BACKGROUND AND AIMS Sodium zirconium cyclosilicate (SZC, LOKELMA®) is an orally administered suspension for treating hyperkalemia that permits selective, monovalent cation exchange and has a high exchange capacity for potassium cations in the gut leading to a net reduction in systemic potassium. The goals of the analysis were to describe the population pharmacodynamic response of serum potassium concentration (S-K) in hyperkalemic subjects treated with SZC; specifically, the longitudinal dose response of SZC for S-K in correction (0–48 h) and maintenance (>48 h) phases of treatment, and the differences in S-K response among populations identified by their intrinsic and extrinsic factors. METHOD A population pharmacodynamic model was developed using serum potassium, dose timing/amount, and patient demographic and medical history data pooled from 7 phase 2 and 3 clinical studies consisting of 2369 subjects with hyperkalemia (screening S K >5 mmol/L) with a total of 25 764 S-K observations. A virtual pharmacokinetic-pharmacodynamic model (K-PD) model was used to mimic a drug exposure without any measurements (none were available, drug was not absorbed). S-K over time was described by an indirect response model where potassium enters/leaves a compartment corresponding to serum potassium measurements. The impact of SZC exposure was modeled as an inhibition of the potassium production rate, consistent with the SZC mechanism of action to bind potassium and lower its absorption and appearance in circulation. RESULTS Diagnostics indicated good agreement between the observed data and the model, for mean S-K and change from baseline, and overall qualified the model as appropriate for predicting S-K changes following SZC dosing. The sigmoid exposure response on Kin was characterized by a high value for the EC50 (32.8 g) and a hill coefficient of 1.36. Not surprisingly, within the range of doses studied (0–15 g), the dose response appeared nearly linear. The placebo adjusted S-K change from baseline for QD maintenance treatment was 0.37 (95% CI –0.42 to –0.32) for 5 g, –0.75 (95% CI –0.82 to –0.65) for 10 g, and –1.06 (95% CI –1.17 to –0.89) for 15 g. Greater treatment response was associated with high S-K at baseline, more advanced age, lower body weight and lower eGFR. Furthermore, greater treatment response was also predicted for Black or African American and Asian patients, compared with Caucasian individuals, but they remained unworthy of dose adjustment. Lastly, dose-response simulations show no clinically meaningful difference between maintenance regimens of 5 g QOD and 2.5 g QD in the studied patient groups. CONCLUSION The population K-PD model of SZC adequately described the changes in serum potassium concentration during correction and maintenance phase dosing in the seven clinical studies. Greater treatment response was associated with high S-K at baseline, more advanced age, lower body weight, lower eGFR, Black or African American race and Asian race; however, the impact of each covariate was modest. The overall results of this analysis support the conclusion that no dose adjustment is needed for SZC based on any of the tested covariates.
