INTRODUCTION AND AIMS: Higher serum phosphate (Pi) and fibroblast growth factor 23 (FGF23) concentrations are associated with cardiovascular disease and death in persons with chronic kidney disease (CKD). We hypothesized that lanthanum carbonate (LC) and nicotinamide (NAM, inhibitor of gastrointestinal sodium phosphate transporter 2b), alone and in combination, would effectively and safely lower serum Pi and FGF23 in non-dialysis CKD.
METHODS: COMBINE is a seven-center, double-blind, placebo-controlled trial that randomized US patients with eGFR 20-45 ml/min/1.73m2 to: NAM (750 mg 2x/d) and LC (1000 mg 3x/d) (“N-L” group), NAM and LC placebo (“N-p”), NAM placebo and active LC (“p-L”), or double placebo (“p-p”) for 12 months. Dual primary endpoints were changes from baseline (Δ) of serum Pi and intact FGF23 (Kainos) concentrations; p<0.0083 was considered statistically significant (6 pairwise comparisons; 0.05/6). Safety and tolerability were secondary endpoints.
RESULTS: Baseline characteristics of the 205 participants were age 68±12 yrs; 62% male; 66% Caucasian; 55% diabetes; eGFR 33±7 ml/min/1.73m2. Baseline serum Pi was 3.7±0.5 mg/dL and FGF23 was 103.6 [IQR 76.8, 145.1] pg/ml. Baseline characteristics were well-balanced across arms. Over 12 months, mean Δ Pi in the active groups were +0.03±0.66, 0.00±0.53, and -0.02±0.60mg/dL (N-L, N-p, and p-L arms, respectively), compared to +0.11±0.59 mg/dL in the p-p arm (p ≥0.38 for all comparisons; Figure 1). Similarly, ΔFGF23 in the 3 active groups vs. p-p did not reach statistical significance (p≥0.037). The ΔFGF23 were + 3.7[-13.7, 27.6], +21.3 [1.8, 45.8], -9.1 [-35.8, 9.7], and +6.0 [-10.6, 34.8] pg/mL in the N-L, N-p, p-L, and p-p groups, respectively (Figure 2). Study drug discontinuations were 42%, 26%, 30%, and 14% in the N-L, N-p, p-L, and p-p arms respectively (p=0.017). Gastrointestinal side effects were the predominant cause for discontinuation. Rates of laboratory- or symptom-related adverse events were similar across arms.
CONCLUSIONS: Randomization to LC, NAM, or dual treatment did not substantially lower serum Pi or FGF23 concentrations in persons with non-dialysis CKD over 12-months. While LC, NAM, and dual treatment appeared safe, GI symptoms were common and limited adherence. Novel strategies to lower serum Pi and FGF23 in non-dialysis CKD are needed.
Comments