Neurogastroenterology & Motility Neurogastroenterol Motil (2015) 27, 805–815
doi: 10.1111/nmo.12553
A randomized, double-blind, placebo-controlled trial to evaluate the efficacy, safety, and tolerability of long-term treatment with prucalopride H. PIESSEVAUX ,* E. CORAZZIARI ,† E. REY ,‡ M. SIMREN ,§ A. WIECHOWSKA-KOZLOWSKA ,¶ R. KERSTENS ,** M. COOLS ,** K. BARRETT ††
& A. LEVINE ‡‡
*Cliniques Universitaires Saint-Luc, Brussels, Belgium †Unit a Operativa Complessa di Gastroenterologia, Universita degli Studi di Roma “La Sapienza”, Rome, Italy ‡Department of Digestive Diseases, Instituto de Investigaci on Sanitaria del Hospital Clınico San Carlos, Hospital Clınico San Carlos, Universidad Complutense de Madrid, Madrid, Spain §University of Gothenburg, Gothenburg, Sweden ¶Niepubliczny Zaklad Opieki Zdrowotnej, Szczecin, Poland **Shire Movetis NV, Turnhout, Belgium ††Shire, Basingstoke, UK ‡‡Shire, Chesterbrook, PA, USA
Key Messages
• In contrast to the results of four previous 12-week trials, this study found no significant improvement in the • • •
proportion of patients achieving three or more spontaneous complete bowel movements (SCBMs) per week over 24 weeks when treated with prucalopride compared with placebo. Despite extensive evaluation, no explanation for these null efficacy results could be found. The aim of this randomized, placebo-controlled trial was to assess the efficacy of 24 weeks of prucalopride treatment in patients with chronic constipation. Patients with chronic constipation who had less than three SCBMs per week were randomized to receive prucalopride 2 mg or placebo daily for 24 weeks. The primary endpoint was the proportion of patients achieving a mean of three or more SCBMs per week over the treatment period. Overall, 361 patients were enrolled in the trial. The proportion of patients achieving the primary endpoint was not statistically different between the prucalopride and placebo groups (25.1% vs 20.7%; p = 0.367). No new safety concerns were identified.
Abstract Background Randomized trials have confirmed the efficacy of prucalopride for the treatment of chronic constipation up to 12 weeks. This study aimed to assess the efficacy of prucalopride over a 24-week period (ClinicalTrials.gov: NCT01424228). Methods Address for Correspondence Dr H. Piessevaux, Cliniques Universitaires Saint-Luc, Avenue Hippocrate 10, 1200 Brussels, Belgium. Tel: +32 2764 2834; fax: + 32 2764 8927; e-mail: [email protected] Received: 16 October 2014 Accepted for publication: 24 February 2015
Adults with chronic constipation and ≤2 spontaneous complete bowel movements (SCBMs)/week were randomized to receive prucalopride 2 mg or placebo daily for 24 weeks. The primary endpoint was the proportion of patients achieving a mean of ≥3 SCBMs/week over the treatment period, assessed using daily e-diaries. Secondary outcomes and safety parameters were assessed throughout the study. Key Results Overall, 361 patients were randomized and received prucalopride or placebo. Baseline characteristics were similar in the prucalopride (N = 181) and placebo (N = 180) groups. Mean age was 48.9 years (standard deviation, 16.0) and most patients were women.
© 2015 The Authors. 805 Neurogastroenterology & Motility published by John Wiley & Sons Ltd. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.
H. Piessevaux et al.
Neurogastroenterology and Motility
The proportion of participants achieving the primary endpoint was not statistically different between the prucalopride and placebo groups (25.1% vs 20.7%; p = 0.367). There was also no statistically significant difference between groups over the first 12-week period (prucalopride, 25.1%; placebo, 20.1%; p = 0.341). There were no statistically significant differences between groups for most secondary endpoints. No new safety concerns were identified. Conclusions & Inferences This trial did not show statistically significant improvements in primary or secondary outcomes with prucalopride compared with placebo over 24 or 12 weeks. This is in contrast to the results of four previous 12-week trials, which demonstrated prucalopride to be significantly more effective than placebo. An extensive evaluation did not provide an explanation for the null efficacy results of this study. Keywords clinical prucalopride.
trial,
constipation,
long-term,
Abbreviations: AE, Adverse event; BM, Bowel movement; CBM, Complete bowel movement; CI, Confidence interval; ECG, Electrocardiogram; GCP, Good clinical practice; HRQoL, Health-related quality of life; ITT, Intent-to-treat; PAC-QOL, Patient Assessment of Constipation-Quality of Life; PAC-SYM, Patient Assessment of Constipation-Symptom; SCBM, Spontaneous complete bowel movement; SD, Standard deviation; TEAE, Treatment-emergent adverse event.
INTRODUCTION Chronic constipation is a common disorder characterized by low stool frequency and gastrointestinal symptoms such as straining, hard stools, bloating, and incomplete evacuation. Initial treatment recommendations include dietary and lifestyle changes, and use of laxatives, which are available both over the counter and on prescription. However, many patients with chronic constipation experience long-term symptoms despite these interventions.1 An Internet survey in the United States indicated that 43% of 557 participants with constipation had experienced their symptoms for more than 4 years,2 and a recent European survey found that 59% of patients with constipation had symptoms for more than 3 years.3 In the latter study, the majority of participants used laxatives, but only 28% were satisfied with their treatment.3 Effective, long-term treatments are therefore needed for patients with chronic constipation. Prucalopride is a 5-hydroxytryptamine receptor 4 agonist that stimulates intestinal motility and is
licensed in the European Union for the symptomatic treatment of chronic constipation in adult women in whom laxatives fail to provide adequate relief. The recommended dose of prucalopride is 2 mg/day; in individuals over the age of 65 years, the starting dose is 1 mg/day.4 Four phase III clinical trials have shown prucalopride to be significantly more effective than placebo at increasing the proportion of patients achieving three or more spontaneous complete bowel movements (SCBMs) per week over 12 weeks of treatment, the primary endpoint of the studies.5–9 These trials also demonstrated that prucalopride was better than placebo at alleviating symptoms of constipation and improving health-related quality of life (HRQoL) during this time. In two long-term, open-label, follow-up studies, patients were enrolled from three of the phase III trials of prucalopride5–7 and reported sustained satisfaction when prucalopride treatment was continued after the end of the double-blind trials.10 Improvements in average Patient Assessment of Constipation-Quality of Life (PAC-QOL) satisfaction scores observed at the end of the 12-week, double-blind trials were maintained during open-label treatment for up to 18 months. While these results are suggestive of long-term efficacy, the limitations of these open-label studies include that the primary efficacy endpoint was evaluated using PAC-QOL, not SCBM frequency, and that there was no placebo comparator group. A randomized, double-blind, placebo-controlled clinical trial was therefore performed to assess further the long-term efficacy and safety of treatment with prucalopride. The primary objective of this study was to assess the efficacy of prucalopride compared with placebo over a 24-week period. The primary endpoint of the study was defined as the proportion of participants with a mean of three or more SCBMs per week evaluated over this 24-week period. Secondary objectives were to evaluate the long-term safety, tolerability, and effect on HRQoL of prucalopride compared with placebo.
METHODS Study population This was a stratified, randomized, parallel-group, double-blind, placebo-controlled, phase IV study (ClinicalTrials.gov Identifier: NCT01424228). It was carried out at 50 sites across Europe (Belgium, Czech Republic, Hungary, Italy, Poland, Romania, Slovakia, Spain, and Sweden) from 6 April 2011 to 19 December 2012. This study was carried out in accordance with local ethical and legal requirements, the International Conference on Harmonization of good clinical practice (GCP)11 and the principles of the Declaration of Helsinki.12 All participants provided written, informed consent before entry into the trial.
806 © 2015 The Authors. Neurogastroenterology & Motility published by John Wiley & Sons Ltd.
Volume 27, Number 6, June 2015
Long-term treatment with prucalopride
Inclusion criteria Men or non-pregnant, non-breastfeeding women aged 18 years and older with chronic constipation were eligible for inclusion. Chronic constipation was defined as two or fewer SCBMs (bowel movements not preceded within 24 h by the use of laxatives or enema that resulted in a feeling of complete evacuation) per week, and one or more of the following for a minimum of 6 months: very hard or hard stools for at least 25% of stools; sensation of incomplete evacuation following at least 25% of stools; and straining at defecation for at least 25% of the time. Exclusion criteria Participants were excluded from the trial if they were considered to have drug-induced constipation, or constipation secondary to causes such as endocrine disorders, metabolic disorders, neurological disorders, or surgery. Additional exclusion criteria included a history of clinically significant (as evaluated by the investigator) cancer or cardiac, vascular, liver, pulmonary, endocrine, neurological or psychiatric disorders, or metabolic disturbances. Participants were also excluded if they had previously used prucalopride. Sample size For this study, a 15% and 30% response rate for the placebo and prucalopride treatment groups, respectively, was assumed based on results from previous large studies of prucalopride in chronic constipation.5–7 The 15% difference in response rates could be detected with 161 patients per treatment arm at a power of 90%. It was assumed that 5% of participants would have insufficient e-diary data to evaluate the primary endpoint; therefore, 340 participants were targeted to be randomized.
Trial design The trial design is shown in Fig. 1. After an initial screening visit, participants underwent a run-in phase of 2 weeks. If patients needed a colonoscopy after screening and/or if they had been using agents that influence bowel habits, the run-in phase could extend to up to 4 weeks. Following the baseline visit, patients returned to the treatment center for assessments at weeks 2, 4, 8, 12, and 24. At baseline, patients were randomized 1 : 1 to receive placebo or prucalopride once daily. They were instructed to take their treatment before breakfast. Randomization was carried out using an interactive web-based/voice-response system (Oracleâ, Vilvoorde, Belgium). Participants younger than 65 years took prucalopride 2 mg once daily. Participants aged 65 years or older started on a daily dose of prucalopride 1 mg, consistent with the labeled dose in Europe for patients of this age. The dose was increased to 2 mg in the case of insufficient response (an average
of fewer than three SCBMs per week in the previous 2 weeks) at weeks 2 or 4. No further dose adjustments were allowed. The use of agents that influence bowel habits, such as prokinetics or anticholinergic drugs, was disallowed during the trial. Laxative use was also disallowed during the trial, except for bisacodyl as a rescue medication (used if a patient had not had a bowel movement for ≥3 consecutive days). Use of rescue medication was not permitted in the 24 h before or the 48 h after the baseline visit.
Assessments Daily e-diaries completed by the patient were used to gather information on intake of study and rescue medication, bowel movement frequency (date and time of each bowel movement) and consistency (using the Bristol Stool Scale13), degree of straining (rated as none, mild, moderate, severe, or very severe), and feeling of complete evacuation after a bowel movement (yes or no). At each visit from screening onwards, patients were asked to score the severity of their stool, abdominal, and rectal symptoms on 5-point scales (from 0 = absent to 4 = very severe) using the validated Patient Assessment of Constipation-Symptom (PAC-SYM) questionnaire.14 Health-related quality of life parameters were assessed using the PAC-QOL questionnaire administered at screening, baseline, and weeks 4, 12, and 24. The PAC-QOL assesses constipation-related HRQoL parameters on four subscales (physical discomfort, psychosocial discomfort, worries and concerns, and satisfaction) that are scored on a 5-point scale (0 = none/not at all, 4 = extremely/all the time).15,16
Adverse events and safety Adverse events (AEs) were recorded from the signing of informed consent at screening to the final patient visit. Treatment-emergent AEs were defined as AEs that occurred after the first administration of the investigational product up to 5 days after the last dose of investigational product. Serious treatment-emergent adverse events (TEAEs) were collected until 30 days after the last dose of investigational product. Vital signs were monitored at every visit. Electrocardiograms (ECGs) were recorded, physical examinations were carried out, and samples for blood and urine tests were collected at screening, baseline, and at the 4- and 24-week visits or at early discontinuation.
Placebo Washout*
Run-in Prucalopride 2 mg†
Visit 1 Visit 2 Visit 3 Visit 4 (Screening) (Baseline) Week 2 Week 4 Week –4 to –2 Day 1
Visit 5 Week 8
Visit 6 Week 12
Visit 7 Week 24
Figure 1 Study design. *Washout if required due to colonoscopy or use of medication affecting bowel habits. †Patients aged ≥65 years started at 1 mg daily. In cases of insufficient response, the dose increased to 2 mg daily at week 2 or 4.
© 2015 The Authors. 807 Neurogastroenterology & Motility published by John Wiley & Sons Ltd.
H. Piessevaux et al.
Neurogastroenterology and Motility
Statistical methods Analysis populations The primary efficacy analysis was carried out in the intent-to-treat (ITT) population, which included all randomized patients who took at least one dose of the trial medication. Twenty-one patients were excluded from the ITT population as the unblinded treatment arm information was inadvertently present in a report sent to several sites due to a programming error. The safety population included all randomized patients who took at least one dose of the trial medication, inclusive of the 21 individuals omitted from the ITT population. The per-protocol population included patients who took their treatment for at least 28 days and excluded those with major protocol violations that could have impacted on efficacy or safety analyses (as determined by the study team prior to database lock). The completers population consisted of patients in the ITT population who completed the study. A site audit revealed two critical findings that were categorized as serious breaches of GCP at one site participating in the study. Patients enrolled at this site were included in the ITT and safety populations, but a post hoc analysis of the primary efficacy variable was carried out using the ITT population excluding the patients from this site. Efficacy endpoints The primary efficacy endpoint was the proportion of responders (participants with an average frequency of three or more SCBMs per week) over 24 weeks of treatment, assessed using the ITT population. The proportion of responders at each 1-, 4-, and 12-week period was also assessed. Demographic data, baseline characteristics, and AE data were presented descriptively. A Cochran–Mantel–Haenszel test, stratified for the randomization stratification factors (0 or >0 complete bowel movements [CBMs] per week during the run-in phase, country and sex), was used to compare the primary endpoint for prucalopride vs placebo. The last non-missing observation was used as the patient’s final on-treatment assessment for visit-based parameters. For efficacy endpoints based on e-diary data, if a participant did not complete the e-diary up to day 168 but had at least 14 non-missing e-diary days after the first week of treatment, the data from the last 14 non-missing diary days were carried forward to impute the missing data up to day 168. For the primary endpoint analysis of the 1–4-, 1–12-, and 1–24-week periods, patients who discontinued the study before day 28 were deemed to be non-responders. Sensitivity analyses of the primary efficacy endpoint were prespecified and described in detail in the statistical analysis plan. These included primary analysis on the per-protocol population and on patients who completed the study (complete-case analysis) and a generalized mixed model for repeated measures and multiple imputation. Multiple imputation was used to impute missing weekly frequencies of SCBMs based on the treatment group to which that patient belonged. The number of imputations was set to 15, and a logistic regression model was applied to each of the 15 imputed datasets, including country, number of CBMs/week during the run-in period (0 or >0), sex, and treatment as factors in the model. Response rates per treatment group and the difference in response rates were estimated per imputation together with the 95% confidence interval (95% CI). In a subsequent step, results were combined into a single outcome with a test for significance of the difference between response rates. A generalized linear mixed model for repeated measures, including treatment, week, and treatment 9 week as factors, and baseline SCBM as a covariate in the model, was used to analyze the proportion of patients who were responders at each week.
Secondary efficacy variables derived from e-diary data were evaluated over the 24-week treatment period and by 1 and 4 week periods in the ITT population. Additionally, the results of the PAC-SYM and PAC-QOL questionnaires were assessed at visits throughout the study from baseline to the final on-treatment assessment.
RESULTS Population Overall, 364 patients were randomized equally to prucalopride (N = 182) or placebo (N = 182; Fig. 2). Three participants did not receive treatment; the remainder were included in the safety population (prucalopride, N = 181; placebo, N = 180). Twentyone patients were excluded from the ITT population as the unblinded treatment arm information was inadvertently present in a report sent to several sites due to a programming error. The final ITT population consisted of 171 patients treated with prucalopride and 169 with placebo. Of the participants in the ITT population, 38 in the prucalopride group and 31 in the placebo group stopped treatment before day 28 or had protocol violations with a potential impact on the primary efficacy variable; these patients were excluded from the per-protocol population. A total of 17 individuals were enrolled at the site at which GCP issues were identified: 5 patients in the prucalopride group and 12 in the placebo group. These patients were included in the ITT and safety populations as per the study protocol, but a post hoc analysis of the primary efficacy variable was also carried out using the ITT population excluding those from this site. The majority of patients (74.2% and 69.2% for prucalopride and placebo groups, respectively) completed the study. Throughout the 24-week study period, the most common reasons for discontinuation were withdrawal of consent (prucalopride, 6.0%; placebo, 14.8%) and AEs (prucalopride, 7.7%; placebo, 5.5%). The discontinuation rate at week 12 revealed that a total of 14.6% of patients discontinued in the prucalopride group, compared with 17.8% in the placebo group. The most common reasons for discontinuation at week 12 were withdrawal of consent, which occurred nearly three times more frequently in the placebo group (10.1%) than in the prucalopride group (3.5%), and AEs (prucalopride, 6.4%; placebo, 3.6%). Baseline demographics for the safety population are shown in Table 1. Similar baseline demographics were seen in the other analysis populations (data not shown). The mean age was 48.9 years (standard
808 © 2015 The Authors. Neurogastroenterology & Motility published by John Wiley & Sons Ltd.
Volume 27, Number 6, June 2015
Long-term treatment with prucalopride
Screened N = 469 Screening failures N = 105
Randomized to placebo N = 182
Randomized to prucalopride N = 182
Excluded from safety population • Did not receive treatment (n = 2)
Excluded from safety population • Did not receive treatment (n = 1)
Safety population N = 180
Safety population N = 181
Excluded from intent-to-treat population • Potential unblinding (n = 11)
Excluded from intent-to-treat population • Potential unblinding (n = 10)
Intent-to-treat population N = 169
Intent-to-treat population N = 171
Excluded from per-protocol population • Stopped treatment before day 28 or protocol violations (n = 31)
Excluded from per-protocol population • Stopped treatment before day 28 or protocol violations (n = 38)
Per-protocol population N = 138
Discontinued • Withdrew consent (n = 27) • Adverse event (n = 10) • Sponsor decision (n = 9) • Other (n = 6) • Did not fulfil eligibility criteria (n = 2) • Non-compliant (n = 2) Completed study N = 126
Per-protocol population N = 133
Discontinued • Adverse event (n = 14) • Sponsor decision (n = 12) • Withdrew consent (n = 11) • Other (n = 8) • Did not fulfil eligibility criteria (n = 2) Completed study N = 135
Figure 2 Patient flow.
deviation [SD]: 16.00 years) and most patients were women (85.3%). The majority of baseline demographic characteristics were similar in both treatment groups: participants in the prucalopride group had a mean duration of constipation of 15.7 years (SD: 15.77 years) and the placebo group had a mean duration of constipation of 13.7 years (SD: 13.33 years).The most common symptoms of constipation were a feeling of not completely emptying bowels (24.7%) and infrequent defecation (24.4%). At baseline, the majority (60.0%) of patients reported having had 0 SCBMs per week over the past 6 months, and only 1.2% reported 3 or more SCBMs per week during this period.
Efficacy Primary endpoint The proportion of responders (patients who achieved an average of ≥3 SCBMs per week) in the ITT population (Fig. 3) over the 24-week, double-blind treatment period was not statistically different (p = 0.367) between the prucalopride (25.1%) and placebo (20.7%) treatment groups. The results were similar for the 1–12 week period (prucalopride, 25.1%; placebo, 20.1%; p = 0.341) and the 13–24 week period (prucalopride, 28.1%; placebo, 23.7%; p = 0.275). Sensitivity analyses of the primary endpoint were performed using the per-protocol population and the
© 2015 The Authors. 809 Neurogastroenterology & Motility published by John Wiley & Sons Ltd.
H. Piessevaux et al.
Neurogastroenterology and Motility
Table 1 Baseline demographics and disease characteristics of the safety population Total (N = 361)
48.3 (16.25) 149 (82.8)
49.4 (15.78) 146 (80.7)
48.9 (16.00) 295 (81.7)
153 (85.0) 27 (15.0) 24.7 (4.36)
155 (85.6) 26 (14.4) 25.5 (4.80)
308 (85.3) 53 (14.7) 25.1 (4.60)
Proportion of responder with three or more SCBMs per week (%)
Prucalopride (N = 181)
Placebo (n = 169) Prucalopride (n = 171)
90 80 70 60 50 40
p = 0.367
144
145
289
13.7 (13.33)
15.7 (15.77)
14.7 (14.61)
169
171
340
p = 0.341 p = 0.275
30 20 10 0
1–24
1–12 Weeks
13–24
Figure 3 Proportion of patients with a mean frequency of three or more SCBMs per week (ITT population). ITT, intent-to-treat; SCBM, spontaneous complete bowel movement.
100
97 41 25 4 2
(57.4) (24.3) (14.8) (2.4) (1.2)
107 34 24 4 2
(62.6) (19.9) (14.0) (2.3) (1.2)
204 75 49 8 4
(60.0) (22.1) (14.4) (2.4) (1.2)
46 (25.6)
43 (23.8)
89 (24.7)
43 35 25 16 15
45 28 25 26 14
88 63 50 42 29
(23.9) (19.4) (13.9) (8.9) (8.3)
(24.9) (15.5) (13.8) (14.4) (7.7)
(24.4) (17.5) (13.9) (11.6) (8.0)
Proportion of patients (%)
Age, years Mean (SD) 0 to
doi: 10.1111/nmo.12553
A randomized, double-blind, placebo-controlled trial to evaluate the efficacy, safety, and tolerability of long-term treatment with prucalopride H. PIESSEVAUX ,* E. CORAZZIARI ,† E. REY ,‡ M. SIMREN ,§ A. WIECHOWSKA-KOZLOWSKA ,¶ R. KERSTENS ,** M. COOLS ,** K. BARRETT ††
& A. LEVINE ‡‡
*Cliniques Universitaires Saint-Luc, Brussels, Belgium †Unit a Operativa Complessa di Gastroenterologia, Universita degli Studi di Roma “La Sapienza”, Rome, Italy ‡Department of Digestive Diseases, Instituto de Investigaci on Sanitaria del Hospital Clınico San Carlos, Hospital Clınico San Carlos, Universidad Complutense de Madrid, Madrid, Spain §University of Gothenburg, Gothenburg, Sweden ¶Niepubliczny Zaklad Opieki Zdrowotnej, Szczecin, Poland **Shire Movetis NV, Turnhout, Belgium ††Shire, Basingstoke, UK ‡‡Shire, Chesterbrook, PA, USA
Key Messages
• In contrast to the results of four previous 12-week trials, this study found no significant improvement in the • • •
proportion of patients achieving three or more spontaneous complete bowel movements (SCBMs) per week over 24 weeks when treated with prucalopride compared with placebo. Despite extensive evaluation, no explanation for these null efficacy results could be found. The aim of this randomized, placebo-controlled trial was to assess the efficacy of 24 weeks of prucalopride treatment in patients with chronic constipation. Patients with chronic constipation who had less than three SCBMs per week were randomized to receive prucalopride 2 mg or placebo daily for 24 weeks. The primary endpoint was the proportion of patients achieving a mean of three or more SCBMs per week over the treatment period. Overall, 361 patients were enrolled in the trial. The proportion of patients achieving the primary endpoint was not statistically different between the prucalopride and placebo groups (25.1% vs 20.7%; p = 0.367). No new safety concerns were identified.
Abstract Background Randomized trials have confirmed the efficacy of prucalopride for the treatment of chronic constipation up to 12 weeks. This study aimed to assess the efficacy of prucalopride over a 24-week period (ClinicalTrials.gov: NCT01424228). Methods Address for Correspondence Dr H. Piessevaux, Cliniques Universitaires Saint-Luc, Avenue Hippocrate 10, 1200 Brussels, Belgium. Tel: +32 2764 2834; fax: + 32 2764 8927; e-mail: [email protected] Received: 16 October 2014 Accepted for publication: 24 February 2015
Adults with chronic constipation and ≤2 spontaneous complete bowel movements (SCBMs)/week were randomized to receive prucalopride 2 mg or placebo daily for 24 weeks. The primary endpoint was the proportion of patients achieving a mean of ≥3 SCBMs/week over the treatment period, assessed using daily e-diaries. Secondary outcomes and safety parameters were assessed throughout the study. Key Results Overall, 361 patients were randomized and received prucalopride or placebo. Baseline characteristics were similar in the prucalopride (N = 181) and placebo (N = 180) groups. Mean age was 48.9 years (standard deviation, 16.0) and most patients were women.
© 2015 The Authors. 805 Neurogastroenterology & Motility published by John Wiley & Sons Ltd. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.
H. Piessevaux et al.
Neurogastroenterology and Motility
The proportion of participants achieving the primary endpoint was not statistically different between the prucalopride and placebo groups (25.1% vs 20.7%; p = 0.367). There was also no statistically significant difference between groups over the first 12-week period (prucalopride, 25.1%; placebo, 20.1%; p = 0.341). There were no statistically significant differences between groups for most secondary endpoints. No new safety concerns were identified. Conclusions & Inferences This trial did not show statistically significant improvements in primary or secondary outcomes with prucalopride compared with placebo over 24 or 12 weeks. This is in contrast to the results of four previous 12-week trials, which demonstrated prucalopride to be significantly more effective than placebo. An extensive evaluation did not provide an explanation for the null efficacy results of this study. Keywords clinical prucalopride.
trial,
constipation,
long-term,
Abbreviations: AE, Adverse event; BM, Bowel movement; CBM, Complete bowel movement; CI, Confidence interval; ECG, Electrocardiogram; GCP, Good clinical practice; HRQoL, Health-related quality of life; ITT, Intent-to-treat; PAC-QOL, Patient Assessment of Constipation-Quality of Life; PAC-SYM, Patient Assessment of Constipation-Symptom; SCBM, Spontaneous complete bowel movement; SD, Standard deviation; TEAE, Treatment-emergent adverse event.
INTRODUCTION Chronic constipation is a common disorder characterized by low stool frequency and gastrointestinal symptoms such as straining, hard stools, bloating, and incomplete evacuation. Initial treatment recommendations include dietary and lifestyle changes, and use of laxatives, which are available both over the counter and on prescription. However, many patients with chronic constipation experience long-term symptoms despite these interventions.1 An Internet survey in the United States indicated that 43% of 557 participants with constipation had experienced their symptoms for more than 4 years,2 and a recent European survey found that 59% of patients with constipation had symptoms for more than 3 years.3 In the latter study, the majority of participants used laxatives, but only 28% were satisfied with their treatment.3 Effective, long-term treatments are therefore needed for patients with chronic constipation. Prucalopride is a 5-hydroxytryptamine receptor 4 agonist that stimulates intestinal motility and is
licensed in the European Union for the symptomatic treatment of chronic constipation in adult women in whom laxatives fail to provide adequate relief. The recommended dose of prucalopride is 2 mg/day; in individuals over the age of 65 years, the starting dose is 1 mg/day.4 Four phase III clinical trials have shown prucalopride to be significantly more effective than placebo at increasing the proportion of patients achieving three or more spontaneous complete bowel movements (SCBMs) per week over 12 weeks of treatment, the primary endpoint of the studies.5–9 These trials also demonstrated that prucalopride was better than placebo at alleviating symptoms of constipation and improving health-related quality of life (HRQoL) during this time. In two long-term, open-label, follow-up studies, patients were enrolled from three of the phase III trials of prucalopride5–7 and reported sustained satisfaction when prucalopride treatment was continued after the end of the double-blind trials.10 Improvements in average Patient Assessment of Constipation-Quality of Life (PAC-QOL) satisfaction scores observed at the end of the 12-week, double-blind trials were maintained during open-label treatment for up to 18 months. While these results are suggestive of long-term efficacy, the limitations of these open-label studies include that the primary efficacy endpoint was evaluated using PAC-QOL, not SCBM frequency, and that there was no placebo comparator group. A randomized, double-blind, placebo-controlled clinical trial was therefore performed to assess further the long-term efficacy and safety of treatment with prucalopride. The primary objective of this study was to assess the efficacy of prucalopride compared with placebo over a 24-week period. The primary endpoint of the study was defined as the proportion of participants with a mean of three or more SCBMs per week evaluated over this 24-week period. Secondary objectives were to evaluate the long-term safety, tolerability, and effect on HRQoL of prucalopride compared with placebo.
METHODS Study population This was a stratified, randomized, parallel-group, double-blind, placebo-controlled, phase IV study (ClinicalTrials.gov Identifier: NCT01424228). It was carried out at 50 sites across Europe (Belgium, Czech Republic, Hungary, Italy, Poland, Romania, Slovakia, Spain, and Sweden) from 6 April 2011 to 19 December 2012. This study was carried out in accordance with local ethical and legal requirements, the International Conference on Harmonization of good clinical practice (GCP)11 and the principles of the Declaration of Helsinki.12 All participants provided written, informed consent before entry into the trial.
806 © 2015 The Authors. Neurogastroenterology & Motility published by John Wiley & Sons Ltd.
Volume 27, Number 6, June 2015
Long-term treatment with prucalopride
Inclusion criteria Men or non-pregnant, non-breastfeeding women aged 18 years and older with chronic constipation were eligible for inclusion. Chronic constipation was defined as two or fewer SCBMs (bowel movements not preceded within 24 h by the use of laxatives or enema that resulted in a feeling of complete evacuation) per week, and one or more of the following for a minimum of 6 months: very hard or hard stools for at least 25% of stools; sensation of incomplete evacuation following at least 25% of stools; and straining at defecation for at least 25% of the time. Exclusion criteria Participants were excluded from the trial if they were considered to have drug-induced constipation, or constipation secondary to causes such as endocrine disorders, metabolic disorders, neurological disorders, or surgery. Additional exclusion criteria included a history of clinically significant (as evaluated by the investigator) cancer or cardiac, vascular, liver, pulmonary, endocrine, neurological or psychiatric disorders, or metabolic disturbances. Participants were also excluded if they had previously used prucalopride. Sample size For this study, a 15% and 30% response rate for the placebo and prucalopride treatment groups, respectively, was assumed based on results from previous large studies of prucalopride in chronic constipation.5–7 The 15% difference in response rates could be detected with 161 patients per treatment arm at a power of 90%. It was assumed that 5% of participants would have insufficient e-diary data to evaluate the primary endpoint; therefore, 340 participants were targeted to be randomized.
Trial design The trial design is shown in Fig. 1. After an initial screening visit, participants underwent a run-in phase of 2 weeks. If patients needed a colonoscopy after screening and/or if they had been using agents that influence bowel habits, the run-in phase could extend to up to 4 weeks. Following the baseline visit, patients returned to the treatment center for assessments at weeks 2, 4, 8, 12, and 24. At baseline, patients were randomized 1 : 1 to receive placebo or prucalopride once daily. They were instructed to take their treatment before breakfast. Randomization was carried out using an interactive web-based/voice-response system (Oracleâ, Vilvoorde, Belgium). Participants younger than 65 years took prucalopride 2 mg once daily. Participants aged 65 years or older started on a daily dose of prucalopride 1 mg, consistent with the labeled dose in Europe for patients of this age. The dose was increased to 2 mg in the case of insufficient response (an average
of fewer than three SCBMs per week in the previous 2 weeks) at weeks 2 or 4. No further dose adjustments were allowed. The use of agents that influence bowel habits, such as prokinetics or anticholinergic drugs, was disallowed during the trial. Laxative use was also disallowed during the trial, except for bisacodyl as a rescue medication (used if a patient had not had a bowel movement for ≥3 consecutive days). Use of rescue medication was not permitted in the 24 h before or the 48 h after the baseline visit.
Assessments Daily e-diaries completed by the patient were used to gather information on intake of study and rescue medication, bowel movement frequency (date and time of each bowel movement) and consistency (using the Bristol Stool Scale13), degree of straining (rated as none, mild, moderate, severe, or very severe), and feeling of complete evacuation after a bowel movement (yes or no). At each visit from screening onwards, patients were asked to score the severity of their stool, abdominal, and rectal symptoms on 5-point scales (from 0 = absent to 4 = very severe) using the validated Patient Assessment of Constipation-Symptom (PAC-SYM) questionnaire.14 Health-related quality of life parameters were assessed using the PAC-QOL questionnaire administered at screening, baseline, and weeks 4, 12, and 24. The PAC-QOL assesses constipation-related HRQoL parameters on four subscales (physical discomfort, psychosocial discomfort, worries and concerns, and satisfaction) that are scored on a 5-point scale (0 = none/not at all, 4 = extremely/all the time).15,16
Adverse events and safety Adverse events (AEs) were recorded from the signing of informed consent at screening to the final patient visit. Treatment-emergent AEs were defined as AEs that occurred after the first administration of the investigational product up to 5 days after the last dose of investigational product. Serious treatment-emergent adverse events (TEAEs) were collected until 30 days after the last dose of investigational product. Vital signs were monitored at every visit. Electrocardiograms (ECGs) were recorded, physical examinations were carried out, and samples for blood and urine tests were collected at screening, baseline, and at the 4- and 24-week visits or at early discontinuation.
Placebo Washout*
Run-in Prucalopride 2 mg†
Visit 1 Visit 2 Visit 3 Visit 4 (Screening) (Baseline) Week 2 Week 4 Week –4 to –2 Day 1
Visit 5 Week 8
Visit 6 Week 12
Visit 7 Week 24
Figure 1 Study design. *Washout if required due to colonoscopy or use of medication affecting bowel habits. †Patients aged ≥65 years started at 1 mg daily. In cases of insufficient response, the dose increased to 2 mg daily at week 2 or 4.
© 2015 The Authors. 807 Neurogastroenterology & Motility published by John Wiley & Sons Ltd.
H. Piessevaux et al.
Neurogastroenterology and Motility
Statistical methods Analysis populations The primary efficacy analysis was carried out in the intent-to-treat (ITT) population, which included all randomized patients who took at least one dose of the trial medication. Twenty-one patients were excluded from the ITT population as the unblinded treatment arm information was inadvertently present in a report sent to several sites due to a programming error. The safety population included all randomized patients who took at least one dose of the trial medication, inclusive of the 21 individuals omitted from the ITT population. The per-protocol population included patients who took their treatment for at least 28 days and excluded those with major protocol violations that could have impacted on efficacy or safety analyses (as determined by the study team prior to database lock). The completers population consisted of patients in the ITT population who completed the study. A site audit revealed two critical findings that were categorized as serious breaches of GCP at one site participating in the study. Patients enrolled at this site were included in the ITT and safety populations, but a post hoc analysis of the primary efficacy variable was carried out using the ITT population excluding the patients from this site. Efficacy endpoints The primary efficacy endpoint was the proportion of responders (participants with an average frequency of three or more SCBMs per week) over 24 weeks of treatment, assessed using the ITT population. The proportion of responders at each 1-, 4-, and 12-week period was also assessed. Demographic data, baseline characteristics, and AE data were presented descriptively. A Cochran–Mantel–Haenszel test, stratified for the randomization stratification factors (0 or >0 complete bowel movements [CBMs] per week during the run-in phase, country and sex), was used to compare the primary endpoint for prucalopride vs placebo. The last non-missing observation was used as the patient’s final on-treatment assessment for visit-based parameters. For efficacy endpoints based on e-diary data, if a participant did not complete the e-diary up to day 168 but had at least 14 non-missing e-diary days after the first week of treatment, the data from the last 14 non-missing diary days were carried forward to impute the missing data up to day 168. For the primary endpoint analysis of the 1–4-, 1–12-, and 1–24-week periods, patients who discontinued the study before day 28 were deemed to be non-responders. Sensitivity analyses of the primary efficacy endpoint were prespecified and described in detail in the statistical analysis plan. These included primary analysis on the per-protocol population and on patients who completed the study (complete-case analysis) and a generalized mixed model for repeated measures and multiple imputation. Multiple imputation was used to impute missing weekly frequencies of SCBMs based on the treatment group to which that patient belonged. The number of imputations was set to 15, and a logistic regression model was applied to each of the 15 imputed datasets, including country, number of CBMs/week during the run-in period (0 or >0), sex, and treatment as factors in the model. Response rates per treatment group and the difference in response rates were estimated per imputation together with the 95% confidence interval (95% CI). In a subsequent step, results were combined into a single outcome with a test for significance of the difference between response rates. A generalized linear mixed model for repeated measures, including treatment, week, and treatment 9 week as factors, and baseline SCBM as a covariate in the model, was used to analyze the proportion of patients who were responders at each week.
Secondary efficacy variables derived from e-diary data were evaluated over the 24-week treatment period and by 1 and 4 week periods in the ITT population. Additionally, the results of the PAC-SYM and PAC-QOL questionnaires were assessed at visits throughout the study from baseline to the final on-treatment assessment.
RESULTS Population Overall, 364 patients were randomized equally to prucalopride (N = 182) or placebo (N = 182; Fig. 2). Three participants did not receive treatment; the remainder were included in the safety population (prucalopride, N = 181; placebo, N = 180). Twentyone patients were excluded from the ITT population as the unblinded treatment arm information was inadvertently present in a report sent to several sites due to a programming error. The final ITT population consisted of 171 patients treated with prucalopride and 169 with placebo. Of the participants in the ITT population, 38 in the prucalopride group and 31 in the placebo group stopped treatment before day 28 or had protocol violations with a potential impact on the primary efficacy variable; these patients were excluded from the per-protocol population. A total of 17 individuals were enrolled at the site at which GCP issues were identified: 5 patients in the prucalopride group and 12 in the placebo group. These patients were included in the ITT and safety populations as per the study protocol, but a post hoc analysis of the primary efficacy variable was also carried out using the ITT population excluding those from this site. The majority of patients (74.2% and 69.2% for prucalopride and placebo groups, respectively) completed the study. Throughout the 24-week study period, the most common reasons for discontinuation were withdrawal of consent (prucalopride, 6.0%; placebo, 14.8%) and AEs (prucalopride, 7.7%; placebo, 5.5%). The discontinuation rate at week 12 revealed that a total of 14.6% of patients discontinued in the prucalopride group, compared with 17.8% in the placebo group. The most common reasons for discontinuation at week 12 were withdrawal of consent, which occurred nearly three times more frequently in the placebo group (10.1%) than in the prucalopride group (3.5%), and AEs (prucalopride, 6.4%; placebo, 3.6%). Baseline demographics for the safety population are shown in Table 1. Similar baseline demographics were seen in the other analysis populations (data not shown). The mean age was 48.9 years (standard
808 © 2015 The Authors. Neurogastroenterology & Motility published by John Wiley & Sons Ltd.
Volume 27, Number 6, June 2015
Long-term treatment with prucalopride
Screened N = 469 Screening failures N = 105
Randomized to placebo N = 182
Randomized to prucalopride N = 182
Excluded from safety population • Did not receive treatment (n = 2)
Excluded from safety population • Did not receive treatment (n = 1)
Safety population N = 180
Safety population N = 181
Excluded from intent-to-treat population • Potential unblinding (n = 11)
Excluded from intent-to-treat population • Potential unblinding (n = 10)
Intent-to-treat population N = 169
Intent-to-treat population N = 171
Excluded from per-protocol population • Stopped treatment before day 28 or protocol violations (n = 31)
Excluded from per-protocol population • Stopped treatment before day 28 or protocol violations (n = 38)
Per-protocol population N = 138
Discontinued • Withdrew consent (n = 27) • Adverse event (n = 10) • Sponsor decision (n = 9) • Other (n = 6) • Did not fulfil eligibility criteria (n = 2) • Non-compliant (n = 2) Completed study N = 126
Per-protocol population N = 133
Discontinued • Adverse event (n = 14) • Sponsor decision (n = 12) • Withdrew consent (n = 11) • Other (n = 8) • Did not fulfil eligibility criteria (n = 2) Completed study N = 135
Figure 2 Patient flow.
deviation [SD]: 16.00 years) and most patients were women (85.3%). The majority of baseline demographic characteristics were similar in both treatment groups: participants in the prucalopride group had a mean duration of constipation of 15.7 years (SD: 15.77 years) and the placebo group had a mean duration of constipation of 13.7 years (SD: 13.33 years).The most common symptoms of constipation were a feeling of not completely emptying bowels (24.7%) and infrequent defecation (24.4%). At baseline, the majority (60.0%) of patients reported having had 0 SCBMs per week over the past 6 months, and only 1.2% reported 3 or more SCBMs per week during this period.
Efficacy Primary endpoint The proportion of responders (patients who achieved an average of ≥3 SCBMs per week) in the ITT population (Fig. 3) over the 24-week, double-blind treatment period was not statistically different (p = 0.367) between the prucalopride (25.1%) and placebo (20.7%) treatment groups. The results were similar for the 1–12 week period (prucalopride, 25.1%; placebo, 20.1%; p = 0.341) and the 13–24 week period (prucalopride, 28.1%; placebo, 23.7%; p = 0.275). Sensitivity analyses of the primary endpoint were performed using the per-protocol population and the
© 2015 The Authors. 809 Neurogastroenterology & Motility published by John Wiley & Sons Ltd.
H. Piessevaux et al.
Neurogastroenterology and Motility
Table 1 Baseline demographics and disease characteristics of the safety population Total (N = 361)
48.3 (16.25) 149 (82.8)
49.4 (15.78) 146 (80.7)
48.9 (16.00) 295 (81.7)
153 (85.0) 27 (15.0) 24.7 (4.36)
155 (85.6) 26 (14.4) 25.5 (4.80)
308 (85.3) 53 (14.7) 25.1 (4.60)
Proportion of responder with three or more SCBMs per week (%)
Prucalopride (N = 181)
Placebo (n = 169) Prucalopride (n = 171)
90 80 70 60 50 40
p = 0.367
144
145
289
13.7 (13.33)
15.7 (15.77)
14.7 (14.61)
169
171
340
p = 0.341 p = 0.275
30 20 10 0
1–24
1–12 Weeks
13–24
Figure 3 Proportion of patients with a mean frequency of three or more SCBMs per week (ITT population). ITT, intent-to-treat; SCBM, spontaneous complete bowel movement.
100
97 41 25 4 2
(57.4) (24.3) (14.8) (2.4) (1.2)
107 34 24 4 2
(62.6) (19.9) (14.0) (2.3) (1.2)
204 75 49 8 4
(60.0) (22.1) (14.4) (2.4) (1.2)
46 (25.6)
43 (23.8)
89 (24.7)
43 35 25 16 15
45 28 25 26 14
88 63 50 42 29
(23.9) (19.4) (13.9) (8.9) (8.3)
(24.9) (15.5) (13.8) (14.4) (7.7)
(24.4) (17.5) (13.9) (11.6) (8.0)
Proportion of patients (%)
Age, years Mean (SD) 0 to
