Trials Bolster Hopes New K-Binding Agents Will Prevent, Treat HF-Related Hyperkalemia

September 17, 2014

LAS VEGAS, NV — Two polymer-based, nonsystemic oral agents that sequester potassium (K) in the gastrointestinal tract may be a stride closer to realizing hopes they can prevent or treat hyperkalemia in patients with heart failure. Subgroup analyses of two randomized studies suggest they can lower serum potassium levels without inducing hypokalemia and help maintain potassium levels in the normal range in patients with heart failure taking guidelines-recommended renin-angiotensin-aldosterone (RAAS)-inhibiting meds.

Hyperkalemia is a frequent concern in heart failure, often complicated by chronic kidney disease (CKD), that "is a barrier to full implementation" of RAAS-inhibiting agents such as ACE inhibitors, angiotensin-receptor blockers (ARBs), and especially aldosterone antagonists like spironolactone, observed Dr Mohamed A El-Shahawy (Academic Medical Research Institute, Los Angeles, CA) here at the Heart Failure Society of America (HFSA) 2014 Scientific Meeting .

If the agents become part of hyperkalemia management, El-Shahawy said when presenting the study on the potassium-absorbing zirconium silicate compound ZS-9 (ZS Pharma), they could allow RAAS therapy when before it couldn't be tolerated or intensification of RAAS therapy that was tolerated only at suboptimal levels. The agents "may also allow for fewer dietary restrictions, with improvement in quality of life for many of our patients."

According to Dr Bertram Pitt (University of Michigan School of Medicine, Ann Arbor), who presented data on another proprietary ion-exchanging polymer, patiromer (Relypsa), the availability of "new and safe therapies to treat hyperkalemia and to maintain patients on RAAS inhibitors opens the possibility of further reduction in cardiovascular events and reduction in healthcare costs in these very high-risk individuals."

Both studies, as reported, had limited numbers of heart-failure patients and follow-up times measured in weeks, and neither had clinical end points, so their findings say little about how the agents might perform long term in practice or whether they can contribute to gains in clinical outcomes.

Speaking as discussant following Pitt's presentation, Dr Mark H Drazner (University of Texas Southwestern, Dallas) pointed out that experience with these agents in heart failure is still limited. In the case of patiromer, "There's a relatively small number, about 200 heart-failure patients, who currently have been treated with this agent. No NYHA class 4 patients have been enrolled, and [aldosterone blockers] have not been well studied in hyperkalemic patients," he said. The seminal clinical heart-failure trials for spironolactone and eplerenone—including RALES, EPHESUS , EMPHASIS-HF , and TOPCAT —excluded patients with hyperkalemia for safety's sake.

"So the fact of the matter is that we really don't know if patients who have potassium over 5 [mEq/L] benefit from blockade of their aldosterone receptor," Drazner said. That could question the extent to which these new agents might be useful in allowing or intensifying heart-failure meds.

Patiromer "Pivotal Phase 3 Trial"

The prespecified patiromer heart-failure subgroup analysis was part of a broader safety and efficacy study of 243 patients with hyperkalemia with CKD who were receiving at least one RAAS-inhibiting agent.

In it, according to Pitt, serum K levels dropped off sharply in the hyperkalemic patients, reduced their risk of recurrent hyperkalemia, and allowed more patients to avoid discontinuation of their RAAS-inhibiting medications due to intolerance.

In the initial single-blind phase of the study, 102 patients with heart failure and 141 without HF took patiromer at two different dosage levels depending on their initial serum K levels. Over the course of four weeks, there were "fairly rapid and sustained reductions in serum potassium both in the patients with [a drop of 1.06 mEq/L, p<0.001] and without [0.98-mEq/L decrease, p<0.001] heart failure," he said. Three-fourths of patients in both groups were free of hyperkalemia at four weeks.

At least one adverse event was seen in 41% and 51% patients respectively, the most common being "mild to moderate constipation" in 11% of both groups; 3% in both groups experienced hypokalemia.

In the second phase of the study, 107 patients from the first phase who entered with serum K >5.5 mEq/L to <6.5 mEq/L and went to 3.8 to <5.5 mEq/L by week 4 and who were still taking patiromer and their RAAS inhibitors were randomized to continue their regimen (n=55) or to switch from patiromer to placebo (n=52) and followed for eight weeks. Together, the two arms included 49 patients with heart failure and 58 without heart failure.

Switching patiromer for placebo led to a significant jump in serum K compared with maintenance of the polymer agent in both the heart-failure patients and the non-HF patients, according to Pitt. Also regardless of HF status, patients on patiromer were significantly less likely to have recurrent hyperkalemia (defined here as serum K >5.5 mEq/L).

Four-Week Median Absolute Change in Serum K vs Baseline in Patients Randomized to Placebo or Patiromer, by HF Status

End Points Placebo Patiromer
Heart-Failure Group n=22 n=27
Serum K, 4-wk change (mmol/L) +0.74 +0.10*
Recurrent hyperkalemia (%) 52 8*
>1 adverse event (%) 64 56
No-Heart-Failure Group n=30 n=28
Serum K, 4-wk change (mmol/L) +0.78 -0.05*
Recurrent hyperkalemia (%) 66 23*
>1 adverse event (%) 40 39
*p<0.001 vs placebo group

Patients were followed for time to RAAS-inhibitor withdrawal, which was generally due to intolerance. Among patients with heart failure, about half of those in the placebo group compared with virtually none of those taking patiromer stopped taking their meds by eight weeks.

Patiromer, Pitt said, "was relatively well maintained and safe," with no important safety differences between the actively treated and control patients. There were few adverse events, including one death in the placebo group and none in the patiromer group. The "minimal GI side effects," he said, included a few manageable cases of diarrhea and nausea. "And there was some transient hypokalemia, which was easily corrected, both in patients with and without heart failure."

As discussant for the patiromer-trial presentation, Drazner was circumspect about the new potassium binders and their promise for HF-related hyperkalemia. In considering their use as an alternative to established measures for lowering serum potassium—low potassium diet, adjustments to loop diuretic dosing, and avoidance of K-raising medications and K-containing table-salt substitutes—the bottom line is, "Which strategy leads to improved outcomes?"

But there are few data on that for patiromer and other new potassium binders. "Whether patiromer improves hard clinical outcomes in heart-failure patients by allowing RAAS blockade or augmentation of RAAS blockade in those who otherwise would not tolerate it due to hyperkalemia remains unanswered," he said.

ZS-9 Trial, Heart-Failure Subgroup

The phase 3 trial tested ZS-9 at four dosage levels, acutely in all patients for the first 48 hours and then with randomization against placebo, in 753 patients with hyperkalemia. In both phases, serum K levels quickly dropped with use of the polymer agent in a dose-dependent fashion without any patients developing significant hypokalemia and with a side-effect profile similar to what was seen with placebo, said El-Shahawy.

Randomized primarily in the US and Australia, the cohort included 40% with heart failure, 65% who were taking RAAS-inhibiting agents, and 29% with heart failure who were also taking the drugs. Patients without heart failure had other disorders associated with hyperkalemia, including CKD and diabetes.

RAAS agents in the total population predominantly included ACE inhibitors and ARBs, but also aldosterone antagonists in 6% of patients.

Partial results of the overall trial, without the analysis of effects in heart failure, were reported recently at the American Heart Association (AHA) High Blood Pressure Research Scientific Sessions 2014 and covered at the time by heartwire .

In the study's initial phase, the primary efficacy end point of change in serum K over 48 hours showed a highly significant effect for ZS-9 at the three highest three-times-daily dosage levels; there were no significant changes in the placebo group, and ZS-9 at the lowest dosage (1.25 g) performed a lot like placebo.

48-Hour Absolute Difference in Serum K vs Placebo (Primary Efficacy End Point) at the Three Highest-Tested ZS-9 Dosages and Placebo

End points 2.5 g 5 g 10 g*
Overall Cohort n=141 n=157 n=143
Serum K, 48-h change (mmol/L) -0.46 -0.54 -0.73
Patients with HF n=54 n=64 n=59
Serum K, 48-h change (mmol/L) -0.46 -0.52 -0.78
Patients with HF on RAAS inhibitors n=38 n=45 n=46
Serum K, 48-h change (mmol/L) -0.51 -0.53 -0.80
*p<0.0001 for all three differences in 10 g ZS-9 vs placebo and in 5 g vs placebo just in the overall group; all other differences were significant at higher p values
RAAS=renin-angiotensin-aldosterone system

In the study's second phase, extending another 12 days, 542 patients from phase 1 who had achieved normokalemia on ZS-9 were randomized to receive the drug (at their dosage from phase 1 but only once daily) or placebo while their RAAS-inhibiting meds were maintained. El-Shahawy's presentation focused on the 132 patients who were randomized to 5-g ZS-9 vs placebo and the 124 randomized to 10-g ZS-9 vs placebo.

Patients maintained normokalemia throughout the 12 days at both the 5-g and 10-g ZS-9 levels, with mean serum K finishing at about 4.7 mEq/L (vs about 5 mEq/L for placebo, p=0.0075) and 4.5 mEq/L (vs about 5 mEq/L for placebo, p<0.0001), respectively, El-Shahawy reported.

The agent's adverse-event profile was "very favorable" compared with placebo, he said. At the 5-g level, the rate of any adverse event was 22% for the agent and 24% for placebo; the rates of adverse GI effects were 8% and 7%, respectively. At the 10-mg level, the rates were 33% and 25%, respectively, for any adverse events and 5% and 0%, respectively, for GI events.

Going beyond the current study, El-Shahawy said, "Long-term experience with ZS-9 may allow the clinician to continue RAAS blockade, with its proven class 1A [guideline] indication, in some target populations—delaying the progression of CKD and improving survival in patients with heart failure."

El-Shahawy discloses receiving research grants from ZS Pharma, Amgen, GlaxoSmithKline, Celgene, and Abbvie, and participating in advisory boards for ZS Pharma. Pitt discloses equity interest in Relypsa and SZ Pharmaceuticals and having consulted for Pfizer, Bayer, Lilly, Relypsa, Sarfez Pharmaceuticals, and SZ Pharmaceuticals. Drazner reports having nothing to disclose. 


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.