Dapagliflozin use in tolvaptan-treated patients with ADPKD: exploring renal outcomes in a retrospective study

Abstract

Background: Despite the widespread use of sodium-glucose cotransporter 2 inhibitors (SGLT2i) in the management of chronic kidney disease, their role in autosomal dominant polycystic kidney disease (ADPKD) remains unclear.

Methods: This observational study evaluated the efficacy of the SGLT2i dapagliflozin in patients with ADPKD receiving tolvaptan. The primary outcome was the chronic estimated glomerular filtration rate (eGFR) slope, modeled using a multivariable linear mixed-effect model; a within-group analysis was also performed using an interrupted time series approach.

Results: A total of 48 patients receiving tolvaptan were analyzed (24 patients in the control group vs 24 patients in the dapagliflozin group). The mean follow-up duration was 649 ± 363 days across all patients. The chronic eGFR slope was -2.30 [95% confidence interval (CI) -3.47, -1.13] in the control group and -1.72 (95% CI -3.48, -0.03) mL/min/1.73 m² per year in the dapagliflozin group (P = .595). In within-group analysis using an interrupted time series approach, the chronic eGFR slope changed from -2.34 (95% CI -3.39, -1.30) to -1.14 (95% CI -2.68, 0.40) mL/min/1.73 m² per year following dapagliflozin initiation (P = .191). No serious adverse events were observed during the follow-up period.

Conclusions: Although no statistically significant differences were observed, both between- and within-group analyses showed a numerically slower decline in eGFR with dapagliflozin. Importantly, no evidence of harm was observed. These findings may contribute to ongoing discussions regarding the potential role of SGLT2i in ADPKD.

Keywords: ADPKD; SGLT2 inhibitor; dapagliflozin; polycystic kidney disease; tolvaptan.

GRAPHICAL ABSTRACT

Figure 1:

Study flow chart.

Figure 2:

Initial eGFR dip during the first 2 months. Paired eGFR values at baseline and at the initial dip are shown. Due to missing data, 22 cases in the control group and 17 cases in the dapagliflozin group were included. The initial dip was defined as the lowest eGFR value recorded between 2 and 8 weeks after time zero. ^**P < .01, ^***P < .001 (Wilcoxon signed-rank test).

Figure 3:

Between-group analysis of eGFR slope using LMM (control vs dapagliflozin). The chronic eGFR slope was estimated using the linear mixed-effects model with all available eGFR data throughout the follow-up period (A, entire follow-up period), as well as with data truncated to the mean follow-up period of the dapagliflozin group (B, restricted follow-up period).

Figure 4:

Within-group analysis of eGFR slope using ITS with LMM (pre-post dapagliflozin). In the dapagliflozin group, the eGFR slope was estimated using an interrupted time series approach with a linear mixed-effects model. The model incorporated data from 2 years before time zero (before dapagliflozin) and from 2 months after time zero to the end of follow-up (after dapagliflozin). The dotted line indicates the counterfactual eGFR slope.

Figure 5:

One-year changes in Hb, Hct and UA. One-year changes in Hb, Hct and UA in each group are shown for each group. ^**P < .01 (Wilcoxon rank sum test).