Ichihara K, Masumori N, Fukuta F, Tsukamoto T, Iwasawa A, Tanaka Y
J Urol 2015;193:921–6
Ichihara et al randomized 94 patients with benign prostatic obstruction and persistent urgency while on tamsulosin treatment to receive either tamsulosin (0.2 mg/d) or tamsulosin plus mirabegron (50 mg/d) for 8 wk in a multicenter open-label study. They report that combination treatment was associated with greater improvements in overactive bladder symptom score, the urinary urgency and daytime frequency part and storage subscore of the International Prostate Symptom Score, quality of life index, and postvoid residual urine. They concluded that such combination treatment is effective and safe in men with prostatic obstruction who have overactive bladder symptoms after tamsulosin monotherapy.
It may be argued whether tamsulosin was given at sufficiently high doses to reach full effects and whether the study was underpowered; however, it illustrates a growing interest in combination treatment of lower urinary tract symptoms (LUTS), whether with α-blockers in combination with 5α-reductase inhibitors , muscarinic antagonists  or phosphodiesterase type 5 inhibitors  or with muscarinic antagonists in combination with β3-adrenoceptor agonists . More often than not, combination treatment appeared more effective than monotherapy in such studies. While adding a second drug to monotherapy may indeed be helpful in patients experiencing insufficient LUTS improvement on monotherapy, parallel-group studies such as all of the above may be inadequate to prove such benefit.
Each of the above drug classes has a different mechanism of action and hence is likely to have a different set of responders and nonresponders. Consider hypothetical drugs A and B with a responder rate of 50% each (not unusual in functional urology once the placebo responder rate has been subtracted) and a symptom score reduction by 6 and 0 points in responders and nonresponders, respectively. In that situation, each monotherapy would reduce the symptom score by 3 points at the group level; if their responder groups do not overlap, combination treatment could reduce symptoms by 6 points, even if not a single patient benefits from the combination. This simply reflects that parallel-group studies cannot distinguish whether an observed symptom improvement at the group level comes from a benefit of combination treatment in individual patients or simply reflects that combination treatment hits two distinct, perhaps overlapping groups of responders. For this reason, we urgently need adequately powered crossover studies to understand whether there indeed is a group truly benefitting from combination treatment at the individual-patient level.