Persistence with mirabegron or antimuscarinic treatment for overactive bladder syndrome: Findings from the PERSPECTIVE registry study

Abstract Objectives This analysis from the PERSPECTIVE (a Prospective, Non‐interventional Registry Study of Patients Initiating a Course of Drug Therapy for Overactive Bladder) study evaluated treatment persistence with mirabegron or antimuscarinics over a 12‐month period. Methods Participants were adults diagnosed with overactive bladder (OAB) by their health care provider (HCP), who were initiating mirabegron or antimuscarinic treatment. The HCP made all treatment decisions, and patients were followed for 12 months with no mandatory scheduled visits. Information requests were sent to patients at baseline and months 1, 3, 6, and 12. Patients were nonpersistent if they switched, discontinued, or added OAB medications/therapies to their initial treatment. Reasons for discontinuation and switching patterns were investigated. Results Overall, 1514 patients were included (613 mirabegron and 901 antimuscarinic initiators). Persistence rates decreased steadily over time in both groups. A low proportion of patients added or switched medication at each time point. Unadjusted Kaplan‐Meier analysis showed similar persistence rates for both groups. When the data were adjusted for patient characteristics (age, sex, and OAB treatment status), mirabegron initiators had higher persistence rates. No significant differences were noted in unadjusted median time to end of persistence. However, end of treatment persistence by any cause was longer with mirabegron (median: 9.5 vs 6.7 months for antimuscarinics). HCPs stated that the most common reasons for nonpersistence were no symptomatic improvement and side effect aversion. Conclusions Treatment persistence was longer for mirabegron compared with antimuscarinic initiators after controlling for patient characteristics. End of treatment persistence by any cause was also longer with mirabegron.


| INTRODUCTION
Overactive bladder (OAB) is a chronic symptom syndrome affecting more than one in ten North American adults, although the condition is particularly common among the elderly. [1][2][3][4] Serious ramifications on the quality of life (QoL) and daily living of affected patients have been noted, 5,6 and the treatment of OAB is estimated to cost over $100 billion in the US each year. 7 Pharmacotherapy approaches are advocated for OAB symptom management if behavioral and conservative measures do not produce adequate improvement. 8,9 Owing to their established efficacy and propensity to improve QoL, oral antimuscarinics and the β 3adrenoreceptor agonist, mirabegron, are the recommended principal pharmacologic interventions in North America for patients with OAB symptoms. [9][10][11][12] Although both types of agent are approved as first-line therapeutic agents, 9 patients typically receive antimuscarinics first in clinical practice, and, as such, patients who receive mirabegron tend to be more treatment experienced than patients treated with antimuscarinics. [13][14][15] However, the clinical utility of antimuscarinic agents is limited by the occurrence of certain adverse events (AEs), including dry mouth, constipation, and urinary retention. 16 Owing to its different mechanism of action, mirabegron is associated with a lower frequency of these AEs and is thought to have a more favorable safety profile than antimuscarinics. [16][17][18][19] AEs that have been commonly reported in mirabegron trials include hypertension, nasopharyngitis, and headache, although similar frequencies of these events were also observed in the placebo groups. [20][21][22] Clinical investigations have shown that patients who continue to persist with OAB medication typically experience significantly improved urinary symptoms and QoL. 12,23,24 However, despite the objective and patient-reported benefits associated with pharmacological agents as treatment modalities for OAB symptoms, persistence with these medications consistently decreases over time. 14,15,25 Furthermore, a retrospective pharmacy claims analysis found that oral antimuscarinics specifically exhibit relatively poor adherence and persistence compared with medications used for other common chronic conditions. 26 Patient-reported reasons for discontinuing antimuscarinic use include the medication not working as expected or because of the occurrence of side effects. 27 In fact, a UK prescription data analysis found that between 65% and 86% of patients discontinue therapy within 1 year depending on the antimuscarinic prescribed. 28 However, real-world data collected from retrospective administrative claims databases and observational studies suggest that treatment persistence with mirabegron may be greater than for antimuscarinics in patients with OAB. 15

| Statistical analyses
Statistical data were analyzed with SAS (version 9. Adding or switching medication was categorized according to the initial OAB treatment and the addition or switch to a second-line OAB medication type. This included adding/switching to mirabegron, adding/switching to antimuscarinics, discontinuation of initial treatment, or switching to onabotulinumtoxin A or sacral neuromodulation.
Rates of switching, adding on, or discontinuing were calculated as: 3 | RESULTS

| Study population
The study population data have been reported previously. 31  concomitant treatment at baseline with α 1 -adrenoreceptor antagonists and 5α-reductase inhibitors, respectively (Table S1).

| Persistence
The unadjusted persistence rates decreased steadily in both the mirabegron and antimuscarinic groups (Table 1) (Table S2).
The unadjusted Kaplan-Meier analysis showed that patients initiating mirabegron or antimuscarinics had similar rates of persistence during follow-up ( Figure 1A). However, when the data were adjusted for baseline age, sex, and prior OAB medication use, the persistence curves for initial mirabegron or antimuscarinic treatment separated by month 2 of the study, with mirabegron initiators showing higher rates of persistence ( Figure 1B).
Although low numbers of patients were included, no statistically significant differences were noted between the mirabegron and antimuscarinic initiators in terms of unadjusted median time to end of persistence for all patients and for each of the subgroup analyses ( Analysis of the end of treatment persistence by any cause data showed that clear separation was observed in favor of mirabegron (median time to end of treatment persistence: 9.5 months vs 6.7 months for antimuscarinics; Figure 2A). This observation may be due to antimuscarinic initiators switching treatment (median time: 4.1 months vs 8.5 months for mirabegron; Figure 2B) or adding a treatment (median time: 6.5 months vs 12 months for mirabegron, Figure 2C) sooner than mirabegron initiators. In contrast, mirabegron initiators who stopped all treatment during follow-up did so sooner than antimuscarinic initiators (median time: 6.1 months vs 9.1 months, respectively; Figure 2D).
Data from the adjusted Cox proportional hazard model also showed that initial treatment had no effect on time to end of persistence ( Figure 3)

| DISCUSSION
In this secondary analysis from the PERSPECTIVE registry, patients with OAB initiating a new course of mirabegron persisted with treatment longer than those initiating antimuscarinics after controlling for differences in patient characteristics between groups. End of treatment persistence by any cause was also longer with mirabegron than with antimuscarinic treatment. It is important to note, however, that no statistical analyses were conducted on these specific results. Furthermore, no additional notable differences in persistence were observed between the two groups.
In agreement with the specific persistence findings mentioned above, previous real-world analyses conducted in Canada, Japan, Spain, the US, and the UK typically showed that patients treated with mirabegron remained on treatment longer than those treated with antimuscarinics. [13][14][15][32][33][34] In these previous studies, 12-month persistence rates of between 14% and 38% were noted for mirabegron, and rates of between 3% and 25% were reported with antimuscarinics. Conversely, the opposite finding was noted in a Japanese urology clinic study, which reported 12-month persistence rates of 12.2% and 20.1% with mirabegron and solifenacin, respectively. 35 It is important to note, however, that the Japanese urology clinic study was a prospective, randomized trial when the other studies mentioned above were retrospective database analyses, and these differences in study design may partially explain the contradictory This investigation showed that treatment persistence was longer with mirabegron when the Kaplan-Meier results were adjusted for baseline age, sex, and prior OAB medication use. However, no difference was noted in the unadjusted results. These findings are potentially due to the fact that mirabegron initiators were more likely to be male and treatment experienced than their antimuscarinic counterparts. In agreement with these findings, previous persistence studies typically showed that higher proportions of males and treatmentexperienced patients were noted in mirabegron compared with antimuscarinic cohorts. [13][14][15]25 In this study, the age of the patients did not appear to have an effect on the treatment that was initiated.
Furthermore, the fact that a greater proportion of mirabegron initiators were treatment experienced may partially explain why end of treatment persistence by any cause was longer in this group than in the antimuscarinic initiators group. The patients who initiated mirabegron may have therefore been more familiar with the efficacy of OAB medication and the potential adverse effects associated with treatment.
No statistically significant differences in the persistence results were observed in PERSPECTIVE. This finding was potentially due to the study design as there was no requirement for patients to attend scheduled visits during the follow-up period. The lack of statistically significant differences in persistence may explain why similar patientreported OAB symptom bother and HRQoL results were observed for both treatments in the present study. 31 In our investigation, older, female, and treatment-naïve patients were more likely to persist with treatment for a longer period of time.
The finding that older patients persisted with OAB therapy longer than younger patients has been previously reported in several persistence studies involving mirabegron and antimuscarinics. 13,15,25 However, a further study found no relationship between patient age and OAB treatment persistence, 14 and a UK-based hospital prescription analysis found that younger age was associated with improved persistence with mirabegron therapy. 36 Furthermore, previous investigations support the results of this study in so much as female patients were found to persist with OAB medication for a longer period of time than their male counterparts. 13,37 Conversely, a UK clinical practice study found no difference in persistence rates between the sexes, 14 and a Japanese real-world analysis study found that males typically persisted with mirabegron and antimuscarinic therapy longer than females. 15 In contrast to the present study, previous persistence studies have typically found that treatment-experienced patients persist with mirabegron and antimuscarinic therapy longer than treatment-naïve patients. [13][14][15]25 The discrepancies between our findings and the results of previous investigations may be partially due to the design of the studies. The previous studies were database analyses, while the present study was a clinical practice investigation. In support of this hypothesis, a retrospective, real-life study found that treatment-naïve patients persisted with mirabegron treatment longer than patients who had received previous anticholinergic therapy. 38 The switching analyses showed that a low percentage of patients   41 Furthermore, a Japanese single-center retrospective study found that the biggest reasons for discontinuing mirabegron treatment were unmet treatment expectations, the occurrence of AEs, and symptom improvement. 42 In our study, the proportion of patients who discontinued following no improvement in their symptoms was slightly higher in the mirabegron group, whereas the proportion of patients who discontinued due to side effects was higher in the antimuscarinics group. The latter finding is potentially due to the higher incidence of anticholinergic side effects that are noted with antimuscarinics compared with mirabegron. 16 Despite the promising results obtained in the present study, long-term treatment of a chronic disease such as OAB represents a challenge for patients in terms of adhering to and persisting with prescribed medication. The decline in patient persistence in both treatment groups over time emphasizes the importance of using strategies that improve patient treatment experiences. For example, patientsupport programs have been shown to improve persistence with and/or adherence to medications for the treatment of other chronic diseases. [43][44][45] Indeed, an initial study found that patients with OAB who followed a patient-support plan found that it was informative and feasible to implement and that they were satisfied with several aspects of the plan. 46 The widespread use of support plans could therefore prove beneficial in this setting to improve treatment persistence.
Owing to the varied population involved, we believe that this study more accurately reflects the real-world situation compared with a clinical trial population, where specific inclusion and exclusion criteria have to be satisfied for enrollment. In addition, this investigation was a prospective study and therefore does not suffer from the potential data deficiencies that are inherent with retrospective database investigations. However, this study does have some limitations.
As specific scheduled follow-up visits were not included as part of the study protocol, the quantity of persistence data available throughout the study varied according to the follow-up time point. Furthermore, data on the reasons for discontinuing, switching, or adding on a treatment were only available for a minority of patients. Additionally, no data were captured on whether the patients actually took the medication. Therefore, patients could have been identified as persisting with treatment when they were not taking the medication as prescribed by their HCP. Lastly, if a patient diary had been used during this study, it may have increased patient awareness about the current status of their condition and may have therefore led to increased persistence with the study medication.
In conclusion, the present study is the first observational study conducted in North America to investigate patient persistence with OAB medication following treatment with either mirabegron or antimuscarinics. Specific results from this study showed that patients who received mirabegron persisted with treatment for a longer period of time than the patients who initiated antimuscarinics, although no statistical differences were observed. Regardless of the treatment used, approximately two-thirds of patients were still persisting with their initial treatment after 12 months. These high rates of persistence could lead to symptomatic improvements in patients with OAB.

ACKNOWLEDGMENTS
The authors would like to thank the PERSPECTIVE study investigators and all patients who took part in the study. This study was funded by