Tolvaptan for Autosomal Dominant Polycystic Kidney Disease: Real-World Insights from a Propensity Score Weighted...

Tolvaptan slows disease progression of autosomal dominant polycystic kidney disease (ADPKD) in clinical trials, but real-world effectiveness remains uncertain. We evaluated the association between tolvaptan use and kidney function decline in a large tertiary-centre cohort.

This retrospective cohort study evaluated adults with ADPKD attending a specialist polycystic kidney disease clinic. Patients with <1 year of follow-up or <3 estimated glomerular filtration (eGFR) measurements were excluded. eGFR slopes were estimated using piecewise linear mixed-effects models separating acute (<60 days) and chronic (≥60 days) phases, with the chronic slope defined as the primary outcome. Multivariable mixed-effects models and propensity score overlap weighting were used to adjust for baseline differences in covariates. Complete case analysis was performed as the primary analysis; multiple imputation was used as a sensitivity analysis. Patients receiving <6 months of tolvaptan were excluded from the primary slope analyses but retained for intention-to-treat comparisons.

Of 327 patients, 119 initiated tolvaptan and 208 did not; mean duration of study follow-up was 61.4 months. Tolvaptan-treated patients had a lower baseline eGFR and trended towards a faster pre-study eGFR decline. After full adjustment, tolvaptan was associated with a modest slowing in the chronic eGFR slope of +0.45 mL/min/1.73 m2/year; however, confidence intervals (CIs) were wide (95% CI: -0.37 to +1.18) and therefore compatible with clinically meaningful benefit or no effect. Findings were directionally consistent across complete-case, overlap-weighted, and multiple-imputation analyses with similarly imprecise CIs. A total of 61 (52.1%) patients discontinued therapy within 12 months of initiation, primarily due to aquaretic side effects. In subgroup analyses, tolvaptan patients treated for ≥1 year and tolvaptan responders treated for ≥2 years showed greater magnitudes of benefit with chronic slopes of +0.58 and +0.78 mL/min/1.73 m2/year, respectively, when compared to untreated matched cohorts.

Tolvaptan was not associated with a statistically significant slowing of kidney function decline. Point estimates suggest a modest potential benefit which is attenuated when compared to controlled trials. Treatment discontinuation substantially limited treatment exposure. Further real-world evidence is required to clarify tolvaptan's true effectiveness outside controlled trial settings.