In Search of the Optimal Target for Phosphate Control: Episode 1

Hyperphosphatemia is near ubiquitous among patients receiving maintenance dialysis. Compelling experimental data suggest that hyperphosphatemia is directly toxic to the vasculature.¹ Numerous epidemiologic studies demonstrated an association between elevated serum phosphate and mortality.² These observations profoundly influenced clinical practice and informed guidelines, which suggest that serum phosphate be lowered “toward the normal range” in patients receiving maintenance dialysis.³ This approach has resulted in $1.5 billion spent in 2015 on phosphate-lowering medications in the US Medicare system alone, and between 2008 and 2013, phosphate binder costs accelerated more rapidly than other drugs prescribed to patients on dialysis.⁴ Despite the implications of serum phosphate lowering, this strategy has never been shown to improve survival nor the way patients function or feel.

In this issue of JASN, Isaka et al.⁵ report the results of the Evaluate the New Phosphate Iron-Based Binder Sucroferric Oxyhydroxide in Dialysis Patients with the Goal of Advancing the Practice of EBM (EPISODE) trial. This multicenter, Japanese 2×2 factorial trial randomized 160 patients on hemodialysis with hyperphosphatemia and coronary artery calcification to one of two phosphate binders (sucroferric oxyhydroxide versus lanthanum) and also, two targets for phosphate control: strict, defined as a serum phosphate concentration of 3.5–4.5 mg/dl, versus standard, defined as a serum phosphate concentration of 5.0–6.0 mg/dl. The primary outcome was the relative change in coronary calcium score over a follow-up period of 12 months. The progression of coronary calcification did not differ between patients allocated to either of the two phosphate binders who achieved a similar reduction in serum phosphate concentration. However, patients allocated to strict phosphate control (mean serum phosphate 4.68 mg/dl) had a significant attenuation in the progression of coronary artery calcification (8.5% versus 21.8%, P=0.006) compared with those allocated to standard control (mean serum phosphate 5.54 mg/dl at 12 months). These findings remained robust when absolute changes in vascular calcification were evaluated as well as in sensitivity analyses that attempted to account for incomplete data.

The EPISODE trial is an important contribution to the study of serum phosphate management in dialysis recipients for several reasons. The vast majority of trials in this area have focused on proving that binders lower serum phosphate concentration (as designed), or they compared performance of novel binders with those of prior generations. However, no trial to date has tested the biologic effect of phosphate lowering. This would be akin to demonstrating that statins lowered cholesterol and then skipping to trials that compare statins with one another without ever first proving that statins prevent cardiovascular events. EPISODE is the first completed trial to address the fundamental question of whether a strategy of guideline-endorsed phosphate normalization, as executed in the strict arm of the trial, has a favorable effect on cardiovascular health.

A second advance of the EPISODE trial is confirmation that it is possible to generate between-group differences in serum phosphate for 12 months, confirming and extending the findings of pilot randomized controlled trials conducted in Canada and the United Kingdom.^6,7 These three trials together demonstrated similar differences in serum phosphate between groups despite each using different binders and protocols, further suggesting the feasibility of large, pragmatic trials addressing the controversy of whether a strategy to normalize phosphate has important clinical benefits. The third advance is the observation that a <1-mg/dl difference in phosphate resulted in a difference in coronary artery calcium scores. As such, even modest achieved differences in serum phosphate might have relevant clinical benefits.

Like all trials, EPISODE had limitations. Notably, it was small (and smaller than planned), was open label, and had relatively high losses to follow-up. These limitations may reduce our confidence in the existence of a true effect of phosphate lowering on coronary artery calcium scores. Most importantly, although coronary artery calcification associates with future cardiovascular events,⁸ it is uncertain whether the magnitude of the intergroup difference in calcification scores described in EPISODE and previous trials involving patients on dialysis⁹ translates into patient-important benefits. This is a vital issue because phosphate lowering may be unpleasant for patients, possibly causes harm, and consumes significant health care resources.^4,10

Two large trials are currently evaluating whether lower phosphate targets have patient-important effects in the dialysis population. HiLo is a pragmatic, cluster-randomized controlled trial that is being conducted at up to 120 hemodialysis units and includes an estimated 4400 patients in the United States.¹¹ Patients who opt in at dialysis units randomized to “low phosphate” will be treated to a target phosphate <5.5 mg/dl, intended to reflect prevailing practice in the United States. Patients who opt into HiLo at units allocated to the “high-phosphate” arm (target phosphate >6.5 mg/dl) will have their phosphate intake derestricted coupled with a de-escalation of phosphate binders. Individual patient consent will be sought to participate in the phosphate control strategy to which one’s dialysis unit was allocated. The trial has a hierarchical primary outcome comprising all-cause mortality and all-cause hospitalizations.

The Pragmatic Randomized Trial of High or Standard Phosphate Targets in End-Stage Kidney Disease (PHOSPHATE) was recently initiated (clinicaltrials.gov NCT03573089). Prevalent recipients of hemodialysis or peritoneal dialysis who are ≥45 years old (≥18 if diabetic) and prescribed at least one phosphate binder will be allocated to either intensive phosphate control (target serum phosphate <4.5 mg/dl using diet, dialysis intensification, and/or binders at the discretion of clinicians) or liberalized control, which entails a discontinuation of phosphate binders and “rescue” reduction if serum phosphate exceeds 7.5 mg/dl. PHOSPHATE will enroll 3600 patients in Australia, New Zealand, Canada, Brazil, and the United Kingdom. This open-label, event-driven trial is designed to detect a 15% relative reduction in the time to the primary composite outcome of cardiovascular mortality and nonfatal cardiovascular events.

The EPISODE trial, as well as the launch of HiLo and PHOSPHATE, reflects the willingness of the nephrology community to finally challenge the long-standing practice of serum phosphate normalization. Observational data, expert opinion, and adherence to tradition cannot be accepted as the basis for the most common treatments we offer our patients. Whether intensive phosphate lowering is beneficial, is harmful, or has no effect on outcomes like death and hospitalization, it is disruptive to our patients’ lives by virtue of the tablet burden, dietary restrictions, and costs. Given this, we must strive to ensure that the highest-quality evidence informs the next iteration of clinical practice guidelines. Our patients deserve no less.

Disclosures

R. Wald and M.W. Walsh are members of the PHOSPHATE Global Management Committee and Co-Chairs of PHOSPHATE-Canada. R. Wald reports research funding from Baxter; reports being a scientific advisor or membership as Editorial Board member of CJASN, Kidney Medicine, Kidney360; and other interests/relationships as a contributor for UpToDate.

Funding

R. Wald and M.W. Walsh are supported by Canadian Institutes of Health Research grant PJT-162095. M.W. Walsh is supported by McMaster Department of Medicine Clive Kearon Mid-Career Award.