New Therapies for Hyperkalemia

Silvia J. Leon; Oksana Harasemiw; Navdeep Tangri


Curr Opin Nephrol Hypertens. 2019;28(3):238-244. 

In This Article

Sodium Zirconium Cyclosilicate

Sodium zirconium cyclosilicate (formerly ZS-9), is an orally administered, insoluble, nonabsorbed cation exchange crystal that exchanges potassium for sodium and hydrogen in the intestinal tract.[26] ZS-9 forms a highly selective ion trap for potassium and has nine-times potassium-biding capacity compared with SPS in in-vitro studies.[26] The recommended dose is 10 g administered three times a day (for up to 48 h) for acute hyperkalemia, and 10 g once a day for chronic management.[27] The safety and efficacy of ZS-9 have been evaluated in three randomized clinical trials (Table 3).

A phase II study evaluated the safety and tolerability of ZS-9 in patients with Stage 3 CKD and hyperkalemia (serum potassium >5.0 mEq/l). At day 2 (hour 38), 41.7% of patients receiving 10 g of ZS-9 had a decrease more than 1.0 mEq/l in serum potassium, compared with 3.4% of patients receiving the placebo. After the last dose (120 h), serum potassium levels remained significantly lower amongst patients who had received ZS-9 compared with patients who received the placebo.[28] In the subgroup of patients concomitantly receiving RAAS inhibitors, those in the ZS-9 study arm (10 g) had a mean reduction of −1.05 ± 0.47 mEq/l, compared with −0.25 ± 0.46 mEq/l with placebo (P < 0.001) on day 2 (hour 14).

Results from a phase III study with 754 patients reported that during the initial phase (at 48 h), the patients receiving 10 g of ZS-9 had a mean decrease in serum potassium of 0.73 mmol/l (95% CI −0.82 to −0.65) compared with 0.25 mmol/l (95% CI −0.32–0.19) in the placebo group (P < 0.0001).[29] Potassium reduction was observed within the first hour (P = 0.002). Normalization of serum potassium (3.5–4.9 mmol/l), was reported at both 5 and 10 g doses for all subgroups. During the maintenance phase (n = 543), 5 and 10 g of ZS-9 were significantly superior to placebo in maintaining normokalemia (P = 0.008 and <0.001).

The HARMONIZE phase III study,[30] evaluated the efficacy and safety of ZS-9 in patients with hyperkalemia. Patients recruited followed an open-label phase receiving 10 g of ZS-9 three times a day for 48 h. Those who achieved normokalemia were further randomized to placebo or ZS-9 once a day in different dose (5, 10 or 15 g) for 28 days. One hour after the first ZS-9 dose (10 g), a significant mean reduction in serum potassium was noted with −0.2 mEq/l (95% CI −0.3 to −0.2, P < 0.001) compared with baseline. At 48 h, the mean change in potassium was −1.1 mEq/l (95% CI −1.1 to −1.0, P < 0.001).[30] During the randomization phase, the mean reduction of serum potassium was significant with lower potassium achieved at higher ZS-9 doses (P < 0.001 for each comparison placebo-ZS-9 dose). The potassium mean during days 8–29 were 4.8, 4.5, 4.4 and 5.1 mEq/l for ZS-9 5, 10, 15 g and placebo respectively (P < 0.001 for all levels). At the end of the randomized phase, normokalemia (3.5–5 mEq/l) was reported in 71%. In this study, 66% patients had CKD, 36% had heart failure and 66% had diabetes. A substantial proportion were also treated with RAAS inhibitors (70%).[30]

A sub-group analysis of the HARMONIZE study focusing on the use of ZS-9 in patients with heart failure (n = 94) showed good potential for use of ZS-9 in this population.[31] During the open-label phase, mean serum potassium was normalized to 4.4 mmol/l (95% CI 4.3–4.5) from baseline after 48 h of treatment. During the randomized phase, patients receiving different doses of ZS-9 (5, 10 and 15 g) achieved mean potassium concentrations of 4.7 mmol/l (95% CI 4.5–4.9), 4.5 mmol/l (95% CI 4.3–4.6) and 4.4 mmol/l (95% CI 4.2–4.5), respectively, compared with 5.2 mmol/l (95% CI 5.0–5.4) placebo (P < 0.001 all comparisons). Efficacy findings were consistent among patients receiving treatment with RAAS inhibitors. During the maintenance phase, edema was the most commonly reported adverse event, observed in 3.8% on placebo patients compared with 13.1% in the ZS-9 arm.

ZS-9 was well tolerated and appears to be safe. Similarly to patiromer, the most common adverse events were gastrointestinal-related (Table 2). Although ZS-9 was associated with fewer gastrointestinal events than patiromer, there were more unique adverse events reported, such as edema and urinary tract infections, as well as QTc-interval prolongation secondary to lowering of serum potassium levels.