Hypothesis / aims of study
Benign prostate hyperplasia (BPH) may lead to lower urinary tract symptoms. Patients with moderate-to-severe symptoms usually start upfront pharmacotherapy (e.g., alpha-blockers, 5-alpha reductase inhibitors, or combination). Pharmacotherapy doesn't necessarily cure BPH & patients may require subsequent surgical interventions such as transurethral resection of the prostate (TURP). An alternative to TURP & pharmacotherapy, Greenlight laser photoselective vaporization of the prostate (GL-PVP), has lower costs compared to TURP & faster symptomatic improvement compared to pharmacotherapy. The purpose of this study was to evaluate the cost-utility of upfront pharmacotherapy (i.e., alpha-blockers or 5-ARI or combination) followed by delayed surgical intervention (i.e., TURP, GL-PVP) for those who fail, compared to receiving an upfront surgical intervention.
Study design, materials and methods
The target population were men with moderate-to-severe symptoms & no contraindications for BPH surgery. Microsimulation was used to model the progression of BPH symptoms, cost projection & quality-adjusted life-years (QALYs). Pharmacotherapy costs were obtained from the Ontario Drug Benefit Formulary. BPH surgeries costs were collected retrospectively. All other outcomes were obtained from the literature. Cost-utility analysis used a Canadian public payer perspective, a life-time time horizon, a discount rate of 1.5% & a willingness-to-pay threshold of $50,000 per QALY. Probabilistic sensitivity analysis (PSA) was performed to estimate parameter uncertainty.
Compared to the upfront pharmacotherapy options, upfront surgical interventions were, on average, more costly & more effective with the incremental cost per QALY gained ranging by drug type from $2,138 to $2,911 for upfront GL-PVP & $5,473 to $6,646 for upfront TURP options. Compared to upfront TURP, upfront GL-PVP was associated with lower costs ($9,468 vs. $11,562) & a marginally lower effectiveness (15.20 vs. 15.24 QALYs) translating to an incremental cost per QALY of $53,417 more gained. PSA indicated that upfront GL-PVP would be cost-effective 47% of the time at a threshold of $50,000/QALY.
Interpretation of results
Sensitivity analysis with a multiplicative approach of utilities for joint health states and a probabilistic analysis demonstrated the robustness of the model. However, the limitation of statistical modelling may introduce bias. Data on clinical effectiveness, adverse events and adherence were mainly obtained from randomised control trials. The study population in trials may not be representative of the real-world BPH/LUTS population and adherence to long-term pharmacotherapy maybe lower in the real-world clinical setting than assumed in our model.