A Propensity Score-Matched Protocol: Terazosin Plus Finasteride For Benign Prostatic Hyperplasia – Sex Hormone And Urinary Control Effects

Summary

The goal of this protocol is to evaluate, using a retrospective propensity score-matched cohort design, whether terazosin combined with finasteride improves long-term urinary control and sex hormone profiles compared with finasteride monotherapy in patients with benign prostatic hyperplasia.

Abstract

To investigate the clinical effect of terazosin hydrochloride combined with finasteride in benign prostatic hyperplasia (BPH), and to analyze its effect on sex hormones and urinary control. This retrospective propensity score‑matched (PSM) cohort study included BPH patients treated between January 2020 and March 2021. 1:1 PSM using logistic regression with prespecified covariates (age, BMI, IPSS, Qmax, PVR, PV, PSA, CRP, T, E2, DHT, incontinence episodes, ICIQ‑SF); inclusion criteria (age 50‑80 years, IPSS ≥ 8, PSA > 1.4 ng/mL, PV ≥ 40 mL, Qmax ≤ 15 mL/s); exclusion of prior prostate surgery or 5α‑reductase/α1‑blocker use; 12‑month treatment and 36 month follow up with outcomes measured at 6, 12, 24, 36 months. Sixty patients were allocated to each group after PSM. The combination group showed significantly greater improvements in IPSS, Qmax, PVR, and PV (all P < 0.05), lower DHT and E2, higher T (all P < 0.05), fewer 24 h incontinence episodes, and better ICIQ‑SF scores (both P < 0.05). Safety profiles were comparable (P > 0.05). The terazosin‑- finasteride regimen is associated with improved long-term outcomes, better urinary control, and favorable hormonal changes, though hormonal effects are primarily attributable to finasteride.

Introduction

Benign prostatic hyperplasia (BPH) is one of the most common urological diseases affecting middle-aged and elderly men. Its incidence increases with age. For example, the prevalence of BPH is about 50% in men aged 60 years and as high as 90% in those aged 81-90 years^1,2. As we age, the prostate increases in size and presses on the urethra, resulting in bladder outlet obstruction (BOO), thereby inducing lower urinary tract symptoms (LUTS), including urinary frequency, urgency, increased nocturia, dysuria, urinary retention, etc^3,4. Such symptoms markedly diminish life quality and could precipitate secondary conditions like impaired bladder contractility, recurrent UTIs, and even compromised renal function⁵. Therefore, it is clinically important to explore safe and effective treatment options for BPH.

Currently, the main treatment modalities for BPH include pharmacologic therapy, minimally invasive surgery, and surgery. For patients with moderate-to-severe LUTS, pharmacologic therapy is usually the preferred option, with α1-adrenergic receptor blockers (e.g., terazosin hydrochloride) and 5α-reductase inhibitors (e.g., finasteride) being the two commonly used classes of drugs in clinical practice. Terazosin hydrochloride, as a representative drug of selective α1-adrenergic receptor blockers, has advantages in the treatment of LUTS caused by BPH^6,7. Compared to conventional non-selective alpha blockers, selective α1‑adrenoceptor blockers are more selective for α1A receptors and improve LUTS with fewer cardiovascular side effects⁸. Recent studies have also suggested that terazosin may exert its therapeutic effects through multiple mechanisms, such as modulation of prostate microcirculation and inhibition of inflammatory responses^9,10. Finasteride, on the other hand, specifically inhibits type II 5α-reductase, reducing DHT levels in serum and prostate tissues by more than 70%, thus inhibiting prostate epithelial cell proliferation, reducing prostate size, and delaying disease progression with long-term use¹¹. However, there are limitations to the efficacy of single-agent therapy. Therefore, in recent years, combination regimens (e.g., α1-blockers combined with 5α-reductase inhibitors) have become a research hotspot, with a view to slowing down disease progression and reducing adverse effects while improving symptoms. Several large-scale clinical studies have confirmed that the combination of α1-blockers and 5α-reductase inhibitors is superior to monotherapy in improving symptom scores, increasing urinary flow rate, and reducing the occurrence of acute urinary retention¹². A systematic review and meta‑analysis has also demonstrated the favorable efficacy of α1-blocker combined with antimuscarinic agents for BPH patients with predominant storage symptoms, further supporting the rationale of multi‑target pharmacological approaches¹³. The advantages of combination therapy are especially pronounced in patients with larger prostate volume (PV) (≥ 40 mL) and higher Prostate-Specific Antigen (PSA) levels (> 1.4 ng/mL). The most recent European urology guidelines, published in 2023, also explicitly recommend that initial combination therapy should be considered for patients with moderate-to-severe LUTS with enlarged prostate¹⁴. Recent studies have identified a unique sex hormone imbalance characterizing patients with BPH: in addition to classic abnormalities of androgen metabolism, an elevated estrogen/androgen ratio may exacerbate disease progression by promoting prostatic interstitial hyperplasia^15,16. A community‑based study in aging Chinese males further confirmed a significant association between symptomatic late‑onset hypogonadism (SLOH) and LUTS, indicating that testosterone deficiency may serve as a risk factor for LUTS¹⁷. These findings highlight the importance of evaluating both androgen and estrogen status when managing BPH patients. Data from longitudinal studies longer than 3 years are lacking on the compounding effect of this multi-targeted pharmacologic intervention on hormone balance. In addition, in terms of safety, there has been concern about the tolerability of combination therapy. Several studies between 2020-2023 have systematically assessed the adverse effect profile of combination regimens. Overall, combination therapy did not significantly increase the known side effects of single agents, and the most common adverse events remained postural hypotension associated with α1-blockers (incidence 3-5%) and sexual dysfunction associated with finasteride (incidence 8-15%)^18,19. These findings provide important safety data for long-term clinical use.

Propensity score matching (PSM) is particularly useful in retrospective studies where treatment assignment is non‑randomized and baseline characteristics may differ between groups, as it helps to reduce selection bias by balancing observed covariates. The efficacy of α‑blocker and 5α‑reductase inhibitor combination therapy in alleviating LUTS and preventing progression is well‑established^20,21. However, a notable gap exists in two areas: 1) the specific, long-term benefits for urinary control, particularly the reduction in 24 h incontinence frequency and the resultant improvement in quality of life (ICIQ-SF), beyond the common focus on voiding symptoms; and 2) the long-term dynamic changes in the hormonal profile. The interaction where finasteride reduces DHT and elevates T, requires longer-term data. Current studies, often limited to 12-month follow-ups, lack the 24-36 month longitudinal observations needed to track T, E2, and DHT changes and their clinical significance.

Based on the research gaps, this PSM based retrospective cohort study compares the real-world effectiveness of terazosin plus finasteride versus finasteride monotherapy. It specifically investigates improvements in urinary continence and tracks dynamic changes in sex hormone levels over 36 months. We hypothesize that the combination therapy will lead to significantly greater improvements in International Prostate Symptom Score (IPSS), maximum urinary flow rate (Qmax), and post-void residual urine volume(PVR), as well as a faster, more pronounced, and sustained reduction in 24 h incontinence episodes and ICIQ-SF scores. Regarding hormonal outcomes, we anticipate that the combination regimen will be associated with more profound DHT suppression, accompanied by a compensatory rise in serum testosterone and a decrease in estradiol, with these hormonal changes remaining stable throughout the follow-up. The findings aim to offer evidence for optimizing BPH management and guiding clinical decisions that weigh efficacy against hormonal impacts.

Protocol

This clinical study was conducted in compliance with the ethical principles outlined in the Declaration of Helsinki and was approved by the Ethics Committee of the Haiyan County People's Hospital(Approval No.: 2025-17).

Study participant assessment and selection

Clinical data were extracted for all patients diagnosed with BPH from the hospital's electronic medical record system, covering the period from January 2020 to March 2021.

PSM was employed using the 1:1 nearest-neighbor method to mitigate potential confounding factors and selection bias inherent in retrospective designs.

This matching process yielded a final cohort of 120 patients, with 60 patients in each group. Patients with missing data at any scheduled time point, including baseline, 6, 12, 24, or 36 months, were excluded from the analysis. Consequently, all 120 matched patients had complete follow-up data. As shown in Figure 1, the study workflow included patient screening, PSM, and final cohort allocation.

The following inclusion criteria were applied to screen potential participants. The diagnosis met the criteria of the Chinese Guidelines for the Diagnosis and Treatment of BPH. The patient's age was verified to be between 50 and 80 years, with an IPSS ≥ 8, a PSA level > 1.4 ng/mL, and a PV ≥ 40 mL as measured by transrectal ultrasound. The Qmax was confirmed to be ≤ 15 mL/s. Voluntary participation and signed informed consent were also verified.

The following exclusion criteria were applied. Patients with comorbid prostate cancer, urinary tract infection, neurogenic bladder, or bladder stones were excluded. Patients with a history of prostate surgery or prior treatment with 5α-reductase inhibitors or α1-blockers were excluded. Patients with severe cardiac, hepatic, or renal insufficiency, such as serum creatinine > 2.0 mg/dL, were excluded. Patients with a known allergy to terazosin hydrochloride or finasteride, as well as those with comorbid severe psychiatric illness or anticipated poor compliance, were also excluded.

The following dropout criteria were defined and applied after enrollment. Patients who did not ultimately meet all inclusion criteria or had critically incomplete information were excluded from the final analysis. Patients who demonstrated poor medication compliance or voluntarily withdrew from the study were excluded. Patients who experienced a sudden onset of other serious illnesses during the study period were excluded. Patients who developed severe psychological or emotional problems that prevented continued participation were excluded. Patients who underwent a change in treatment, such as the addition of a new surgical procedure, were also excluded.

The required sample size was calculated prior to matching. An a priori sample size calculation was performed using dedicated software. An independent-samples t-test model was used. The effect size (d) was set to 0.5, the significance level (alpha) was set to 0.05 (two-tailed), and the desired statistical power (1-β) was set to 0.8. The calculated sample size was inflated to account for an estimated 10% data attrition rate and potential losses during the PSM process. The initial calculation indicated a need for 54 participants per group, for a total sample size of 108. The final cohort was expanded to 120 individuals to meet statistical requirements.

Treatment protocol assignment and execution

Patient-specific treatment details were identified and compiled from Electronic Medical Record (EMR) prescription data.

For patients assigned to the combination therapy group, the administration of terazosin hydrochloride (2 mg) orally once daily at bedtime and finasteride (5 mg) orally once daily in the morning was confirmed from the records.

For patients assigned to the monotherapy group, the administration of finasteride (5 mg) orally once daily in the morning was confirmed from the records.

The treatment period was defined as 12 months, and the total follow-up period was defined as 36 months. Patients were instructed not to take any other medications affecting urinary function or sex hormones during the study.

Data collection for outcome measures

All study data were collected retrospectively from institutional electronic health records, laboratory databases, and nursing documentation systems.

Data for the primary outcome measures were collected at baseline, 6, 12, 24, and 36 months. Sex hormone levels, including dihydrotestosterone, serum testosterone, and estradiol, were extracted and recorded. Urinary control function metrics, including the number of urinary incontinence episodes within 24 h and the ICIQ-SF score, were also extracted and recorded.

Data for the secondary outcome measures were collected at the same time points: baseline, 6, 12, 24, and 36 months. The IPSS, Qmax, PVR, and PV were extracted and recorded. All reported adverse events, such as dizziness, hypotension, and sexual dysfunction, were extracted and recorded from adverse event report forms, along with relevant liver and kidney function indicators from laboratory reports.

Statistical analysis

All statistical analyses were conducted using dedicated software. To mitigate selection bias, 1:1 nearest-neighbor PSM without replacement was performed. The propensity score was estimated by a logistic regression model, with treatment assignment, defined as combination therapy versus monotherapy, as the dependent variable. The following baseline covariates were forced into the model based on clinical plausibility and prior literature: age, BMI, IPSS, Qmax, PVR, PV, PSA, C-reactive protein, total testosterone, estradiol, DHT, 24 h incontinence episodes, and ICIQ-SF score. Matching was performed on the logit of the propensity score using a caliper of 0.0224. Balance after matching was assessed using standardized mean differences (SMDs), with SMD < 0.1 indicating negligible imbalance. As shown in Table 1, all SMDs after matching were < 0.1, confirming adequate confounding control.

Continuous variables were tested for normality using the Shapiro-Wilk test. For baseline comparisons between the matched groups, normally distributed continuous variables were presented as mean ± standard deviation (x̄ ± s) and were compared using the paired t-test to account for the matched pairs. Categorical variables were presented as numbers and percentages [n (%)] and were compared using McNemar's test for paired binary data.

For longitudinal outcomes measured at multiple time points, including baseline, 6, 12, 24, and 36 months, linear mixed-effects models were employed, with treatment group, time, and their interaction as fixed effects, and subjects as random intercepts. This approach accounted for the matched design and within-subject correlation. When appropriate, post-hoc comparisons with Bonferroni correction were performed.

All statistical tests were performed as two-sided. A P-value of less than 0.05 was defined as the threshold for statistical significance.

Results

The study initially enrolled 148 patients (76 in the combination group; 72 in the monotherapy group). To minimize potential confounding, 1:1 propensity score matching was applied, resulting in 120 successfully matched patients (60 per group). As presented in Table 1, significant baseline discrepancies (P < 0.05) existed before matching, with the monotherapy group being older and having higher C-reactive protein levels, total IPSS, voiding subscores, post-void residual urine, and nocturia episodes, suggesting ...

Discussion

The prostate is a gonadal organ located at the bladder neck, traversed centrally by the urethra. After age 45, age‑related changes in sex hormone metabolism are associated with prostatic growth, which may progress to BPH. The condition typically evolves slowly; patients may remain asymptomatic as long as there is no lower urinary tract obstruction. As prostate volume increases, or when complicated by stones or infection, urinary obstruction and voiding difficulties may develop, potentially compromising patient heal...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
Automated Chemiluminescence Immunoassay Analyzer	F. Hoffmann-La Roche Ltd.	7682913001 (e 801 Module)	Used for quantitative detection of serum testosterone (T), estradiol (E2), and dihydrotestosterone (DHT).
Dihydrotestosterone (DHT) Detection Reagent Kit	Beijing Solarbio Science & Technology Co., Ltd.	SEKSM-0030	Used for quantitative detection of DHT concentration in serum and plasma.
Hospital Electronic Medical Record System	Winning Health Technology Group Co.,Ltd.	HIS V5.0	Used for retrospective extraction of patient data.
International Consultation on Incontinence Questionnaire-Short Form	—	ICI-Q-SF	Used to assess impact of urinary incontinence on quality of life.
International Prostate Symptom Score Questionnaire	—	IPSS (LOINC 80976-4)	Used to assess lower urinary tract symptoms (LUTS).
Laboratory Centrifuge	Thermo Fisher Scientific Inc.	75007201 (230V)	Used for serum separation. Standard conditions: 1500 × g for 10 minutes.
Sample Size Calculation Software	G*Power	Version 3.1	Used for a priori power analysis.
Sex Hormone Detection Reagent Kit	F. Hoffmann-La Roche Ltd.	Testosterone II: 05200067	Used for quantitative detection of testosterone or estradiol in serum/plasma on Roche analyzers.
Statistical Analysis Software	IBM SPSS Statistics	Version 25.0	Used for all data analysis.
Transrectal Ultrasound Diagnostic System	Koninklijke Philips N.V.	EPIQ 7 model 795200	Used to measure prostate volume (PV) and post-void residual (PVR).
Urinary Flowmeter	Guangzhou Pudong Medical Equipment Co., Ltd.	PT-UFM-A	Used to measure maximum urinary flow rate (Qmax).