NDT Advance Access originally published online on July 24, 2009
Nephrology Dialysis Transplantation 2009 24(10):2970-2972; doi:10.1093/ndt/gfp366
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Sevelamer versus calcium-based phosphate binders in chronic kidney disease: what should we conclude from the evidence to date?
Department of Renal Medicine, University of Queensland at Princess Alexandra Hospital, Brisbane, Australia
Correspondence and offprint requests to: David W. Johnson; E-mail: david_johnson{at}health.qld.gov.au
Keywords: chronic kidney disease; lanthanum; phosphate binders; sevelamer; vascular calcification
Calcium-based phosphate binders have traditionally been considered standard first-line treatment for control of hyperphosphataemia in chronic kidney disease (CKD). However, several cross-sectional studies have observed that prescribed calcium intake from phosphate binders is dose-dependently associated with the degree of vascular calcification in CKD patients [1,2]. Other studies have demonstrated an association between vascular calcification scores and an increased risk of cardiovascular disease [3–5]. Although there is presently no proof that interventions that reduce the surrogate outcome of vascular calcification lead to reduced cardiovascular events in CKD patients, the K/DOQI guidelines have recommended that the ‘total dose of elemental calcium provided by the calcium-based phosphate binders should not exceed 1500 mg/day (opinion), and the total intake of elemental calcium (including dietary calcium) should not exceed 2000 mg/day (opinion)’ [6]. If these guidelines are followed, there is likely to be a substantial shift in clinical practice away from the use of calcium-based phosphate binders to non-calcium-based phosphate binders, such as sevelamer, which are considerably more expensive. This raises the important question of just how good is the evidence that sevelamer therapy results in superior patient-level outcomes (i.e. ‘hard’ clinical end-points) compared with calcium-based phosphate binder therapy?
To date, there has only been one randomized controlled trial (RCT) of sevelamer versus calcium-based binders, which has been adequately powered for mortality. The Dialysis Clinical Outcomes Revisited (DCOR) study involved 75 centres and 2103 haemodialysis patients randomly allocated to receive either sevelamer or a calcium-based phosphate binder for a mean follow-up period of 
20 months [7]. Sevelamer-treated patients experienced comparable all-cause mortality [hazard ratio (HR) 0.93, 95% confidence interval (CI) 0.79–1.10] and cardiovascular mortality (HR 0.93, 95% CI 0.74–1.17) to patients receiving calcium-based binders. Based on a significant interaction between the pre-specified variable of age and treatment effect (P = 0.02), a subsequent age-stratified analysis demonstrated that sevelamer was associated with a reduction in mortality in patients aged 65 years or older (HR 0.77, 0.61–0.96). However, these results were limited by the fact that the authors did not correct their interaction assessments for multiple comparisons, which would have yielded a non-significant result. A subsequent intention-to-treat analysis of the DCOR study using pre-planned secondary analyses and Centers for Medicare & Medicaid Services data demonstrated no significant differences in overall or cardiovascular mortality and no significant interaction between age and treatment effect [8].
A meta-analysis of RCTs, including the DCOR study, observed that sevelamer treatment was associated with similar all-cause mortality (risk difference –2%, 95% CI –6%–2%; 5 RCTs and 2429 participants) and cardiovascular mortality (risk difference –1%, 95% CI –4%–2%; 3 RCTs and 2102 participants) compared with calcium-based phosphate binder therapy [9]. The authors concluded that there was ‘no convincing evidence that sevelamer improves clinically relevant outcomes in ESRD patients’ and that ‘recommendations for the routine use of sevelamer in dialysis patients are not supported by current data.’
In this issue of Nephrology Dialysis and Transplantation, Jamal et al. report the results of an updated systematic review of the effects of calcium-based versus non-calcium-based phosphate binders on mortality in CKD patients [10]. Eight RCTs of sevelamer versus calcium-based binders were selected. The study was sponsored by Genzyme (manufacturers of sevelamer) and one of the authors declared a potential conflict of interest with Genzyme. Similar to the findings of the previous meta-analysis, Jamal et al. reported no significant differences between patients treated with sevelamer or calcium-based phosphate binders with respect to coronary artery calcification (WMD –76.35, 95% CI –158.25–5.55; 5 RCTs and 469 participants), cardiovascular events [relative risk (RR) 0.85, 95% CI 0.35–2.03; two RCTs and 153 participants] or all-cause mortality (RR 0.68, 95% CI 0.41–1.11; eight RCTs and 2873 participants). In spite of the non-significant differences between the groups, the authors concluded that there were ‘trends’ towards decreased all-cause mortality, cardiovascular mortality and coronary artery calcification associated with non-calcium-based phosphate binders and that future studies are needed to conduct ‘research on the mechanism by which non-calcium-based phosphate binders reduce mortality or alternatively, how calcium-based binders increase mortality’. This article raises important questions regarding interpretation of the results of meta-analyses and how broad or limited the subsequent conclusions/recommendations should be.
In my opinion, it is inappropriate to state that there were trends towards a reduction in all-cause or cardiovascular mortality with sevelamer when the confidence limits were relatively wide and appreciably overlapped a RR of 1.0. Although the authors state that the ‘confidence intervals do not exclude a potentially important beneficial effect’, they also do not confidently exclude an 11% increased risk of all-cause mortality or a doubling of cardiovascular events associated with sevelamer. The argument for a trend towards a reduction in coronary artery calcification with sevelamer is further undermined by the unproven clinical validity of vascular calcification as a surrogate outcome for cardiovascular and overall mortality [11]. Moreover, the meta-analysis had significant limitations, which reduced confidence in its conclusions of trends favouring sevelamer.
Ideally, a meta-analysis should specify stringent criteria used for subject inclusion, experimental intervention and outcome determination a priori to ensure that these are similar in all selected trials [12]. The study by Jamal et al. indicates that a complete search strategy is available online as an appendix, but does not clearly specify the selection criteria. Although RCTs using lanthanum as the non-calcium-based binder formed part of the inclusion criteria, it is not clear why certain published RCTs comparing lanthanum with calcium-based binders [13,14] were not included. Moreover, the I2 statistic for the primary all-cause mortality analysis (47.8%) suggests at least a moderate degree of trial heterogeneity and raises questions about the validity of summarizing all of the mortality information contained in the currently available trials into a single odds ratio.
Although it is estimated that only one-half to two-thirds of all RCTs are published [15,16], the systematic review by Jamal et al. only considered published RCTs. This raises the possibility of reporting bias, since RCTs with significant and positive results are likely to be published more often (publication bias) and more rapidly (time lag bias) than those with non-significant or ‘negative’ results [15]. Indeed, the potential for publication bias was suggested by an asymmetrical funnel plot and a supportive weighted regression test (P = 0.07). A previous meta-analysis by Tonelli et al. [9] found statistically significant evidence of publication bias.
The meta-analysis of Jamal et al. was also limited by the generally poor quality of included studies. Three of the eight RCTs (contributing 2200 of the 2873 subjects) were appraised as having a high risk of bias due to inadequate sequence generation, allocation concealment and/or blinding, whilst two RCTs (242 patients) had an unclear risk and only three RCTs (431 participants) had a low risk of bias. Study quality was further limited by high drop-out rates (ranging between 8% and 38% for coronary calcification studies) and frequent lack of intention-to-treat analyses leading to possible informative censoring bias.
Meta-analyses should also ideally consider harms as well as benefits of interventions. Whilst these were not considered by Jamal et al., a previous meta-analysis observed that one RCT reported a significant difference for serious gastrointestinal complaints (33%, 95% CI 9–58%) favouring calcium-based phosphate binders over sevelamer [9,17]. Clearly, more study is needed in this area.
More study is also required in the area of cost-effectiveness of phosphate binder therapy. A previous economic evaluation of sevelamer in CKD patients considering four separate Markov modelling strategies based on data obtained from the DCOR study [18] found that the use of sevelamer was associated with a cost per quality-adjusted life year (QALY) gained that exceeded what would usually be considered a good value for money (CAN$105 000–CAN$278 100). The authors concluded that ‘this strategy remains economically unattractive, particularly given the uncertainty of clinical benefit in this group’.
So, to come back to the question posed in the title, how much can we conclude from the available studies of sevelamer versus calcium-based phosphate binders in CKD patients? I believe that the evidence from one large, adequately powered RCT and two meta-analyses containing a limited number of additional small, generally poor quality RCTs would suggest that there is no significant difference between sevelamer and calcium-based phosphate binders with respect to the important patient level outcomes of overall and cardiovascular mortality in either the total CKD population or in sub-groups. Interpreting multiple non-significant results in the most recent systematic review as trends favouring sevelamer is neither helpful nor justified based on the imprecision of the results and the limitations of the meta-analysis. The available evidence also raises significant doubts about the value of vascular calcification as a valid surrogate outcome of mortality in phosphate binder trials. There has been inadequate study to date of both the relative harms and cost-effectiveness of sevelamer versus calcium-based phosphate binders in CKD patients. Although a net benefit of sevelamer with respect to patient level outcomes cannot be conclusively excluded, the available evidence would suggest that any such benefit, if it exists at all, is modest at best and must be carefully weighed against the considerable additional costs of this therapy. A further large, well-designed and executed RCT of sevelamer versus calcium-based phosphate binders adequately powered for a 15–20% reduction in mortality with consideration of adverse events and costs as secondary outcomes and a pre-specified analysis of mortality in an elderly sub-group would be helpful to finally lay the matter to rest, one way or the other. Until then, recommendations for the routine or preferential use of sevelamer in CKD patients are not justified and may ultimately prove to be highly cost-ineffective.
‘It is the peculiar and perpetual error of the human understanding to be more moved and excited by affirmatives than negatives’. Sir Francis Bacon (1561–1626)
Conflict of interest statement. D.J. has served on the consultant advisory boards for Genzyme (manufacturer of sevelamer) and Amgen (manufacturer of cinacalcet). He has received speakers’ honoraria, research funding and travel sponsorships from Amgen.
(See related article by S. A. Jamal et al. The effects of calcium-based versus non-calcium-based phosphate binders on mortality among patients with chronic kidney disease: a meta-analysis. Nephrol Dial Transplant 2009; 24: 3168–3174.)
 |
References |
|---|
 Top References |
|---|
- Goodman WG, Goldin J, Kuizon BD, et al. Coronary-artery calcification in young adults with end-stage renal disease who are undergoing dialysis. N Engl J Med (2000) 342:1478–1483.
[Abstract/Free Full Text] - Guerin AP, London GM, Marchais SJ, et al. Arterial stiffening and vascular calcifications in end-stage renal disease. Nephrol Dial Transplant (2000) 15:1014–1021.
[Abstract/Free Full Text] - Winkelmayer WC, Tonelli M. Phosphate binder choice in dialysis patients: a call for evidence-based rather than marketing-based clinical practice. Am J Kidney Dis (2008) 51:362–365.[CrossRef][Web of Science][Medline]
- Matsuoka M, Iseki K, Tamashiro M, et al. Impact of high coronary artery calcification score (CACS) on survival in patients on chronic hemodialysis. Clin Exp Nephrol (2004) 8:54–58.[CrossRef][Medline]
- Block GA, Raggi P, Bellasi A, et al. Mortality effect of coronary calcification and phosphate binder choice in incident hemodialysis patients. Kidney Int (2007) 71:438–441.[CrossRef][Web of Science][Medline]
- K/DOQI. Clinical practice guidelines for bone metabolism and disease in chronic kidney disease. Am J Kidney Dis (2003) 42:S1–S201.[Medline]
- Suki WN, Zabaneh R, Cangiano JL, et al. Effects of sevelamer and calcium-based phosphate binders on mortality in hemodialysis patients. Kidney Int (2007) 72:1130–1137.[CrossRef][Web of Science][Medline]
- St Peter WL, Liu J, Weinhandl E, et al. A comparison of sevelamer and calcium-based phosphate binders on mortality, hospitalization, and morbidity in hemodialysis: a secondary analysis of the Dialysis Clinical Outcomes Revisited (DCOR) randomized trial using claims data. Am J Kidney Dis (2008) 51:445–454.[CrossRef][Web of Science][Medline]
- Tonelli M, Wiebe N, Culleton B, et al. Systematic review of the clinical efficacy and safety of sevelamer in dialysis patients. Nephrol Dial Transplant (2007) 22:2856–2866.
[Abstract/Free Full Text] - Jamal SA, Fitchett D, Lok CE, et al. The effects of calcium-based versus non-calcium-based phosphate binders on mortality among patients with chronic kidney disease: a meta-analysis. Nephrol Dial Transplant (2009) 24:3168–3174.
[Abstract/Free Full Text] - O’Shea S, Johnson DW. Addressing risk factors in CKD-MBD: can we influence patient-level outcomes? Nephrology (2009) 14(4):416–427.
- LeLorier J, Gregoire G, Benhaddad A, et al. Discrepancies between meta-analyses and subsequent large randomized, controlled trials. N Engl J Med (1997) 337:536–542.
[Abstract/Free Full Text] - Hutchison AJ, Maes B, Vanwalleghem J, et al. Efficacy, tolerability, and safety of lanthanum carbonate in hyperphosphatemia: a 6-month, randomized, comparative trial versus calcium carbonate. Nephron Clin Pract (2005) 100:c8–c19.[CrossRef][Web of Science][Medline]
- D’Haese PC, Spasovski GB, Sikole A, et al. A multicenter study on the effects of lanthanum carbonate (Fosrenol) and calcium carbonate on renal bone disease in dialysis patients. Kidney Int Suppl (2003) S73–S78.
- Hopewell S, Loudon K, Clarke MJ, et al. Publication bias in clinical trials due to statistical significance or direction of trial results. Cochrane Database Syst Rev (2009; Jan 21 (1):MR000006).
- Scherer RW, Langenberg P, von Elm E. Full publication of results initially presented in abstracts. Cochrane Database Syst Rev (2007; Apr 18 (2):MR000005).
- Sadek T, Mazouz H, Bahloul H, et al. Sevelamer hydrochloride with or without alphacalcidol or higher dialysate calcium vs calcium carbonate in dialysis patients: an open-label, randomized study. Nephrol Dial Transplant (2003) 18:582–588.
[Abstract/Free Full Text] - Manns B, Klarenbach S, Lee H, et al. Economic evaluation of sevelamer in patients with end-stage renal disease. Nephrol Dial Transplant (2007) 22:2867–2878.
[Abstract/Free Full Text]
Accepted in revised form: 1. 7.09
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||