Patient‐reported outcomes in men with lower urinary tract symptoms (LUTS) due to benign prostatic hyperplasia (BPH) treated with intraprostatic OnabotulinumtoxinA: 3‐month results of a prospective single‐armed cohort study

Study Type – Therapy (prospective cohort)


OBJECTIVE
• To evaluate patient-reported and objective outcomes after intraprostatic injection of OnabotulinumtoxinA (BTX-A) in men with lower urinary tract symptoms (LUTS) due to benign prostatic hyperplasia (BPH).

PATIENTS AND METHODS
• A prospective single-armed cohort study was designed. • Patients diagnosed with LUTS due to BPH and unsatisfactory response to medical therapy, were recruited between November 2010 and July 2011.

RESULTS
• Of 75 screened patients, 64 with a mean ( SD ) age of 63 (9.3) years were available for the outcome assessment.

CONCLUSIONS
• Intraprostatic injection of BTX-A in men with LUTS due to BPH provides clinically signifi cant short-term subjective and objective benefi t. • Increasing severity of baseline LUTS appears moderately associated with the patient-perceived benefi t from the treatment.
• Although the non-randomised design and short-term assessment limit the level of evidence of our study, intraprostatic BTX-A seems a promising, safe and minimally invasive option for patients with BPH with unsatisfactory response to standard drug therapy.

KEYWORDS
botulinum toxin , lower urinary tract symptoms , benign prostatic hyperplasia , patient-reported outcome , patient ' s satisfaction Study Type -Therapy (prospective cohort) Level of Evidence 3a Patient-reported outcomes in men with lower urinary tract symptoms (LUTS) due to benign prostatic hyperplasia (BPH) treated with intraprostatic OnabotulinumtoxinA: 3-month results of a prospective single-armed cohort study INTRODUCTION BOO secondary to BPH with benign prostatic enlargement represents the most common cause of LUTS in older men [ 1 ] . Major treatment goals are fast and sustained long-term relief of LUTS, improvement of health-related quality of life (HRQL), prevention of disease progression and complications, and as important, patient ' s satisfaction with treatment. Currently, BPH is commonly treated with pharmacological therapy including α -blockers, which reduce urethral resistance, and 5 α -reductase inhibitors (5-ARIs), which reduce prostate volume. However, these drugs must be taken continuously, cannot always control the long-term urinary symptoms and progression, and are associated to several side-effects [ 2 -4 ] . For patients with severe LUTS, prostate reduction can be obtained surgically by TURP or by minimally invasive surgical techniques. Each surgical option has specifi c adverse consequences and > 20% of patients undergoing TURP do not have satisfactory long-term outcomes [ 5 ] .
Botulinum toxin (BTX) is an emerging, non-invasive and potentially targeted therapy for BPH. BTX blocks the release of neurotransmitters, e.g. acetylcholine and norepinephrine, from pre-synaptic nerves [ 6,7 ] . The therapeutic benefi t of intraprostatic BTX injection is based on its organ-specifi c, chemical denervation leading to inhibition of smooth muscle contraction and tissue atrophy [ 6,7 ] . Animal experiments have suggested that the mechanism of action in the prostate gland is induction of atrophy, diffuse apoptosis, reduced cell proliferation and decreased expression of α -adrenergic receptors [ 8 -10 ] . As a result, BTX may have effects on both the dynamic and the mechanic component of BPH.
Several cohort and small randomised studies evaluating the effects of intraprostatic injection of BTX have been published and recently summarised in systematic reviews [ 11,12 ] . There is substantial agreement on the safety and effectiveness of BTX in improving both LUTS and BOO parameters. However, these studies focused mainly on objective measures and limited the evaluation of subjective health outcomes to the IPSS [ 13 ] . The patients ' opinions about the treatment were given little consideration for assessing treatment outcome. The value of patient-reported outcomes (PROs), such as patients ' perceptions, preferences and satisfaction with therapy, is increasingly accepted as part of the clinical decisionmaking process, and has been increasingly acknowledged in clinical practice guidelines for the management of BPH [ 14 -17 ] . Consequently, we considered it of interest to appraise the PROs of this investigational novel treatment for male LUTS, to provide researchers with pre-trial data relevant for the design of future randomised trials.
The purpose of the present study was to evaluate safety and effi cacy of intraprostatic injection of OnabotulinumtoxinA (BTX-A) by assessing both objective and subjective health outcomes, including the impact on sexual function and patients perceptions about their treatment.

PATIENTS AND METHODS
A prospective, single-armed cohort study was designed. The study was conducted in one tertiary referral centre (Urological Clinic, ' Agostino Gemelli ' Universitary Hospital, Rome, Italy). Ambulatory patients were recruited between November 2010 and July 2011, after signing written research consent. Patients were followed-up for ≥ 3 months after treatment.
The study was conducted in accordance with good clinical practice guidelines and the last version of the Declaration of Helsinki and our local Institutional Review Board approved the protocol. The study has been reported according to the STROBE (STrengthening the Reporting of OBservational studies in Epidemiology) checklist. The companies marketing BTX-A had no role in the funding, design and data analysis of the present study.

BASELINE EVALUATION AND RECRUITMENT OF PARTICIPANTS
Before treatment, all patients underwent a comprehensive urological evaluation, including self-administration of the IPSS [ 13 ] and the International Index of Erectile Function (IIEF) [ 18 ] , DRE, TRUS of the prostate to measure total prostate volume (TPV) and transitional zone volume (TZV), urofl owmetry with postvoid residual urine volume (PVR) determined ultrasonographically, and serum PSA assay (for measurement of total PSA, tPSA).

Inclusion criteria
Men aged 50 -80 years, diagnosed with LUTS due to BPH with unsatisfactory response to combined α -blockers and 5-ARIs therapy for ≥ 6 months, prostate enlargement ( > 30 mL), tPSA level of < 4 ng/mL or of 4 -10 ng/mL but with a biopsy showing no malignancy, and consenting to the study treatment. Patients were considered with unsatisfactory response to medical therapy if they had an IPSS of ≥ 8, IPSS-HRQL item score of ≥ 2, and a maximum urinary fl ow rate (Q max ) of ≤ 15 mL/s with a voided volume of ≥ 150 mL.

Exclusion criteria
A PVR of > 250 mL, previous prostate ablative treatments, therapy with α -blockers for other conditions, neurogenic voiding disorders, prostate or bladder cancer, bladder stones, urethral stricture, chronic bladder catheterisation, prostatitis, myasthenia gravis or other conditions that precluded the administration of BTX.

INTERVENTION
For the treatment, a solution of BTX-A (Botox ® , Allergan, Irvine, CA, USA) was used. With the patient lying in lithotomy position, transperineal intraprostatic injection was carried-out using a 22-G spinal needle under TRUS guidance and local anaesthesia. The dosage of BTX-A was 200 U, diluted in 6 mL saline. Three injections of 1 mL for each lobe were performed. The treatment was administered on a day-hospital regimen. Each treatment took ≈ 20 min. Each patient underwent antibiotic prophylaxis (single tablet of prulifl oxacin 600 mg). Patients discontinued drug therapies for BPH 20 days after treatment, when clinical benefi ts of BTX injection were expected to be achieved according to previous studies [ 11,12 ] .

OUTCOME ASSESSMENT AND MEASURES
The outcome assessment was performed in our Outpatient Clinic at 3 months after treatment, when maximum effects have been achieved according to most previous studies [ 11,12 ] .
The primary outcome effi cacy measure was the improvement of LUTS as measured by a ≥ 4-point reduction in the IPSS.
Secondary outcomes were rate of complications and adverse events, change of HRQL as measured by IPSS-HRQL item score, and change of Q max , PVR, tPSA and IIEF score. Furthermore, patient impression of improvement (Patient Global Impression of Improvement [ PGI-I ] ; 7-point Lickert-type response scale), of treatment satisfaction (PGI of Satisfaction [ PGI-S ] ; 6-point Lickert-type response scale) and of effi cacy (PGI of Effi cacy [ PGI-E ] ; 6-point Lickert-type response scale) were assessed by three questions of a PROs questionnaire (PROs-q; Appendix S1), higher scores indicating better outcome. The questionnaire also asked patients to report if they would repeat the treatment under the same circumstances and if they would recommend the treatment to another person with the same problem ( ' yes ' or ' no ' ).

SAMPLE SIZE
We calculated that a sample of 65 patients would allow detection of a 4-point difference in the IPSS score with a statistical power of 90% and α set at 0.05 (two-sided). This sample size would allow us to test the hypothesis that at least 70% of treated patients will report a positive perception in at least one question of the PRO-q, setting the null hypothesis value at 50%, with a statistical power of 91% and α set at 0.05 (two-sided). Assuming a rate of loss to follow-up of 10%, the study required 72 patients.

STATISTICAL ANALYSIS
All statistical analyses were performed using MedCalc software for Windows. The results are presented as the mean ( SD ). Differences between variables were compared using the Wilcoxon test. The degree of association between variables was evaluated using rank correlation and calculating the Spearman ' s rank correlation coeffi cient ( ρ ) with 95% CI. A correlation analysis was performed to study the association between the satisfaction score (dependent variable) and patients baseline characteristics (covariates). A P < 0.05 (two-sided) was considered to indicate statistical signifi cance.

RESULTS
Of 75 consecutive screened patients, 68 were eligible and treated with intraprostatic BTX-A injection. Four patients were not available for the outcome assessment, leaving 64 patients for the fi nal analysis. Figure 1 shows the study fl ow chart and Table 1 shows the patients ' baseline characteristics.

PRIMARY OUTCOME
The 3-months analysis showed a mean reduction of 49.2% in the IPSS, which decreased from a mean ( SD ) of 19.7 (7.7) to 10.0 (7.1) ( P < 0.001), and a mean reduction of 44.8% in IPSS-HRQL item score, from 4.17 (1.2) to 2.3 (1.6) ( P < 0.001). The statistically signifi cant mean IPSS reduction of 9.7 (8.9) points supports the rejection of the hypothesis of no difference between the IPSS before and after treatment. A more detailed analysis of the IPSS, performed by evaluating separately the voiding and the storage subscores, showed a 48.0% reduction of the IPSS-voiding subscore, from 11.8 (5.4) to 6.1 (5.3) and a 49.1% reduction of the IPSS-storage subscore, from 7.2 (3.1) to 3.9 (3.7) (both P < 0.001).

SECONDARY OUTCOMES
During treatment and follow-up, no local complications or systemic adverse events related to treatment were reported. There was postoperative mild haematuria and stranguria in seven (10.9%) and 14 (21.9%) patients, respectively, and both resolved spontaneously within a few days. No patient had urinary retention or acute prostatitis.
In all, 50 (78.1%) patients would repeat the same treatment under the same circumstances, while 54 (84.4%) patients would recommend the treatment to another person with the same diagnosis. Table 2 shows the univariate correlation analysis between the PGI-S score and the patients ' baseline characteristics. There was a statistically signifi cant positive correlation between PGI-S and both the baseline IPSS and IPSS-HRQL score. As a result, a higher proportion of patients with baseline severe LUTS (IPSS > 28) and low HRQL reported satisfaction with the treatment, compared with patients with moderate (IPSS 20 -27) or mild (IPSS < 19) LUTS and better HRQL ( Fig. 6 ).
The IPSS reduction rate was also positively correlated to the baseline IPSS ( ρ 0.346, P = 0.006). For instance, patients with a baseline IPSS of > 20 (34 patients) had a mean reduction rate of 53.5% (from 25.6 to 13.7 mean score), and patients with baseline IPSS < 20 (30) had a mean reduction rate of 39.7% (from 13.1 to 7.9 mean score). None of the published studies assessed PROs, such as patient-perceived level of improvement and success of therapy, as well as the patient satisfaction with intraprostatic BTX-A therapy. BPH might be defi ned by the LUTS whose severity is not strictly related to objectively measurable parameters and whose impact on patients is heterogeneous and diffi cult to predict [ 15 ] . As a result, evaluations from the patient ' s perspective, using information gathered directly from the treated patients, are of utmost importance in BPH research and management [ 14 -17 ] . PROs, e.g. patient ' s satisfaction with treatment, have been showed to have signifi cant implications in making a full assessment of overall treatment success in patients with BPH. Studying the patient-perceived effi cacy of doxazosin, Cam et al .

DISCUSSION
[ 28 ] reported that, after 1 year, the probability of surgery was signifi cantly higher in patients who considered treatment ineffective than in the those who considered the treatment effective or reported no change ( P < 0.05), and was also signifi cantly higher in those who felt that their condition remained unchanged than in those who considered treatment effective ( P < 0.05).
The present study is the fi rst to provide data useful for gaining insight into the way patients perceive the effect that treatment with intraprostatic BTX has on their symptoms and HRQL. The assessment of patients ' perception of treatment outcome, for patient-reported improvement, satisfaction and effectiveness, showed promising results, comparable with those reported in the studies on α -blockers and 5-ARIs [ 28 -31 ] . A subjective improvement of LUTS was reported by more than half of the present patients and nearly two-thirds of them reported some degree of satisfaction and effectiveness.
In the present study, there were also a not negligible proportion of patients reporting no satisfaction or no improvement from the treatment or even a worsening of their LUTS. Considering that patients had to compare their clinical status at 3 months with that before treatment (when they were still on medical therapy), these results could be explained by an effi cacy of BTX-A similar (in patients reporting no change) or lower (in patients reporting worse symptoms) than that of medical therapy in these patients. In the present study, a correlation, although moderate, between the reduction in IPSS after treatment and more severe baseline symptoms is in agreement with the fi ndings of some previous studies on the effect of tamsulosin [ 32 -36 ] and dutasteride [ 37 ] , and is worthy of further investigation.
We also tried to investigate what drove patient satisfaction. The present data showed an apparent positive correlation between satisfaction with treatment and both the baseline IPSS (moderate correlation) and IPSS reduction from baseline (strong correlation), indicating higher satisfaction in patients with increasing severity of baseline symptoms and more pronounced symptom reduction. Accordingly, a positive correlation between the IPSS reduction and the patient-perceived improvement and satisfaction score has been previously reported in patients treated with dutasteride, tamsulosin or combined therapy [ 30,38 ] . Conversely, in the present study there were no correlations between patient-reported satisfaction and baseline objective parameters, e.g. age, prostate volume, PSA level, Q max and PVR. Overall, these fi ndings suggest a clinical benefi t with BTX-A therapy in all age ranges and also in patients with known risk factors for disease progression, e.g. severe LUTS, large prostate volume, low Q max , high PVR and PSA level [ 39 ] . Together with the quick relief of both voiding and storage symptoms and the improvement of HRQL reported in most studies [ 11,12 ] , these fi ndings may indicate, if confi rmed in large randomised studies, that BTX therapy has the potential to be a treatment option capable of delivering a ' total ' management approach for both low/ intermediate and high-risk patients with BPH [ 40 ] .
According to Silva et al . [ 41 ] , we confi rmed in the present study that intraprostatic BTX-A injection in patients with BPH does not impair erectile, orgasmic or ejaculatory functions and does not change libido. This is of utmost importance because sexual dysfunctions commonly occur in parallel with BPH [ 42 ] and are often associated with α -blockers and 5-ARIs.
The present study does have some limitations. The open-label/non-comparative design, the lack of a placebo arm, and the assessment of PROs with a non-validated questionnaire reduce the generalizability of the present fi ndings and the level of evidence provided. Recall bias also has to be considered when asking patients to compare their present health status with a previous one. In the present study, it was not possible to study the biological effects of BTX-A on human prostate. This is an interesting issue, as little is known about the mechanism of action, which is hypothesised to be related to the induction of apoptosis [ 8,43 ] . The short follow-up of the present study did not allow elucidation of the long-term effi cacy of BTX treatment and we plan to report the longer term results. According to international guidelines, the treatment strategy for BPH involves the sustained improvement of LUTS and HRQL, and the reduction of the risk of acute urinary retention and BPH-related surgery [ 16,17 ] . It would be very interesting to study the long-term effects of BTX therapy, not only on LUTS and HRQL of patients with BPH, but also on the risk of acute urinary retention and BPH-related surgery, especially because the available data suggest that patients are more worried about long-term risks of BPH and prefer therapies that affect long-term disease progression over those that provide short-term symptoms relief [ 14,31,44 ] .
In conclusion, the present study confi rmed that intraprostatic injection of BTX-A is a promising, safe and minimally invasive therapeutic option able to improve both objective and subjective outcomes in patients with LUTS due to BPH with unsatisfactory response to medical therapy.
The study also provided important outcome information from the patient ' s perspective by showing that most of the treated patients reported favourable subjective outcomes at 3-months follow-up. Noteworthy, PROs appeared not to be associated in this cohort with age and objective baseline parameters, but there was a moderate correlation between PROs and increasing severity of baseline LUTS. This could be explained by the subjective nature of both PROs and the IPSS.
The use of BTX-A in the setting of patients with BPH remains off-label in all countries and further basic research and randomised trials are necessary in order to: (i) identify predictive parameters clinically useful for a better selection of patients, (ii) achieve an adequate level of scientifi c evidence supporting the introduction of this treatment in clinical practice, and (iii) defi ne the best treatment technique (dose, injection site, route of administration).