Soc Nephrol 15:22082210, 2004 5. Ganesh SK, Stack AG, Levin NW, Hulbert-Shearon T, Port FK: Association of elevated serum PO4, Ca ⫻ PO4 product, and parathyroid hormone with cardiac mortality risk in chronic hemodialysis patients. J Am Soc Nephrol 12:2131-2138, 2001 6. Malluche HH, Mawad H, Monier-Faugere MC: The importance of bone health in end-stage renal disease: Out of the frying pan, into the fire? Nephrol Dial Transplant 19:i9i13, 2004
CALCIMIMETICS IN CHRONIC KIDNEY DISEASE
7. Kestenbaum B, Sampson JN, Rudser KD, et al: Serum phosphate levels and mortality risk among people with chronic kidney disease. J Am Soc Nephrol 16:520-528, 2005 8. Stevens LA, Djurdjev O, Cardew S, Cameron EC, Levin A: Calcium, phosphate, and parathyroid hormone levels in combination and as a function of dialysis duration predict mortality: Evidence for the complexity of the association between mineral metabolism and outcomes. J Am Soc Nephrol 15:770-779, 2004 9. Hutchison AJ, Whitehouse RW, Boulton HF, et al: Correlation of bone histology with parathyroid hormone, vitamin D3, and radiology in end-stage renal disease. Kidney Int 44:1071-1077, 1993 10. Qi Q, Moniere-Faugere MC, Geng Z, Malluche HH: Predictive value of serum parathyroid hormone levels for bone turnover in patients on chronic maintenance hemodialysis. Am J Kidney Dis 26:622-631, 1995 11. Wang M, Hercz G, Sherrard DJ, Maloney NA, Segre GV, Pei Y: Relationship between intact 1-84 parathyroid hormone and bone histomorphometric parameters in dialysis patients without aluminum toxicity. Am J Kidney Dis 26:836844, 1995 12. Ziolkowska H, Panczyk-Tomaszewska M, Debinski A, Polowiec Z, Sawicki A, Sieniawska M: Bone biopsy results and serum bone turnover parameters in uremic children. Acta Paediatr 89:666-671, 2000 13. National Kidney Foundation: K/DOQI Clinical Practice Guidelines for Bone Metabolism and Disease in Chronic Kidney Disease. Am J Kidney Dis 42:S1-S201, 2003 (suppl 3) 14. CARI Guidelines. Available at: http://www.kidney. org.au/cari/guidelines.php. Accessed July 1, 2005 15. Young EW, Akiba T, Albert JM, et al: Magnitude and impact of abnormal mineral metabolism in hemodialysis patients in the Dialysis Outcomes and Practice Patterns Study (DOPPS). Am J Kidney Dis 44:S34-S38, 2004 (suppl 3) 16. Albaaj F, Hutchison A: Hyperphosphatemia in renal failure: Causes, consequences and current management. Drugs 63:577-596, 2003 17. Courant O, Letessier E, Moutel MG, Hamy A, Paineau J, Visset J: [Surgical treatment of secondary hyperparathyroidism in chronic kidney failure. Results of total parathyroidectomy with parathyroid autotransplantation]. J Chir 130:327334, 1993 18. Goodman WG, Frazao JM, Goodkin DA, Turner SA, Liu W, Coburn JW: A calcimimetic agent lowers plasma parathyroid hormone levels in patients with secondary hyperparathyroidism. Kidney Int 58:436-445, 2000 19. Ott SM: Calcimimetics—New drugs with the potential to control hyperparathyroidism. J Clin Endocrinol Metab 83:1080-1082, 1998 20. Amgen Inc: Sensipar (Cinacalcet) HCl Tablets. Available at: http://www.fda.gov/cder/foi/label/2004/21688_ Sensipar_lbl.pdf. Accessed July 1, 2005 21. Higgins JP, Thompson SG, Deeks JJ, Altman DG: Measuring inconsistency in meta-analyses. BMJ 327:557560, 2003 22. Coburn JW, Charytan C, Chonchol M, et al: Cincalcet HCl is an effective treatment for secondary hyperparathyroidism (HPT) in patients with chronic kidney disease (CKD) not yet receiving dialysis. J Am Soc Nephrol 14:460A, 2003 (abstr)
725
23. Goodman WG, Fadda GZ, Finkelstein FO, et al: Cincalcet HCl is an effective primary therapy for the management of secondary hyperparathyroidism. J Am Soc Nephrol 14:460A, 2003 (abstr) 24. Martin KJ, Sherrard DJ, Nassar G, et al: Control of bio-intact parathyroid hormone (PTH) levels with the calcimimetic cinacalcet HCl in hemodialysis patients with secondary hyperparathyroidism (HPT). J Am Soc Nephrol 14: 462A, 2003 (abstr) 25. Quarles LD, Zeig S, Spiegel DM, et al: Cincalcet HCl controls secondary hyperparathyroidism (HPT) in dialysis patients regardless of disease severity. J Am Soc Nephrol 14:463A, 2003 (abstr) 26. Sterrett JR, Strom J, Stummvoll HK, et al: Long-term treatment of secondary hyperparathyroidism (HPT) with cinacalcet HCl (Sensipar) in patients receiving dialysis. J Am Soc Nephrol 15:275A, 2004 (abstr) 27. Cunningham J, Holzer H, Reichel H, et al: Sustained, long-term reductions in parathyroid hormone (iPTH) after 3 years of treatment with the calcimimetic cinacalcet HCl in patients with secondary hyperparathyroidism (SHPT). Nephrol Dial Transplant 18:S150A-S151A, 2003 (suppl 4; abstr) 28. Cunningham J, Chertow GM, Goodman WG, et al: The effect of cincalcinet HCl on parathyroidectomy, fracture, hospitalization, and mortality in dialysis subjects with secondary hyperparathyroidism (HPT). Proceedings of the European Renal Association-European Dialysis and Transplant Association XLI Congress, May 15-18, 2004, Lisbon, Portugal, p 219 (abstr) 29. Fournier A, Pontoiero G, Walker R, et al: Cinacalcet reduces parathyroid hormone (PTH) and calcium-phosphorus product (Ca ⫻ P) regardless of concurrent changes in vitamin D administration. Proceedings of the European Renal Association-European Dialysis and Transplant Association XLI Congress, May 15-18, 2004, Lisbon, Portugal, p. 319 (abstr) 30. De Francisco A, Suranyi MG, Cunningham J, et al: Greater achievement of NKF/DOQI bone metabolism and disease targets with oral cinacalcet HCl in an EU/Australian phase 3 study in hemodialysis subjects. Proceedings of the European Renal Association-European Dialysis and Transplant Association XLI Congress, May 15-18, 2004, Lisbon, Portugal, p 319 (abstr) 31. Frazao JM, Nicolini M, Torregrosa JV, et al: Cinacalcet HCl effectively reduces intact parathyroid hormone (iPTH) and Ca ⫻ P irrespective of the severity of secondary hyperparathyroidism (HPT). Proceedings of the European Renal Association-European Dialysis and Transplant Association XLI Congress, May 15-18, 2004, Lisbon, Portugal, p 219 (abstr) 32. Block GA, Martin KJ, de Francisco AL, et al: Cinacalcet for secondary hyperparathyroidism in patients receiving hemodialysis. N Engl J Med 350:1516-1525, 2004 33. Goodman WG, Hladik GA, Turner SA, et al: The calcimimetic agent AMG 073 lowers plasma parathyroid hormone levels in hemodialysis patients with secondary hyperparathyroidism. J Am Soc Nephrol 13:1017-1024, 2002 34. Harris RZ, Padhi D, Marbury TC, Noveck RJ, Salfi M, Sullivan JT: Pharmacokinetics, pharmacodynamics, and safety of cinacalcet hydrochloride in hemodialysis patients at
726
doses up to 200 mg once daily. Am J Kidney Dis 44:1070-1076, 2004 35. Lindberg JS, Moe SM, Goodman WG, et al: The calcimimetic AMG 073 reduces parathyroid hormone and calcium x phosphorus in secondary hyperparathyroidism. Kidney Int 63:248-254, 2003 36. Lindberg JS, Culleton B, Wong G, et al: Cinacalcet HCl, an oral calcimimetic agent for the treatment of secondary hyperparathyroidism in hemodialysis and peritoneal dialysis: A randomized, double-blind, multicenter study. J Am Soc Nephrol 16:800-807, 2005 37. Malluche HH, Monier-Faugere MC, Wang G, et al: Cinacalcet HCl reduces bone turnover and bone marrow fibrosis in hemodialysis patients with secondary hyperparathyroidism (HPT). Proceedings of the European Renal Association-European Dialysis and Transplant Association XLI Congress, May 15-18, 2004, Lisbon, Portugal, p 218 (abstr) 38. Quarles LD, Sherrard DJ, Adler S, et al: The calcimimetic AMG 073 as a potential treatment for secondary hyperparathyroidism of end-stage renal disease. J Am Soc Nephrol 14:575-583, 2003 39. Chertow GM, Corpier C, Zeig S, et al: Preliminary results from TARGET: Treatment strategies to achieve recommended K/DOQI™ goals in ESRD patients on cinacalcet. J Am Soc Nephrol 15:863A, 2004 (abstr) 40. Reed J, Keightley G, Blumenthal S, et al: The CONTROL Study: Enhanced achievement of NKF-K/ DOQI™ bone metabolism and disease targets using cinalcalcet HCl (Sensipar™). J Am Soc Nephrol 15:280A, 2004 (abstr) 41. Urena P, Frazao JM: Calcimimetic agents: Review and perspectives. Kidney Int Suppl 85:S91-S96, 2003 42. Moher D, Schulz KF, Altman D, for the CONSORT Group (Consolidated Standards of Reporting Trials): The CONSORT statement: Revised recommendations for improving the quality of reports of parallel-group randomized trials. JAMA 285:1987-1991, 2001 43. ANZDATA: 27th Annual Report. Available at: http:// www.anzdata.org.au/anzdata/AnzdataReport/download.htm. Accessed August 8, 2005 44. Sherrard DJ: Calcimimetics in action. Kidney Int 53:510-511, 1998 45. Bucher HC, Guyatt GH, Cook DJ, et al: Users’ guides to the medical literature: XIX. Applying clinical trial results. A. How to use an article measuring the effect of an intervention on surrogate endpoints. Evidence-Based Working Group. JAMA 282:771-778, 1999
STRIPPOLI ET AL
46. Psaty BM, Weiss NS, Furberg CD, et al: Surrogate endpoints, health outcomes, and the drug-approval process for the treatment of risk factors for cardiovascular disease. JAMA 282:786-790, 1999 47. Temple R: Are surrogate markers adequate to assess cardiovascular disease drugs? JAMA 282:790-795, 1999 48. Churchill DM, Thorpe KE, Vonesh EF, Keshaviah PR: Lower probability of patient survival with continuous peritoneal dialysis in the United States compared with Canada. Canada-USA (CANUSA) Peritoneal Dialysis Study Group. J Am Soc Nephrol 8:965-971, 1997 49. Besarab A, Bolton WK, Browne JK, et al: The effects of normal, as compared with low, hematocrit values in patients with cardiac disease who are receiving hemodialysis and epoetin. N Engl J Med 339:584-590, 1998 50. McMahon LP, Roger SD, Levin A, for the SLIMHEART Investigators Group: Development, prevention and potential reversal of left ventricular hypertrophy in chronic kidney disease. J Am Soc Nephrol 15:1640-1647, 2004 51. Borrows R, Loucaidou M, van Tromp J, et al: Steroid sparing with tacrolimus and mycophenolate mofetil in renal transplantation. Am J Transplant 4:1845-1851, 2004 52. Avram MM, Sreedhara R, Avram DK, Muchnick RA, Fein P: Enrollment parathyroid hormone level is a new marker of survival in hemodialysis and peritoneal dialysis therapy for uremia. Am J Kidney Dis 28:924-930, 1996 53. Ganesh SK, Stack AG, Levin NW, Hulbert-Shearon T, Port FK: Association of elevated serum PO4, Ca ⫻ PO4 product, and parathyroid hormone with cardiac mortality risk in chronic hemodialysis patients. J Am Soc Nephrol 12:2131-2138, 2001 54. Wanner C, Krane V, Marz W, et al, for the German Diabetes and Dialysis Study Investigators: Atorvastatin in patients with type 2 diabetes mellitus undergoing haemodialysis. N Engl J Med 353:238-248, 2005 55. Schulz KF, Chalmers I, Hayes RJ, Altman DG: Empirical evidence of bias. Dimensions of methodological quality associated with estimates of treatment effects in controlled trials. JAMA 273:408-412, 1995 56. Simes RJ: Publication bias: The case for an international registry of clinical trials. J Clin Oncol 4:1529-1541, 1986 57. Egger M, Smith GD: Meta-analysis bias in location and selection of studies. BMJ 316:61-66, 1998 58. Juni P, Sterchi R, Dieppe P: Systematic review of celecoxib for osteoarthritis and rheumatoid arthritis. BMJ 326:334, 2003