Binder Blunder in CKD

Persons living with CKD are at high risk for cardiovascular disease (CVD) and mortality. Although mechanisms remain uncertain, alterations in mineral metabolism represent potential causal intermediates. Reductions in GFR are associated with altered phosphate homeostasis, and even modest elevations in serum phosphate concentrations are associated with arterial calcification, stiffness, and CVD. Higher phosphate concentrations also stimulate upregulation of counter-regulatory hormones, including parathyroid hormone (PTH) and fibroblast growth factor 23 (FGF23), which may have their own adverse effects on the vasculature. In the United States, intestinal phosphate binders are approved by the Food and Drug Administration (FDA) to lower phosphate concentrations in patients with ESKD, but to date, none are approved for this indication in CKD. However, because alterations in mineral metabolism are evident earlier in the CKD spectrum and because many persons with CKD die prior to progressing to ESKD, there has been a consistent push by researchers, consensus statements, and international guideline committees to suggest use of binders in predialysis CKD.^1,2 Data presented by the IMpact of Phosphate REduction on Vascular End-points in CKD (IMPROVE-CKD) investigators in this issue of JASN suggest that such recommendations should be re-evaluated, as these new data provide the most convincing evidence to date suggesting minimal efficacy of binders in CKD.³

Over the past decade, an increasing number of clinical trials have tested the efficacy of binders in CKD, largely with disappointing results. Initial trials suffered from small sample sizes and single-center experiences. Most provided null results, and a few found very modest reductions in serum phosphate when prescribed in high doses (typically <0.3 mg/dl).⁴ More recently, several larger, multicenter trials have addressed this question. In 2019, we published our experience in a double-blind, randomized trial from seven United States centers, where we evaluated the effects of lanthanum carbonate 1000 mg three times daily on changes in serum phosphate and FGF23 concentrations among 205 participants with eGFR of 20–45 ml/min per 1.73 m². Participants were treated for 12 months. We found no effect on serum phosphate concentrations. The primary analyses were also null for changes in FGF23, albeit prespecified secondary analyses suggested a modest reduction. Unfortunately, this came at the cost of high pill burden and substantial gastrointestinal side effects, which made it difficult for participants to comply with the study medications, despite constant coaching and interaction with our study team. Notwithstanding, the results remained similar in participants who remained on treatment.⁵ In this issue of JASN, our prior experience is corroborated and substantially extended by the IMPROVE-CKD investigators.³ Their carefully designed and well-conducted multinational, double-blind, randomized trial evaluated 278 participants with CKD and eGFR of 15–44 ml/min per 1.73 m². A dose titration protocol was used, so most participants took between three and six lanthanum tablets daily throughout the trial and were treated for nearly 2 years.³ The IMPROVE-CKD trial is therefore the largest trial, with the longest follow-up, of any to date conducted in CKD. The primary end point of IMPROVE-CKD was change in arterial stiffness measured by pulse wave velocity (PWV). The trial did not demonstrate significant improvements in the PWV end point. This finding is unlikely to be due to statistical power as the point estimate for PWV was slightly (not significantly) worse in the lanthanum versus placebo group by the end of the trial. Unfortunately, similar to prior trials, there was also no evidence that lanthanum significantly lowered serum phosphate, FGF23, PTH, arterial calcification, or trajectory of eGFR during follow-up compared with placebo. Results were similar in the subset with higher phosphate concentrations and more advanced CKD at baseline, corroborating findings in other trials.⁵ Long-term adherence of the IMPROVE-CKD participants to the binder regimen can be questioned. However, it seems implausible that if participants could not adhere to binders in a carefully conducted clinical trials with frequent follow-up and contact with study staff all designed to promote adherence, that binders would ultimately prove efficacious when prescribed in routine clinical practice.

Prescription of binders to patients with CKD is not without costs. Binders have relatively low efficacy for phosphate lowering and therefore induce high pill burden. Pill burden is inversely related to medication adherence and self-reported quality of life.⁶ Many patients do not understand or remember the indications for one medication versus another. Thus, high pill burden induced by binders may lead to lower adherence to other medications with proven efficacy, such as angiotensin converting enzyme inhibitors or sodium-glucose cotransporter 2 inhibitors. Many binder formulations lead to gastrointestinal side effects. Calcium-based binders are generally less expensive, and they are used widely in CKD but seem to induce vascular calcification, an effect that is well established in ESKD but is also evident in CKD.⁴ Finally, there is considerable financial cost, with recent studies estimating $1.5 billion spent annually on binders in the United States.⁷

As new data have come forward, potential treatment targets for binder use in CKD have become a moving goalpost. The 2017 Kidney Disease Improving Global Outcomes guidelines recommend “lowering elevated serum phosphate towards the normal range in CKD stage 3-5D” and suggest a combination of nutrition interventions and binders to achieve this goal.² However, the discovery that PTH and FGF23 concentrations are elevated earlier in the course of CKD has led some to advocate for earlier use of binders, even when serum phosphate concentrations remain normal.⁸ Some have suggested targeting 24-hour urine phosphate (a marker of intestinal absorption) rather than serum phosphate in an effort to stave off subsequent increases in serum phosphate, FGF23, or PTH that may manifest as CKD progresses.^1,2 Unfortunately, in CKD, 24-hour urine phosphate levels are not at all correlated with serum phosphate concentrations,⁹ demonstrating that dietary phosphate intake is but one of many factors that influence serum phosphate concentrations in CKD, and it is likely to be a minor player. Moreover, studies evaluating relationships of 24-hour urine phosphate with CVD and all-cause mortality have been null,⁹ and interventions that markedly lower 24-hour urine phosphate do not meaningfully translate to lower serum phosphate concentrations or to improved clinical outcomes.¹⁰ Overall, data supporting use of binders to target measures other than serum phosphate in CKD are thin, at best.

There are important questions that remain unresolved. Emerging studies evaluating new formulations of phosphate binders, particularly those that incorporate iron, may have greater potency.¹¹ Persons with stage 5 CKD approaching dialysis often become overtly hyperphosphatemic, have high PTH and FGF23 concentrations, and are more likely to require binders when they initiate dialysis. The benefits of using binders shortly before initiating dialysis may well be advantageous and deserve further study.¹¹ Novel approaches to phosphate lowering have recently been discovered, and these drugs may work in concert with binders.¹² For now, however, these all remain hypotheses.

In summary, IMPROVE-CKD provides a relatively large, multicenter, double-blind, randomized trial with nearly 2 years of follow-up, and it failed to demonstrate that binders improved surrogate markers of vascular disease, lowered serum phosphate, or improved other markers of mineral metabolism in predialysis CKD. Binders are not FDA approved for phosphate lowering in CKD in the United States, and data evaluating effects of binders on FGF23, PTH, and urine phosphate have not been consistent. Binders can induce high pill burden, have bothersome side effects, and add considerable financial costs. Thus, until proven to be efficacious, binders should not be routinely used in predialysis CKD for the purposes of phosphate lowering, improvements in other mineral metabolism parameters, or CVD prevention.

Disclosures

The author has nothing to disclose.

Funding

J.H. Ix is supported by a contract from the National Institute of Diabetes and Digestive and Kidney Diseases (U01 DK097093).