Article Text

PTU-187 Sample size calculation in confirmatory trials in irritable bowel syndrome with constipation: experience with linaclotide (constella®)
  1. M Falques1,
  2. C Díaz1,
  3. D Vilardell1,
  4. J Fortea1,
  5. JM Johnston2
  1. 1Almirall S. A., Barcelona, Spain
  2. 2Ironwood Pharmaceuticals, Cambridge, USA


Introduction Sample size (SS) calculation with sufficient power to detect significant and clinically meaningful differences vs placebo is necessary to ensure the validity of confirmatory Phase 3 pivotal trials in irritable bowel syndrome with constipation (IBS-C).

Method Using data from a pilot Phase 2b study of linaclotide (Constella®; LIN) in IBS-C (Study 202) and the pivotal Phase 3 LIN trials (Trial 302; Trial 31), we compared 2 statistical assumptions for the Phase 3 SS calculation for the European Medicines Agency (EMA) co-primary endpoints: 1) minimally important clinical differences (MICD) and 2) replication of Phase 2b 290 µg/day results. Protocol design, patient population and endpoints of the trials were also compared.

Results There were no relevant differences in protocol design or patient population in the Phase 2b/3 trials. For endpoint assessments, only the range of the pain/discomfort scale differed (Phase 2b: range 1–5; Phase 3: range 0–10). For both EMA endpoints, Phase 2b data showed a higher placebo effect and greater improvement vs placebo compared with Phase 3 data (both not statistically significant; Table 1). The SS for the Phase 3 trials was estimated to have ≥95% global power to detect an MICD of 15% in both endpoints. Based on replication of Phase 2b results, an SS of 134 patients per treatment arm would have been sufficient to achieve ≥95% global power to detect an effect size comparable to Phase 2b. However, this sample size would have only 46% and 67% overall power to detect the actual observed results of Trials 31 and 302, respectively, and even less overall power (13% and 47%, respectively) when the MICD is reduced to 10%.

Abstract PTU-187 Table 1

Responder rates in Phase 2b/3 studies of LIN

Conclusion In the LIN IBS-C clinical programme, Phase 2b results were not a good predictor of Phase 3 outcome, even without major differences in study design. Caution is needed when calculating SS for Phase 3 IBS-C trials to ensure sufficient power to detect clinically relevant differences vs placebo.

Disclosure of interest M. Falques Employee of: Almirall S. A., C. Díaz Employee of: Almirall S. A., D. Vilardell Employee of: Almirall S. A., J. Fortea Employee of: Almirall S. A., J. Johnston Employee of: Ironwood Pharmaceuticals.

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.