A pooled analysis of three phase III studies to investigate the efficacy, tolerability and safety of darifenacin, a muscarinic M3 selective receptor antagonist, in the treatment of overactive bladder

Authors


Christopher Chapple, Urology Research, Royal Hallamshire Hospital, J Floor, Glossop Road, Sheffield, S10 2JF, UK.
e-mail: c.r.chapple@shef.ac.uk

Abstract

An international group of authors present a pooled analysis of data from their phase III multicentre double-blind clinical trials in patients with overactive bladder, which evaluated the efficacy, tolerability and safety of darifenacin. They found the drug, a muscarinic M3 selective receptor antagonist, to be effective in the treatment of this condition, with excellent tolerability and safety.

A paper from Denmark compares the efficacy and safety of alfuzosin and tamsulosin in a large randomized, double-blind, placebo-controlled, multicentre study. There were similar improvements in urinary symptoms and maximum urinary flow with the two drugs compared to placebo, but the incidence of sexual function adverse events was higher with tamsulosin than placebo.

OBJECTIVE

To evaluate the efficacy, tolerability and safety of darifenacin, a muscarinic M3 selective receptor antagonist (M3 SRA), from an analysis of pooled data from three phase III, multicentre, double-blind clinical trials in patients with overactive bladder (OAB).

PATIENTS AND METHODS

After a 4-week washout/run-in period, 1059 adults (85% women) with symptoms of OAB (frequency and urgency with urge incontinence) for ≥ 6 months were randomized to once-daily oral treatment with darifenacin (7.5 mg, 337; or 15 mg, 334) or matching placebo (388) for 12 weeks. Efficacy was evaluated using electronic patient diaries that recorded incontinence episodes (including those resulting in a change of clothing or pads), frequency and severity of urgency, voiding frequency, and bladder capacity (volume voided). Safety was evaluated by analysis of adverse events (AEs), withdrawal rates and laboratory tests.

RESULTS

Relative to baseline, 12 weeks of treatment with darifenacin resulted in a significant reduction in the median (% change, interquartile range) number of incontinence episodes per week; 7.5 mg (−8.8, −68.4%, −15.1 to −4.4); 15 mg; (−10.6, −76.8%, −17.3 to −5.8: both P < 0.01 vs placebo). There was a significant dose–response trend in each study for which darifenacin 7.5 and 15 mg were evaluated (P < 0.01). There were also significant decreases in the frequency and severity of urgency, voiding frequency, and number of significant leaks (incontinence episodes resulting in a change of clothing or pads; both P ≤ 0.001 vs placebo), together with an increase in bladder capacity (both P < 0.01 vs placebo). Darifenacin was well tolerated; the most common AEs were dry mouth and constipation, although together these resulted in few discontinuations (darifenacin 7.5 mg 0.6% of patients; 15 mg 2.1%; placebo 0.3%). The incidence of peripheral/central nervous system and cardiovascular AEs were comparable with those on placebo.

CONCLUSIONS

Darifenacin (7.5 and 15 mg once daily) is effective in the treatment of patients with OAB. As predicted by its M3 selectivity and associated M1/M2-sparing profile, darifenacin was well tolerated with no central nervous system or cardiovascular safety concerns.

Abbreviations
SRA

selective receptor antagonist

OAB

overactive bladder

AE

adverse event

CYP

cytochrome P450.

INTRODUCTION

Overactive bladder syndrome (OAB) is a term used to describe a collection of urinary symptoms composed of urgency (a sudden and strong desire to void) with or without urge incontinence, usually with increased frequency of voiding and nocturia, in the absence of pathological or metabolic factors that would explain these symptoms [1]. Such symptoms are known to be highly prevalent within the general population, contributing to a significant impairment in quality of life [2].

Antimuscarinic agents block the stimulation of muscarinic receptors in the detrusor muscle of the bladder, but existing antimuscarinic agents are not highly selective for this effect. This explains their ability to cause a range of side-effects, such as dry mouth and constipation. A drug with selectivity for the muscarinic M3 receptor, the subtype primarily responsible for mediating human detrusor contraction [3], might be effective in treating OAB symptoms and with a narrower side-effect profile. Such a drug might reduce tolerability and safety concerns related to blockade of other receptor subtypes, including cognitive impairment (primarily M1-receptor mediated) and cardiac effects (primarily M2-receptor mediated) [4].

Darifenacin is a new muscarinic M3 selective receptor antagonist (M3 SRA) for treating OAB. Unlike less selective agents (e.g. oxybutynin) and nonselective agents (e.g. tolterodine), darifenacin has up to 59-fold selectivity for M3 receptors relative to other receptor subtypes [5]. This paper details the efficacy, tolerability and safety findings from an analysis of pooled data from three phase III clinical trials of darifenacin. The phase III studies were undertaken to evaluate darifenacin over a wide range of clinically important variables in a large population of patients with OAB, and to further explore the dose–response relationship subsequent to initial observations from phase II studies, in which the 7.5 mg dose had not been fully evaluated. A pooled analysis of these trials is appropriate because of the near-identical design and enrolment criteria for patients with OAB.

PATIENTS AND METHODS

Efficacy and safety data from three randomized, double-blind, placebo-controlled, parallel-group, multicentre, fixed-dose, 12-week studies with darifenacin were pooled for this analysis. The 15 mg once-daily dose was investigated in all three studies; two of the studies also investigated the 7.5 mg once-daily dose.

All studies were approved by local ethics committees/institutional review boards at each study centre. Each study was preceded by a screening visit, 2-week washout (if required, for patients on prohibited medications) and a 2-week treatment-free or placebo run-in period, after which patients were randomized to study medication. Further study visits were after 2, 6 and 12 weeks of treatment. Blinding was maintained in each study by the use of matching placebo tablets for the nonactive dosage forms (double-dummy design). Compliance was determined by counting the returned study medication.

Men and women aged ≥ 18 years with symptoms of OAB for ≥ 6 months, and capable of independent toileting, were enrolled into each study. They were eligible for inclusion if they had 5–50 episodes of incontinence per week during the run-in period, and a high voiding frequency (a mean of ≥ 8 voids/24 h) and urgency (a mean of ≥ 1 episode/24 h). Women of childbearing potential were required to use an adequate method of contraception throughout the study. Those taking hormone-replacement therapy had to have received such therapy for ≥ 2 months before entering the study. Similarly, men taking finasteride for BPH had to be on a stable dose for ≥ 6 months. Those receiving long-term therapy with diuretics, antihypertensive medications, benzodiazepines or antihistamines had to be taking a stable dose before study recruitment, with no plans to change treatment during the study. Initiation of a bladder training programme was not permitted during the study; patients already on such a programme were not to modify or discontinue their training during the course of the study. Exclusion criteria included: pregnancy and lactation; clinically significant stress incontinence (i.e. ≥1 episode of stress incontinence per week), BOO and/or a postvoid residual urine volume of >200 mL (as measured by pelvic ultrasonography); clinically important medical problems that would interfere with the patient's participation in the study; patients with interstitial cystitis, severe constipation (two or fewer bowel movements per week), haematuria or intermittent UTI; cystocele or other clinically significant pelvic prolapse; patients with an indwelling catheter and those who practised intermittent self-catheterization; urogenital surgery in the previous 6 months; patients with contraindications to antimuscarinic therapy (e.g. uncontrolled narrow-angle glaucoma, urinary retention, gastric retention); history of alcohol/drug abuse; and known hypersensitivity to study medications.

Concomitant treatment with cytochrome P450 (CYP) 2D6 inhibitors such as cimetidine, fluoxetine and paroxetine (one study only), potent CYP3A4 inhibitors (e.g. ketoconazole), opioids (or other drugs that could cause significant constipation), was not permitted during each study. Antimuscarinic agents not in the study, or other drugs with significant anticholinergic effects (e.g. tricyclic antidepressants) were also prohibited during each study, although patient selection was not based on previous responsiveness to antimuscarinic therapy. All patients provided written informed consent before study entry.

EVALUATION OF EFFICACY

Efficacy was determined with the use of electronic diaries (MiniDoc®, MiniDoc AB, Stockholm, Sweden) [6] that were completed daily by patients for 2 weeks (in two studies) or 1 week (one study) before each study visit at 0, 2, 6 and 12 weeks. Data on volume voided per void were collected for at least 2 days per diary-monitoring period. The primary efficacy variable was the median change in the number of incontinence episodes per week. Secondary efficacy variables included: the number of significant leaks per week (i.e. number of incontinence episodes that resulted in a change of clothing or pads); voiding frequency; bladder capacity (mean volume voided); frequency and severity of urgency (measured on a 100-mm visual analogue scale, where 0 = mild, 100 = severe); and the number of nocturnal awakenings caused by OAB (defined as the total number of times a patient woke for voiding and/or incontinence and/or urge).

As an additional secondary analysis, efficacy was also determined in terms of responder rates, i.e. the proportion of patients with a ≥ 30%, ≥ 50%, ≥ 70% or ≥ 90% reduction from baseline in the number of incontinence episodes per week, or a ≥ 30%, ≥ 50%, ≥ 70% or ≥ 90% reduction from baseline in the number of urgency episodes per week, and those who achieved a normal voiding frequency (<8 voids/day, on average). To support the clinical relevance of these changes, the effect of darifenacin on the proportion of patients experiencing three or more dry days/week, or at least 7 consecutive dry days, in the last 2 weeks of study treatment (11–12 weeks), was also assessed.

EVALUATION OF SAFETY

Safety was determined from adverse events (AEs), withdrawal rates, vital signs and routine laboratory measurements. All observed or volunteered AEs were recorded and evaluated by the investigator in terms of seriousness, intensity and causal relationship to study medication. Changes in vital signs (blood pressure, heart rate) and laboratory findings (clinical chemistry, haematology, urine analysis) between screening and study end were analysed for any clinically relevant trends.

STATISTICAL ANALYSIS

Efficacy was analysed for the full dataset, defined as all participants who received at least one dose of randomized study medication and had efficacy assessed both at baseline (where applicable) and afterward. The ‘last observation carried forward’ approach was used where on-treatment efficacy data were missing.

Differences from placebo for the median change from baseline to the end of treatment for the voiding diary variables were tested using the Wilcoxon rank-sum test, stratified by study (Van Elteren's test) [7]. Nonparametric tests were used because the distribution of the outcome data was not normal. The 7.5 mg dose was not used in all three studies and therefore the relevant placebo comparison for each dose used only the placebo response from the matching studies. For each active treatment group, the Hodges-Lehmann estimate, stratified by study, of the median difference from the corresponding placebo group, and its 95% CI, were derived [7,8]. Response rates for the darifenacin groups were compared against placebo by fitting a logistic model of treatment group with study and baseline value of the corresponding variable included as a covariate on the analyses of the pooled dataset. The median change from baseline is calculated as the median of the individual changes from baseline of the patients, and does not correspond to the difference of the median values from baseline to post-baseline.

Post-hoc statistical analysis for a dose–response trend was conducted by study, on the two studies that evaluated both the 7.5 and 15 mg doses. The P value for the dose–response trend test was one-sided and based on Jonckheere's test, and tested for change from baseline in the number of incontinence episodes per week for each dose. In view of the variability in baseline number of incontinence episodes per week, median differences from placebo were examined when considering the dose–response trend. A confirmatory analysis was also conducted fitting a Poisson generalized linear mixed-effects model and by testing a dose–response relationship using the slope parameter of dose, treated as a continuous covariate.

RESULTS

In all, 1059 patients (most of whom were female, 85%) were included in the pooled-data analysis. Most patients (94.8%) had a primary diagnosis of idiopathic OAB, the remainder (5.2%) having OAB of neurogenic origin, and 22% had previously received drug therapy for their OAB symptoms in the 3 months before study inclusion. Overall, demographic characteristics were similar across the three treatment groups (Table 1), reinforcing the validity of the comparison of data across studies, the pooling of relevant data for meta-analysis and the applicability of the trial results to the general OAB population. Most patients (>99%) in each treatment group complied satisfactorily during the study (>80% prescribed medication taken).

Table 1.  Baseline demographics and clinical characteristics (full analysis set)
CharacteristicDarifenacin, mg once dailyPlacebo
7.515
  • *

    Data missing for two patients;

  • †data missing for four patients.

N337334388
Females, n (%)288 (85.5)281 (84.1)331 (85.3)
Mean (range)
age, years 57 (22–88) 57 (24–85) 56 (19–88)
weight, kg 74 (42–129) 75 (38–154) 76 (40–163)
Idiopathic OAB diagnosis, n (%)313 (92.9)320 (95.8)371 (95.6)
History of previous drug treatment for OAB, n (%) 68 (20.2) 70 (21.0) 96 (24.7)
Median no. of incontinence episodes/week 16.0* 16.9 16.6

Rates of completion and reasons for discontinuation are summarized in Fig. 1; while most patients (87–94%) in each treatment group completed the study, AEs and withdrawal of consent were the most common reasons for discontinuation.

Figure 1.

Summary of patient flow. *All causes; †Lack of efficacy, consent withdrawn, protocol violation, lost to follow-up.

EFFICACY

After 12 weeks of treatment, both doses of darifenacin achieved a significantly greater reduction in the median number of incontinence episodes per week than did placebo; the median (interquartile range, ie Q25–Q75) change in weekly incontinence episodes from baseline was −8.8 (−15.1, −4.4) for 7.5 mg darifenacin (− 68.4%, 335 patients) and −10.6 (−17.3, −5.8) for 15 mg darifenacin (−76.8%, 330 patients), with P = 0.004 and <0.001 against placebo, respectively (corresponding changes of −53.8% and − 58.3%, respectively; 271 and 384 patients). There was a significant dose–response trend using both statistical tests and in both studies for which 7.5 and 15 mg darifenacin were evaluated (P < 0.001 and P = 0.002 using the one-sided analysis based on Jonckheere's test, and P < 0.001 for both studies at every visit using the Poisson generalized linear mixed-effects model). There was also a significant improvement in other diary variables compared with placebo (Table 2), and included decreases in the number of significant leaks (i.e. number of incontinence episodes that resulted in a change of clothing or pads), voiding frequency and number/severity of urgency episodes, and increased bladder capacity. The decrease in the number of nocturnal awakenings per week caused by OAB was also numerically greater with darifenacin than placebo, although the between-group difference was not statistically significant.

Table 2.  Effect of once daily treatment with either darifenacin or placebo for 12 weeks on secondary efficacy variables
Variable/treatment groupN evaluable patientsMedian baselineMedian change from baseline (%)Median difference from placebo (95% CI)P*
  • *

    calculated using Wilcoxon test, stratified by study;

  • Significantly different between doses (P = 0.0093), Wilcoxon test stratified by study.

Mean N voids/day
Darifenacin 7.5 mg335 10.2−1.6 (−16.6)−0.8 (−1.1, −0.4)<0.001
Placebo271 10.1−0.9 (−9.1)  
Darifenacin 15 mg330 10.6−1.9 (−17.4)−0.8 (−1.1, −0.4)<0.001
Placebo385 10.2−1.0 (−9.9)  
Mean N urgency episodes/day
Darifenacin 7.5 mg335  8.0−2.0 (−29.0)−0.8 (−1.3, −0.4)<0.001
Placebo271  8.2−1.0 (−14.3)  
Darifenacin 15 mg330  8.4−2.3 (−29.0)−0.9 (−1.3, −0.5)<0.001
Placebo384  8.4−1.2 (−16.7)  
Mean severity of urgency episodes/day
Darifenacin 7.5 mg335 54.4−7.8 (−14.2)−4.0 (−6.6, −1.6)0.001
Placebo270 56.3−4.2 (−7.8)  
Darifenacin 15 mg329 58.5−9.3 (−16.1)−4.9 (−7.1, −2.7)<0.001
Placebo383 56.5−4.5 (−8.0)  
Bladder capacity (average volume voided/void, mL)
Darifenacin 7.5 mg322 16115 (9.6)10 (3, 17)0.007
Placebo255162 8 (4.9)  
Darifenacin 15 mg32015527 (17.5)20 (14, 27)<0.001
Placebo3661576 (3.9)  
Mean N incontinence episodes/week resulting in change of clothing/pads
Darifenacin 7.5 mg333  8.1−4.0 (−77.1)−1.8 (−2.8, −0.9)<0.001
Placebo270  7.4−2.0 (−47.7)  
Darifenacin 15 mg324  8.0−4.8 (−78.6)−2.0 (−3.0, −1.1)<0.001
Placebo378  7.2−2.7 (−55.1)  
Mean N nocturnal awakenings/week for OAB
Darifenacin 7.5 mg334  11.6−1.7 (−14.9)−0.7 (−1.4, 0.2)0.13
Placebo269  11.0−0.8 (−9.5)  
Darifenacin 15 mg329  11.5−1.9 (−20.5)−0.7 (−1.4, 0.0)0.06
Placebo382  11.3−1.1 (−13.4)  

Analysis of response rates for the reduction in number of incontinence episodes yielded equally striking results. The proportion of patients who achieved a ≥ 70% reduction from baseline was 48% for 7.5 mg and 57% for 15 mg darifenacin, compared with only 33% and 39% of patients in the corresponding placebo groups (P < 0.001). Furthermore, 27% and 28% of patients in each of these groups achieved a ≥ 90% reduction from baseline, respectively, compared with only 17% in each of the placebo groups (P < 0.005) (Fig. 2a). The odds ratios for improvement compared with placebo were consistent for both doses across all responder rates analysed (odds ratio 1.8–1.9 for 7.5 mg and 1.8–2.2 for 15 mg darifenacin). Thus at any responder rate selected the magnitude of effect compared with placebo was maintained and there was a statistically significant difference (P < 0.005). Similarly, responder rates for the reduction in urgency episodes also showed significant differences from placebo (P < 0.05) for both doses of darifenacin at all levels of response (Fig. 2b) (≥30%, ≥50%, ≥70%, ≥90%).

Figure 2.

The percentage of patients who achieved a ≥30%, ≥ 50%, ≥ 70% or ≥ 90% reduction from baseline in; a, the number of incontinence episodes; and b, the number of urgency episodes at 12 weeks, for darifenacin 7.5 mg (red, matching placebo green) and 15 mg (light red, placebo light green). P values are shown over the bars, calculated using the Wilcoxon rank-sum test and stratified by study.

In addition to improvements in the number of episodes of incontinence and urgency, the proportion of patients who attained a normal voiding frequency (<8 voids/day) after 12 weeks of treatment was significantly greater with both doses of darifenacin (7.5 mg, 34%; P = 0.029 vs placebo; and 15 mg, 35%; P = 0.007 vs placebo) than in the corresponding placebo groups (27% and 28%, respectively).

The reduction in the number of episodes of urgency and incontinence, together with the lower frequency of voiding, experienced by patients taking darifenacin for 12 weeks, was paralleled by a significant increase in numbers of patients achieving ≥7 consecutive dry days, such that 24% of patients treated once daily with 15 mg darifenacin were ‘dry’ for at least 7 days, compared with 16% in the corresponding placebo group (P = 0.011; Table 3). Significantly more patients (55% and 61%) had ≥ 3 dry days per week in the darifenacin 7.5 mg and 15 mg groups, respectively, than in those taking placebo (43% and 48%, respectively, both P ≤ 0.001). These findings show a marked improvement from baseline, when <10% of patients in any group were found to have ≥ 3 dry days per week.

Table 3.  Percentage of patients with ≥ 3 dry days per week or ≥ 7 consecutive dry days in the 14 days before the 12-week visit
Treatment group% of patients responding
≥3 dry days/week≥7 consecutive dry days
Darifenacin 7.5 mg 5519
Placebo 4315
P vs placebo  0.001 0.108
Darifenacin 15 mg 6124
Placebo 4816
P vs placebo<0.001 0.011

TOLERABILITY AND SAFETY

The overall incidence of AEs of any cause (i.e. whether or not considered to be treatment-related) was 54% with 7.5 mg and 65.6% with 15 mg darifenacin, vs 48.7% with placebo (Table 4). A substantial proportion of AEs was considered to be unrelated to treatment, such that 37.1% and 52.1% of patients given darifenacin 7.5 mg or 15 mg, vs 20.9% given placebo, had AEs deemed to be treatment-related, respectively. The most common all-cause AEs were dry mouth and constipation, most of which were mild to moderate. Constipation infrequently necessitated laxative use, which included stool-softeners, bulk-forming agents (fibre supplements), osmotic and stimulant laxatives, during treatment with either darifenacin 7.5 mg or 15 mg, similar to placebo (22%, 30% and 25% of patients reporting constipation used laxatives, respectively).

Table 4.  A summary of most common all-causality AEs (reported by ≥3% of patients in any treatment group)
Adverse event, n (%)Darifenacin, mg once dailyPlacebo
7.515
N337334388
Dry mouth 68 (20.2) 118 (35.3) 32 (8.2)
Constipation 50 (14.8) 71 (21.3) 24 (6.2)
Headache 15 (4.5) 17 (5.1) 21 (5.4)
Dyspepsia  9 (2.7) 28 (8.4) 10 (2.6)
Respiratory tract infection  9 (2.7) 17 (5.1) 26 (6.7)
UTI 16 (4.7) 15 (4.5) 10 (2.6)
Abdominal pain  8 (2.4) 13 (3.9)  2 (0.5)
Back pain  8 (2.4)  5 (1.5) 12 (3.1)
≥1 AE182 (54.0)219 (65.6)189 (48.7)

Overall, the incidence of nervous system AEs (peripheral and/or CNS events) reported by patients taking 7.5 mg or 15 mg of darifenacin was comparable to placebo, whether assessed by all causalities (3.6% and 7.5% vs 7.2%, respectively), or as treatment-related events (1.5% and 2.7% vs 2.3%, respectively). The most common nervous system AEs were CNS-related: dizziness (darifenacin 7.5 mg, 0.9%; 15 mg, 2.1%; vs placebo 1.3%) and somnolence (0.3% and 0.9% vs 0.8%, respectively). The incidence of all-cause cardiovascular AEs with darifenacin 7.5 mg (6.2%) or 15 mg (3.6%) was also comparable with that of placebo (2.3%). Similarly, treatment-related cardiovascular AEs were reported by 2.1% and 0.9% vs 0.3% of patients given darifenacin 7.5 or 15 mg or placebo, respectively. No patient taking darifenacin had treatment-related tachycardia. There were no concerns from laboratory data and no clinically relevant changes in vital signs.

There were few discontinuations as a result of all-causality AEs (1.5% with 7.5 mg and 5.1% with 15 mg darifenacin, vs 2.6% with placebo), and even fewer for treatment-related AEs (1.2% and 4.5% vs 1.3%, respectively). Dry mouth and constipation were the most common reasons for discontinuation, although together these resulted in few discontinuations (darifenacin 7.5 mg, 0.6%; 15 mg, 2.1%; placebo 0.3%). Three patients had at least one treatment-related serious AE resulting in hospitalization; two placebo recipients (UTI and oesophagitis) and one patient treated with darifenacin 7.5 mg (second-degree atrioventricular block). The latter patient had a history of atrioventricular block before starting the study, and it was subsequently concluded that the study drug was an unlikely cause of this AE. The patient continued treatment with study drug with no further problem, subsequently having a pacemaker inserted.

DISCUSSION

Darifenacin is an M3 SRA developed for treating OAB; the findings of the present pooled analysis of three phase III clinical trials with this novel agent show it to be well tolerated and significantly better than placebo for improving OAB symptoms. Indeed, treatment with darifenacin was associated with significantly fewer weekly incontinence episodes, the primary efficacy variable in these studies. The median percentage reduction with the 7.5 and 15 mg doses was 68% and 77%, respectively, vs 54–58% among placebo recipients, i.e. half of all patients treated with darifenacin had >68–77% fewer weekly incontinence episodes than placebo. Furthermore, this reduction in incontinence with darifenacin had a statistically significant dose-dependent trend in individual studies (all P ≤ 0.002).

The large placebo response in this analysis was anticipated, in view of similar findings in other studies in patients with OAB [9], and is probably explained by the ‘bladder-training effect’ of completing a voiding diary, coupled with close patient monitoring and encouragement by nurses within the clinical trial setting. The placebo response in these studies may have been further accentuated by the use of electronic diaries, necessitating real-time input of data, thus prompting immediate consideration of bladder symptoms [6]. Despite the marked placebo effect, analysis of responder rates confirmed the clinical relevance of the efficacy of darifenacin. These analyses show that large proportions of patients achieved substantial benefits at both doses of darifenacin, with over a quarter of patients achieving a ≥ 90% reduction in incontinence. This is an important finding, given that incontinence is often regarded as the most distressing and lifestyle-limiting symptom reported by patients with OAB, and is the symptom that leads most patients to seek treatment. Furthermore, in addition to significantly fewer incontinence episodes, the increase in bladder capacity (volume voided), the most objective secondary variable, was significantly greater for patients treated with darifenacin than with placebo, further showing a real benefit with darifenacin, beyond the result of any behavioural changes. However, the magnitude of the additive effect of bladder training, including pelvic floor exercises, in connection with active treatment is unknown.

In the absence of effective treatment, patients having leakage often develop elaborate coping strategies, e.g. frequent voiding [10] and restricting fluid intake. However, in this analysis, patients treated with darifenacin had a significant decrease in voiding frequency and an increase in bladder capacity (volume voided), effects that are presumably related to inhibition of urgency. Indeed, the present analysis showed a significant decrease in both the number of episodes and average severity of urgency during darifenacin therapy, with responder rates for the reduction in number of urgency episodes showing significant differences from placebo for both doses of darifenacin at all levels of response analysed. This is a key finding, given that urgency, in view of its unpredictability, is a particularly troublesome aspect of OAB. Despite the potential for reducing the fear of an incontinence episode occurring, the efficacy of antimuscarinic agents for improving this symptom has been largely ignored in the medical literature.

The reduction in incontinence episodes, urgency and frequency after treatment with darifenacin clearly has a positive effect on the number of dry days experienced by patients. As there is currently no ‘cure’ for OAB, the improvement in the number of dry days that would be acceptable to a patient on treatment has not yet been established. However, the dose-related significant increase in the proportion of patients who achieved ≥3 dry days per week suggests that darifenacin has a significant effect on this debilitating symptom, with almost a quarter of patients on 15 mg achieving a level of continence close to normal (≥7 consecutive dry days). The improvements in symptoms observed in this pooled analysis were paralleled by improvements in quality of life, assessed for the same study population using the King's Health Questionnaire, as reported elsewhere [11], confirming that the benefits for patients were indeed clinically meaningful.

Successful antimuscarinic therapy for OAB, which in some patients may mean life-long treatment, requires the achievement of an acceptable balance between positive effects on the bladder and undesirable AEs. However, with existing agents poor tolerability and safety have often compromised the patients’ satisfaction with treatment, leading many to discontinue therapy within a few months [12]. In the present analysis, both doses of darifenacin were effective in improving OAB symptoms; furthermore, although the incidence of AEs was higher with active treatment, the AE-related withdrawal rates and need for corrective measures (e.g. laxative use) for both doses of darifenacin were not dissimilar to placebo. This suggests that AEs during treatment with darifenacin 7.5 mg and 15 mg once daily, most commonly dry mouth and constipation, are manageable and well tolerated. For example, in the case of dry mouth one possible explanation is that several muscarinic receptor subtypes are involved in salivary secretion, with M3 receptors having a functional role in producing low-viscosity secretions by serous cells, and both M3 and M1 subtypes being functional in the production of lubricating secretions by mucous cells [4]. This is exemplified by findings in the M3‘knockout’ mouse, in which pilocarpine-stimulated salivary secretion is suppressed, yet the oral cavity remains lubricated [13]. As such, selective blockade of M3 receptors would be expected to affect salivary secretion, but with less effect on M1 receptors, a significant proportion of lubricating function should remain. This probably explains why reports of dry mouth were typically rated as being mild to moderate and were infrequent reasons for discontinuation during darifenacin therapy. Similarly, low rates of constipation-related withdrawal during treatment with darifenacin at both doses are probably explained in that numerous mechanisms within the enteric nervous system contribute to physiological function of the gastrointestinal tract, which could partially mitigate the effect of M3 receptor antagonism.

In view of its M3 receptor selectivity and associated M1/M2-sparing profile, agents such as darifenacin are expected to minimize the risk of AEs associated with blockade of non-M3 receptor subtypes, e.g. cognitive impairment (primarily M1-mediated) and cardiac effects (primarily M2-receptor-mediated) [4]. Indeed, the potential for impairment of cognitive function is becoming increasingly recognized among less M3-selective antimuscarinics such as oxybutynin [14,15] and nonselective antimuscarinics such as tolterodine [16]. Furthermore, in a recent study, the long-term use of antimuscarinic agents, including oxybutynin, by patients with Parkinson's disease was associated with exacerbation of pathologies associated with Alzheimer's disease [17], raising concerns about the use of nonselective agents by the elderly population with OAB in particular. Such safety problems are difficult to manage other than by discontinuing treatment, which denies the patient an effective therapy for their OAB symptoms. In the present analysis the incidence of nervous system and cardiovascular AEs in the groups on darifenacin was similar to that for placebo. Such findings are consistent with phase I and II data showing that darifenacin is not associated with impairment of cognitive or cardiac function [18,19], as expected from its M3 selective, M1/M2-sparing profile. Whether the results are similar with newer agents such as SVT-40766 and solifenacin (for which clinical efficacy has recently been reported [20]) will be interesting. Both of these agents show selectivity for M3 over M2 receptors, but relatively low selectivity for M3 over M1 receptors [21,22]. On the other hand, trospium is a nonselective antimuscarinic agent and is thought to have a lower propensity for penetration of the blood–brain barrier in view of its physicochemical properties [23]. Although CNS AEs have been noted with older antimuscarinic agents, and newer agents may have a reduced propensity for these effects, to date only darifenacin has been the subject of published cognitive function studies.

In conclusion, the findings of this pooled data analysis of three phase III studies show that darifenacin, an M3 SRA, provides a significant improvement in the debilitating symptoms of OAB. As predicted by its M3 selectivity and M1/M2-sparing profile, both doses of darifenacin were well tolerated with no CNS or cardiovascular safety concerns.

ACKNOWLEDGEMENTS

The studies were funded by Pfizer Inc. Preparation of the manuscript was supported by an educational grant from Novartis Pharma AG. Editorial and project management services were provided by Thomson ACUMED®. We are grateful for the support of our co-investigators in the studies:

Argentina: ON Contreras Ortiz, OH Damia, O Mazza; Australia: PL Dwyer, RA Gardiner, RJ Millard, J Stanley, BGA Stuckey, PD Sutherland, AGS Tulloch; Belgium: H Cammu, A Demelenne, K Everaert, R-J Opsomer, J-J Wyndaele; Brazil: H Brusschini, C Teloken; Canada: R Casey, J Corcos, HP Drutz, M Elhilali, S Herschorn, A Morales, E Schick, LM Tu, L Valiquette; Czech Republic: J Hiblbauer, P Verner, F Zatura; Denmark: KM Bek, K Glavind, G Lose, J Nordling, WF Rasmussen, M Rudnicki, S Walter; France: F Haab; Germany: W Dierkopf, T Kränzlin, R Landthaler, H Litschmann, G Merder, M Zellner, TD Zöller; Hungary: T Ali, Z Fábos, I Romics, F Törzök; Israel: J Abarbanel, G Gillon, D Gordon, R Langer Y Vardi; Italy: P Di Benedetto, J Golomb, O Nativ, E Usai; Lithuania: VM Cigriejiene, KK Jocius, H Ramonas; Mexico: M Del Pilar, Velazquez, A Sedano; the Netherlands: JH Bosch, GA Smits; Norway: M Andersen, JH Blom, E Borstad, A Gustavsen, I Hoye, M Kristoffersen, JJ Kums G Overland, A Schultz; Poland: A Borkowski, A Dobrowolski, T Engebretsen, P Jarzemski, K Krajka, A Majek, E Miekos, A Prajsner, T Rechberger, Z Salska; South Africa: L Aldera, IH Breytenbach, FJ Van Wijk; Sweden: L Angelsiöö, M Ankardal, C Falconer, A Fianu-Jonasson, L Henriksson, P Kjolhede, G Holmberg, L Lanner, J Pedersen, G Tegerstedt; UK: L Cardozo, CR Chapple, CG Fowler, S Hill, A Joyce, V Khullar, J Maroni, AR Mundy, DE Neal, J Robinson, M Salman, MC Slack, MJ Speakman, LH Stewart; USA: R Ackerman, JP Antoci, SM Auerbach, HJ Bashein, TD Beam, SA Benjamin, BH Blank, RB Bracken, A Burnett, WM Cohen, G Faerber, JE Foote, SJ Freedman, PC Ginsberg, DM Gleason, HB Goldman, TL Griebling, DV Jablonski, RJ Kahnoski, LI Karsh, JM Kaufman, DD Kidd, IW Klimberg, CG Klutke, LD Knoll, RM Kroeger, G Leach, DU Lipsitz, LK Lloyd, KT McVary, DF Mobley, WG Moseley, MI Murdock, BP Murphy, PA Norton, TC Regan, SF Richardson, EJ Sacknoff, KC Shandera, PJ Shenot, R Salant, WD Steers, CP Steidle, ML Wiatrak, S Vick, PJ Vitale, BH Wachs, JM Young, CK Yuen, EJ Zeidman.

CONFLICT OF INTEREST

All authors are investigators in the study and/or have acted as consultants to Pfizer or Novartis.

Ancillary