Daniel M. Keller, PhD

June 27, 2012

June 27, 2012 (Paris, France) — Patients receiving hemodialysis may benefit greatly, with a lower risk for cardiovascular (CV) events, from the use of sevelamer to treat hyperphosphatemia.

Antonio Bellasi, MD, from the Ospedale Sant'anna in Como, Italy, reported the results of a prospective, open-label, randomized controlled trial that compared sevelamer with calcium-based phosphorus binding agents (PBAs) here at the XLIX European Renal Association-European Dialysis and Transplant Association (ERA-EDTA) Congress.

Sevelamer does not contain calcium, and when taken with meals, it binds to dietary phosphate and prevents its absorption, Dr. Bellasi pointed out.

The investigators randomly assigned 466 hemodialysis patients to receive either sevelamer (n = 232) or calcium-containing PBAs (n = 234) and followed them until study completion or until CV-related death occurred. Thirty-three patients in the sevelamer group and 36 in the calcium PBA group were lost to follow-up.

Patients were evenly divided between men and women and were generally of late middle age (mean, 65 ± 14 years), and 79% of them had hypertension, 36% CV disease, and 29% diabetes at baseline.

At study initiation, patients in the sevelamer group had higher serum phosphorus levels but lower coronary artery calcification (CAC) scores compared with patients in the calcium PBA group.

Allocation to either group was blinded, but the trial was open-label. "Because of the impact of sevelamer on lipids, it was deemed impossible to pursue a double-blind trial," Dr. Bellasi explained. However, investigators blinded to the treatment allocation adjudicated the outcomes.

Multiple Benefits With Sevelamer

At 36 months of follow-up (mean, 28 ± 10 months), the patients using sevelamer had almost a 10-fold lower risk for CV-related mortality than the patients taking the calcium-containing PBA (relative risk [RR], 0.11; 95% confidence interval [CI], 0.05 - 0.22; P < .001).

There were 9 CV deaths in the former group vs 79 in the latter out of the entire study population of 466 patients. The relative risk held constant at 0.11, whether it was unadjusted or adjusted for age and sex or for age, sex, diabetes, systolic, and diastolic blood pressure. Even adding stratification for CAC scores at baseline had no effect on the outcome; that is, sevelamer was still associated with a very large amount of CV-related risk reduction.

Similarly, the sevelamer group did much better in terms of all-cause survival, with a greater than 70% reduction in the risk for all-cause death regardless of whether the RR was unadjusted or adjusted for the previously mentioned factors (RR range, 0.26 - 0.29; all P < .001). There were 28 deaths in the sevelamer group at 36 months compared with 100 deaths in the calcium PBA group.

Sevelamer use was linked to less progression of CAC when looked at over the course of 24 months of the study.

At 12 months, half of the calcium PBA group had CAC progression, whereas only 20.9% had progression in the sevelamer group (P < .001). At 24 months, CAC progression affected 66.3% of patients taking calcium PBA but only 38.4% of those receiving sevelamer (P < .001).

Even adjusting for baseline CAC score, sevelamer was associated with a 72% reduction in the risk for CAC progression (RR, 0.28; 95% CI, 0.17 - 0.47; P < .001). The beneficial effect of sevelamer persisted throughout all of the CAC strata, indicating that the beneficial effect of sevelamer was independent of the coronary artery calcium burden.

It appeared that cardiac arrhythmias were a significant contributor to the CV deaths because sevelamer was associated with a 93% reduction in the risk for arrhythmias (unadjusted RR, 0.07; 95% CI, 0.01 - 0.31; P < .001). Adjusting for the other parameters yielded essentially the same result. The corrected QT interval remained fairly constant in the sevelamer group but rose over time for the patients assigned to a calcium PBA (P < .001).

The pulse wave velocity, an indicator of arterial stiffness, rose significantly in the calcium PBA group vs the sevelamer group during the first 24 months of the trial (P < .001).

Sevelamer did not appear to reduce the risk for death from causes other than CV, so all of its beneficial effect on all-cause mortality risk appeared to come from its effect on CV mortality. Non-CV deaths, which were a secondary endpoint of the trial, were essentially no different at 36 months between the sevelamer group (19 deaths) and the calcium PBA group (21 deaths), with relative risks for the sevelamer group ranging from 0.83 to 1.18 (log-rank test, P = .188).

Session moderator David Goldsmith, MB BChir, consultant nephrologist in the Renal Unit at Guy's and St. Thomas' National Health Service Foundation Hospital in London, United Kingdom, called the study findings "very interesting" and praised it for its size and duration. However, he advised caution in interpreting the results, as the trial was not fully blinded and some important factors remain unknown, so the very positive effect of sevelamer should not be generalized yet.

"Were this to be true, then it will be remarkable because there are virtually no other treatments that have been shown to reduce mortality in this group of patients," Dr. Goldsmith told Medscape Medical News. "But I think there are some idiosyncrasies in the way it was done and some uncertainties in my mind about the way in which it's been analyzed that leave me not completely convinced."

He said he would be more satisfied if the study is given a thorough review and is then published in a good journal.

He said that in addition to uncertainty about compliance in the trial, "We don't know things about hypercalcemic episodes. We don't really know why the phosphate didn't change in [the calcium PBA group from baseline] and did in [the sevelamer group]."

Dr. Goldsmith delineated some of the things he would still like answered: "What's the mortality incidence by calcium and phosphate achieved, what's the mortality incidence by baseline coronary artery calcification score, what were the interactions between those 2 things, and was there any dose effect in terms of sevelamer or calcium in terms of mortality or calcification because we could look for a threshold to see whether a small amount of calcium is safe?" he asked.

Many of the answers are probably in the data set. "I think there's a great deal of useful information, which hopefully will come out in a primary publication."

Dr. Goldsmith said calcium-containing PBAs are still used based on "tradition and cost, but mainly cost.... So they are very, very cheap, and it's uncertain whether or not they are harmful, and the sort of studies that would answer those questions have really never been done. This one is the closest anybody's got by administering them for long enough to see an impact because it clearly doesn't happen in a year, any mortality impact, and all we've had so far is indirect surrogates like [CAC], but we haven't had hard endpoint studies."

He advised that more information is needed before anyone should accept the study and its apparently very positive results as conclusive, "and I suspect it will be controversial in some quarters. It was a real-life study with all the uncertainties that flow from that. Therefore, can you really infer from it that the only difference between the 2 groups was truly the phosphate binding strategy or whether there were other unknown things that influenced directly or indirectly what was going on?"

Although the results are important to know, the focus in the field has probably shifted. "I don't think there's any appetite now for doing these sorts of big trials with different binders," Dr. Goldsmith said. "I think the world has moved on to things like calcium sensing receptor [studies]."

The study was not commercially funded. Dr. Bellasi has received honoraria from Genzyme and Amgen. Dr. Goldsmith was not involved in the study. He has received research support from Genzyme, the maker of sevelamer, and is a consultant to Sanofi, the parent company of Genzyme.

XLIX European Renal Association-European Dialysis and Transplant Association (ERA-EDTA) Congress: Abstract SAO059. Presented May 26, 2012.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....