Refers to article:
Systematic Review of Combination Drug Therapy for Non-neurogenic Male Lower Urinary Tract Symptoms
Accepted 15 January 2013
August 2013 (Vol. 64, Issue 2, pages 228 - 243)
A clinically significant drug response rarely achieves adequate disease control, resulting in the need for combination pharmacotherapy. Virtually every prevalent chronic condition including hypertension, coronary artery disease, hyperlipidemia, chronic obstructive pulmonary disease, rheumatoid arthritis, and diabetes are commonly treated with combination drug regimens. Non-neurogenic male lower urinary tract symptoms (NNMLUTS) should be no exception. α-Blockers, 5α-reductase inhibitors (5-ARIs), antimuscarinics (anti-Ms), and phosphodiesterase (PDE) inhibitors are effective monotherapies for subsets of NNMLUTS. These monotherapies rarely “cure” LUTS. Therefore, combining these agents is the next obvious step in pharmacologic management of NNMLUTS.
Füllhase et al.  examined the medical evidence supporting combination medical treatment of NNMLUTS. The authors are to be congratulated for their comprehensive and scholarly systematic review. Unfortunately, there are at least three fundamental limitations of the medical literature for NNMLUTS combination therapy. First, there is a paucity of randomized, double-blind, placebo-controlled trials for the majority of the combinations. Second, there are no cost-effectiveness studies to assess whether observed benefits justify the added cost. Third, the definition of NNMLUTS is unclear. The differential diagnosis of NNMLUTS includes a multitude of non-neurogenic factors that contribute to lower urinary tract symptoms (LUTS). For several of the α-blocker/5-ARI studies identified in the review, NNMLUTS includes clinical benign prostatic hyperplasia (BPH). For other α-blocker/5-ARI studies, NNMLUTS includes clinical BPH and large prostates. For most α-blocker/anti-M studies, NNMLUTS includes clinical BPH refractory to α-blockers. For the α-blocker/PDE studies, NNMLUTS includes BPH and erectile dysfunction. Every combination examined by Füllhase et al. defines a unique cohort of men with NNMLUTS.
I have been an active participant in the evolving story of medical therapy since 1983, when I characterized the human prostatic α1 adrenoceptors. I will use this editorial as an opportunity to express my perspective about the role of combination medical therapy for BPH (NNMLUTS) based on my interpretation of the medical evidence with, at times, deviation from the conclusions of Füllhase et al. . I will limit my comments to the α-blocker plus 5-ARI and α-blocker plus anti-M combinations because the medical evidence for all other combinations is deficient. I will also limit my analysis to LUTS and disease progression due to space limitations.
1. α-Blockers plus 5α-reductase inhibitors
Füllhase et al.  cite three randomized studies (two placebo controlled) that compared an α-blocker, a 5-ARI, and the combination of an α-blocker plus a 5-ARI (references 15, 17, and 22 in the review). All three studies enrolled men with clinical BPH independent of prostate volume. None of the studies showed any advantage of combination over α-blocker alone. In the two placebo-controlled studies (references 15 and 22 in the review), combination therapy was not significantly different than α-blocker alone, and the 5-ARI was not significantly different than placebo. There was no advantage to combination because the 5-ARI was totally ineffective as a monotherapy at alleviating LUTS.
Füllhase et al.  agree that these 1-yr studies failed to show a benefit of combination over α-blocker monotherapy at relieving LUTS. However, they claim that combination therapy was ultimately found to be superior to monotherapy at relieving LUTS in both the long-term Medical Therapy of Prostatic Symptoms (MTOPS) study (reference 19 in the review) and the Combination of Avodart and Tamsulosin (CombAT) study (reference 28 in the review). Let's examine these study designs before interpreting the data. The MTOPS trial was sponsored by the US National Institutes of Health, included a placebo arm, and enrolled men with BPH independent of prostate volume. The CombAT trial was supported by the company promoting the 5-ARI arm, did not include a placebo, and enrolled men with prostate volumes > 30 ml. The mean prostate volume in CombAT was almost 80% greater than that in MTOPS. Therefore, CombAT is relevant to the subset of men with large (or very large) prostates. The primary end point for both the MTOPS and CombAT trials was disease progression, not LUTS improvement. The individual treatment groups in the MTOPS and CombAT trials had approximately 750 and 1600 subjects, respectively, which is grossly overpowered for ascertaining improvements in symptom scores. Füllhase et al. conclude that MTOPS demonstrated a statistically significant long-term advantage of combination therapy over α-blocker alone for improving LUTS. The treatment-related advantage of combination over α-blocker was only 0.8 unit of the American Urological Association symptom score, which I interpret as clinically insignificant. CombAT showed a more robust long-term advantage of combination over α-blocker, but the results are applicable only to men with large or very large prostates.
Both MTOPS and CombAT were designed to investigate disease progression. In MTOPS, all treatment groups significantly decreased overall disease progression relative to placebo, and combination therapy was significantly more effective than the monotherapies. The α-blocker and 5-ARI were equally effective at preventing LUTS progression, which accounted for 80% of the progression events. The 5-ARI was significantly more effective at preventing acute urinary retention (AUR). The observed 66%, 64%, 81%, and 67% risk reductions of combination therapy for overall clinical BPH progression, symptom progression, development of AUR, and progression to invasive therapy of BPH, respectively, have been used to justify combination therapy. An analysis of the numbers needed to treat to prevent disease progression puts these impressive risk-reduction outcomes into clinical perspective. Overall, 12, 56, and 29 men with clinical BPH would need to be treated with combination therapy for a mean of 4.5 yr to prevent a single episode of symptom progression, AUR, or invasive therapy for BPH, respectively . If one assumes that an α-blocker is the first-line treatment for clinical BPH (a reasonable assumption because all combination therapies examined by Füllhase et al.  include an α-blocker), then the addition of a 5-ARI to 36, 151, and 54 men on an α-blocker for a mean of 4.5 yr would prevent a single episode of symptom progression, AUR, or invasive treatment for BPH, respectively.
CombAT defined disease progression as development of AUR or progression to invasive treatment. Based on the results of the CombAT trial, 30 and 18 men with large or very large prostates on an α-blocker would need to be treated with combination therapy for 4 yr to prevent a single episode of AUR or invasive BPH surgery, respectively .
Füllhase et al.  cited the SMART-1 trial in which 327 men with clinical BPH were treated with the combination of dutasteride and tamsulosin for 24 wk, followed by a randomized, placebo-controlled withdrawal of the tamsulosin for an additional 12 wk (reference 32 in the review). In fairness, why was there no arm withdrawing the dutasteride? Simply because this is another combination trial supported by the promoter of the 5-ARI arm (dutasteride). The prostate volume was not reported; however, the baseline mean serum prostate-specific antigen level of 4.3 is higher than in any other 5-ARI study and suggests an even greater propensity to enroll men with very large prostates. The inherent bias of this study, in my opinion, precludes meaningful conclusions.
My conclusion regarding the advantages of combination α-blocker plus 5-ARI differs from that of Füllhase et al. . I interpret the data to show that in the entire cohort of men with clinical BPH, the addition of a 5-ARI to an α-blocker has negligible clinical benefit in the short term or in preventing disease progression. In the cohort of men with clinical BPH and large or very large prostates, there is a modest clinical benefit in adding a 5-ARI to an α-blocker, for both short-term control of LUTS and long-term disease progression. A cost-effectiveness assessment of this modest benefit would likely limit enthusiasm for combination therapy in most cases with prostate volumes > 30 ml.
2. α-Blocker plus antimuscarinic
Füllhase et al.  cite several studies in which men with NNMLUTS on an α-blocker were randomized to anti-M versus placebo. This study design selects the subset of men with clinical BPH whose storage LUTS are refractory to α-blockers.
The Tolterodine and Tamsulosin in Men With LUTS Including OAB: Evaluation of Efficacy and Safety (TIMES) study (reference 42 in Füllhase et al. ) was a 12-wk, multicenter, randomized, placebo-controlled study comparing the safety and efficacy an α-blocker, an anti-M, a combination of these drugs, and placebo in 879 men with overactive bladder (OAB) and BPH. At 12 wk, the International Prostate Symptom Scores of the α-blocker and the combination were identical and were significantly lower than placebo, indicating that combination therapy is no better than α-blocker monotherapy at relieving LUTS in men with OAB and BPH. The percentages of men qualitatively exhibiting an improvement in LUTS in the placebo, α-blocker, anti-M, and combination groups were 62%, 65%, 71%, and 80%, respectively. Only the proportion of men qualitatively rating their symptoms as improved in the combination group was significantly greater than the placebo group.
The study failed to show a greater qualitative symptom improvement for combination relative to α-blocker alone (reference 42 in Füllhase et al. ). Ad hoc analysis demonstrated that in men with prostate volumes > 29 ml, combination therapy significantly reduced 24-h urinary frequency by 1.1 relative to α-blocker, and in prostates < 29 ml, the advantage of combination therapy was 0.6 over the α-blocker (reference 47 in the review). First, post hoc analysis has inherent limitations. Second, I do not interpret these post hoc analyses to show a robust advantage of combination therapy over α-blocker alone. I agree with the authors that a 12-wk interval is inadequate to assess the risks of combination α-blocker plus anti-M in the subgroup of men with NNMLUTS who have BPH and OAB.
Despite its inherent limitations, how do I interpret the medical evidence for combination therapy for NNMLUTS? For men with clinical BPH, with clinical BPH and large or very large prostates, and with clinical BPH and OAB, an α-blocker should be first-line therapy for relief of LUTS. In men whose LUTS is improved but bothersome, a 5-ARI or an anti-M should be added for men with clinical BPH and large prostates or with clinical BPH and OAB, respectively. In those men with clinical BPH and large prostates (>50 ml), one should consider adding a 5-ARI to reduce the risk of progression to AUR.
Conflicts of interest
Dr. Lepor is a consultant for Watson Pharmaceuticals.
Department of Urology, New York University School of Medicine, NYU Langone Medical Center, New York, NY, USA
Department of Urology, New York University School of Medicine, NYU Langone Medical Center, 150 East 32nd Street, Second Floor, New York, NY 10016, USA. Tel. +1 646 825 6340.
© 2013 European Association of Urology, Published by Elsevier B.V.