Skip to content
Medical InsightsKoreaMedical Insights Korea

Overactive Bladder

A new bladder drug has a 99.5 percent probability of working. The probability it cuts urination by more than once a day is 14.2 percent

A Bayesian synthesis of three phase III trials, one of them the Korean bridging study, separates two questions that usually get answered as one: does it work, and by how much.

Reported by

Reporter

Sources
2 cited
Published
Reading time
6 minutes

In short

In a Bayesian cross-trial synthesis of three randomised, double-blind, placebo-controlled phase III trials of vibegron monotherapy for overactive bladder — the Japanese phase III trial, the Korean bridging study at 50 mg and the EMPOWUR trial at 75 mg — the pooled placebo-adjusted effects at week 12 favoured vibegron on all four voiding-diary endpoints with posterior probabilities of benefit of 99.5 percent for micturitions per day, 99.3 percent for urgency episodes, 98.0 percent for urgency urinary incontinence episodes and above 99.9 percent for voided volume, while the probability that the effect exceeded a one-per-day reduction was only 14.2 percent for micturitions and 12.7 percent for urgency episodes.

Key takeaways

  • Three placebo-controlled phase III trials of vibegron monotherapy were pooled: the Japanese phase III trial, the Korean bridging study at 50 mg and the EMPOWUR trial at 75 mg.
  • Pooled effects at week 12 favoured vibegron on all four endpoints: micturitions per day -0.76, urgency episodes -0.69, urgency incontinence episodes -0.44, and voided volume +24.09 mL.
  • Posterior probabilities of any benefit were 99.5, 99.3, 98.0 and above 99.9 percent respectively.
  • The probability of exceeding a reduction of one micturition per day was 14.2 percent; for a reduction of half a micturition it was 90.3 percent.
  • For urgency episodes the probability of exceeding a one-per-day reduction was 12.7 percent, against 82.8 percent for half a episode per day.
  • For voided volume, the probability of exceeding 20 mL was 88.4 percent and of exceeding 30 mL was 5.8 percent.
  • The decision thresholds were defined by the analysis authors rather than taken from an external standard.

Almost every drug trial reports whether an effect is real. Very few report how big it probably is, in a form a patient could use. This analysis does both for vibegron, and the two answers are very far apart.

Two probabilities, one drug

99.5% vs 14.2%

probability that vibegron reduces daily urination at all, against the probability it reduces it by more than once a dayYamamoto T et al., Lower Urinary Tract Symptoms, 2026

The pooled effect on micturitions was -0.76 per day, with a credible interval from -1.33 to -0.28. That interval sits entirely below zero, which is why the probability of benefit is 99.5 percent. It also sits mostly above -1, which is why the probability of a benefit larger than one urination a day is 14.2 percent.

Both statements describe the same result. A patient told the drug works with 99.5 percent probability and a patient told it will probably save them roughly three-quarters of a trip to the toilet per day have been told the truth, and will make different decisions.

The same shape across every endpoint

Urgency episodes fell by 0.69 a day with a 99.3 percent probability of benefit, and a 12.7 percent probability of exceeding one episode a day. Urgency incontinence episodes fell by 0.44 with a 98.0 percent probability of benefit. Voided volume rose by 24.09 mL with better than 99.9 percent certainty of some increase, an 88.4 percent probability of exceeding 20 mL, and a 5.8 percent probability of exceeding 30 mL.

Why the Korean trial matters to the pooling

One of the three trials is the Korean bridging study at the 50 mg dose, and the EMPOWUR trial used 75 mg. A pooled estimate across different doses in different populations is doing real work to produce a single number, and the random-effects model is what accommodates that — it estimates between-trial heterogeneity rather than assuming it away. The cost is that the pooled figure is not the expected effect of any one dose in any one population.

Set against the real-world Korean mirabegron data, where 43.4 percent of 4,009 patients reached a three-point symptom-score improvement at six months, the two evidence types are answering different questions in different currencies. Neither supports the idea that this drug class transforms the condition. Both support the idea that it helps, modestly and reliably.

Sources

  1. 1

    Yamamoto T, Yoshida S, Yajima S, Fukushima H, Yasuda Y, Waseda Y, Sato H, Hirakawa A, Honda H, Komai Y, Matsuo T, Wada N, Masuda H, Fujii Y

    Efficacy of Vibegron in Overactive Bladder: A Bayesian Cross-Trial Synthesis of Benefit Probabilities and Decision Thresholds

    Lower Urinary Tract Symptoms 2026;18(5):e70095

    doi:10.1111/luts.70095 · PMID:42665534

    https://doi.org/10.1111/luts.70095
  2. 2

    Park JS, Kim J, Park MH, Jang WS

    Effectiveness and Safety of Mirabek Sustained-release 50 mg Tablets in Korean Patients with Overactive Bladder: A Multicenter Observational Study at Tertiary General Hospitals

    European Urology Open Science 2025;82:65-72

    doi:10.1016/j.euros.2025.10.004 · PMID:41190053

    https://doi.org/10.1016/j.euros.2025.10.004