Patient population
A total of 246 720 patients with ≥ 1 medical claim for a primary diagnosis for BPH were identified from the HIRA-NPS database from January 1, 2012 to December 31, 2017. Of these, 14,755 (6.0%) patients were eligible and included in the analysis (Additional file 3): 1529 (10.4%) patients received daily treatment with free combination therapy, 6660 (45.1%) received dutasteride monotherapy 0.5 mg daily, and 6566 (44.5%) received tamsulosin monotherapy 0.4 mg daily.
Prior to IPTW adjustment, several baseline demographic and clinical characteristic imbalances were observed between treatment groups with standardized differences of 10% or above. Patients treated with free combination therapy were more likely to have polyuria (13.9% vs 9.9%, standardized difference = 12.6%), a higher number of symptoms or findings associated with BPH (0.3% vs 0.2%, standardized difference = 11.0%), and BPH with LUTS (17.9% vs 12.3%, standardized difference = 15.7%), than those treated with dutasteride 0.5 mg. The free combination and tamsulosin groups were more balanced; however, patients treated with free combination therapy were less likely to be 50–59 years of age (7.7% vs 12.9%, standardized difference = 17.3%) and more likely to be ≥ 70 years of age (60.6% vs 51.6%, standardized difference = 18.3%) at baseline, than those treated with tamsulosin 0.4 mg.
Following IPTW adjustment, baseline demographics and clinical characteristics were balanced across all treatment cohorts with standardized differences of < 10% (Table 1). Most patients (> 87%) were aged ≥ 60 years; BPH with LUTS affected 13.3–18.6% of patients, and polyuria was the most common symptom associated with BPH, affecting 10.6–14.6% of the IPTW-adjusted samples. The most common medications used during the baseline period were non-steroidal anti-inflammatory drugs (NSAIDS; 34.6 − 37.9%), calcium channel blockers (21.0 − 24.2%), and antihypertensives (21.8–23.4%).
Table 1
Baseline characteristics of patients treated with free combination therapy compared with those treated with dutasteride 0.5 mg or tamsulosin 0.4 mg monotherapy, after adjustment using inverse probability of treatment weight.
|
Free combination therapy vs
dutasteride monotherapy
|
Free combination therapy vs
tamsulosin monotherapy
|
|
Free combination of dutasteride plus tamsulosin therapy
(n = 1527)
|
Dutasteride
monotherapy
(n = 6661)
|
Std. diff*
(%)
|
Free combination of dutasteride plus
tamsulosin therapy
(n = 1544)
|
Tamsulosin
monotherapy
(n = 6574)
|
Std. diff*
(%)
|
Age, n (%)
|
40–49 years
|
29 (1.9)
|
150 (2.2)
|
2.3
|
20 (1.3)
|
90 (1.4)
|
0.7
|
50–59 years
|
165 (10.8)
|
704 (10.6)
|
0.8
|
182 (11.8)
|
777 (11.8)
|
0.2
|
60–69 years
|
488 (32.0)
|
2133 (32.0)
|
0.1
|
517 (33.5)
|
2201 (33.5)
|
0.0
|
≥70 years
|
844 (55.3)
|
3675 (55.2)
|
0.3
|
826 (53.5)
|
3506 (53.3)
|
0.3
|
Clinical characteristics, n (%)
|
Any AE†
|
422 (27.7)
|
1857 (27.9)
|
0.5
|
476 (30.8)
|
2030 (30.9)
|
0.1
|
Cardiovascular disease‡
|
565 (37.0)
|
2475 (37.2)
|
0.3
|
622 (40.3)
|
2612 (39.7)
|
1.2
|
Hyperlipidemia
|
304 (19.9)
|
1361 (20.4)
|
1.3
|
338 (21.9)
|
1433 (21.8)
|
0.3
|
Chronic pulmonary disease
|
221 (14.5)
|
988 (14.8)
|
0.9
|
258 (16.7)
|
1088 (16.5)
|
0.4
|
BPH with LUTS
|
210 (13.7)
|
889 (13.3)
|
1.1
|
288 (18.6)
|
1193 (18.2)
|
1.2
|
Polyuria§
|
167 (10.9)
|
709 (10.6)
|
1.0
|
225 (14.6)
|
936 (14.2)
|
1.0
|
Concomitant medications, n (%)
|
NSAIDs
|
530 (34.7)
|
2302 (34.6)
|
0.4
|
585 (37.9)
|
2417 (36.8)
|
2.3
|
Calcium channel blockers
|
330 (21.6)
|
1400 (21.0)
|
1.4
|
374 (24.2)
|
1500 (22.8)
|
3.3
|
Antihypertensives
|
334 (21.9)
|
1454 (21.8)
|
0.1
|
361 (23.4)
|
1459 (22.2)
|
2.8
|
AE, adverse event; BPH, benign prostatic hyperplasia; LUTS, lower urinary tract symptoms; NSAIDs, nonsteroidal anti-inflammatory drugs; |
Std. diff, standardized difference.
*For continuous variables, the standardized difference was calculated by dividing the absolute difference in means of the free combination therapy cohort and reference monotherapy cohorts by the pooled standard deviation (SD) of both groups, for each comparison. The pooled SD was the square root of the average of the squared SD. For dichotomous variables, the standardized difference was calculated using the following equation where P is the respective proportion of participants in each treatment cohort: [(Pfreecombination therapy- Preference)/ √(Pfreecombinationtherapyx(1 – Pfreecombinationtherapy) + Preference x (1 – Preference))/ 2].
†See Supplementary Appendix 2 for list of AEs
‡Three categories of Quan–Charlson comorbidities (ie, congestive heart failure, peripheral vascular disease, and myocardial infarction) are listed under cardiovascular disease.
§Polyuria includes nocturia and urinary frequency.
Treatment frequency and duration
The proportion of patients treated with free combination therapy was 10.4%, compared with 45.1% (dutasteride monotherapy) and 44.5% (tamsulosin monotherapy); this exceeded the threshold of 3% selected to define frequent combined use. The mean ± standard deviation (SD) duration of treatment was similar for patients treated with free combination therapy (292.5 ± 54.1 days) and either monotherapy (297.1 ± 54.0 days for dutasteride; 295.8 ± 54.0 days for tamsulosin) (Table 2).
Table 2
Frequency and duration of treatment with free combination therapy, dutasteride 0.5 mg monotherapy, or tamsulosin 0.4 mg monotherapy in patients with prevalent BPH in South Korea.
|
|
Free combination therapy vs
dutasteride monotherapy
|
Free combination therapy vs
tamsulosin monotherapy
|
|
Free combination of dutasteride plus tamsulosin therapy
(n = 1 529)
|
Dutasteride monotherapy
(n = 6 660)
|
Std. diff*
|
Tamsulosin monotherapy
(n = 6 566)
|
Std. diff*
|
Treatment duration (days)
|
Mean ± SD
|
292.5 ± 54.1
|
297.1 ± 54.0
|
8.6
|
295.8 ± 54.0
|
6.1
|
Median, IQR
|
305.0 (249.0, 341.0)
|
310.0 (260.0, 342.0)
|
310.0 (255.0, 343.0)
|
Treatment duration, n (%)
|
6–9 months
|
519 (33.9)
|
2020 (30.3)
|
7.7
|
2085 (31.8)
|
4.7
|
9–12 months
|
1010 (66.1)
|
4640 (69.7)
|
7.7
|
4481 (68.2)
|
4.7
|
BPH, benign prostatic hyperplasia; IQR, interquartile range; SD, standard deviation; Std. Diff, standardized difference. |
*For continuous variables, the standardized difference was calculated by dividing the absolute difference in means of the free combination therapy cohort and reference monotherapy cohorts by the pooled standard deviation (SD) of both groups, for each comparison. The pooled SD was the square root of the average of the squared SD. For dichotomous variables, the standardized difference was calculated using the following equation where P is the respective proportion of participants in each treatment cohort: [(Pfreecombination therapy- Preference)/ √(Pfreecombinationtherapyx(1 – Pfreecombinationtherapy) + Preference x (1 – Preference))/ 2]. |
Overall risk of any AE or SAE
The risk of any AE occurring was 71.5% (free combination therapy), 64.6% (dutasteride 0.5 mg) and 71.6% (tamsulosin 0.4 mg). The adjusted RR for any AE over the treatment observation period comparing free combination therapy with dutasteride 0.5 mg monotherapy was 1.07 (95% CI: 1.03, 1.12) (Fig. 2A) and comparing free combination therapy with tamsulosin 0.4 mg monotherapy was 0.98 (95% CI: 0.95, 1.02) (Fig. 2B). The risk of any AE with free combination therapy was non-inferior to that of both monotherapy groups based on non-inferiority margins (non-inferiority P < 0.001 each comparison).
The incidence of any SAE in the overall patient population was rare and similar across treatment groups: 1.6% (free combination therapy), 1.3% (dutasteride 0.5 mg monotherapy), and 1.7% (tamsulosin 0.4 mg monotherapy). The adjusted RR for any SAE over the treatment observation period comparing free combination therapy with dutasteride 0.5 mg monotherapy was 1.07 (95% CI: 0.66, 1.74) (Fig. 2C), and comparing free combination therapy with tamsulosin 0.4 mg monotherapy was 0.90 (95% CI: 0.56, 1.45) (Fig. 2D). Non-inferiority was not shown for either comparison (non-inferiority P = 0.852 and 0.762 for comparison with dutasteride and tamsulosin monotherapies, respectively). However, the 95% CI for the RR estimate included the null value of one and had a lower limit below the non-inferiority margins, indicating no meaningful differences in the risk of SAE between treatment groups.
Sensitivity analysis among patients without baseline AEs or SAEs
Results relating to the frequency and duration of treatment and risk of any AE or SAE were consistent with the primary results when patients with AEs or SAEs during the baseline period were excluded from the analysis. For comparison of AEs with free combination therapy versus dutasteride 0.5 mg monotherapy, the adjusted RR was 1.10 (95% CI: 1.04, 1.16), indicating a small meaningful difference in favor of dutasteride monotherapy. For comparison of free combination therapy with tamsulosin 0.4 mg monotherapy, the adjusted RR for any AE was 0.98 (95% CI: 0.93, 1.04), indicating no meaningful difference. The RR for any SAE of free combination therapy versus dutasteride 0.5 mg monotherapy was 1.13 (95% CI: 0.58, 2.24), and versus tamsulosin 0.4 mg monotherapy was 0.80 (95% CI: 0.42, 1.51), indicating no meaningful differences.
Subgroup analysis by age
Treatment frequency with free combination therapy was lowest in patients 40–59 years of age (6.3%) and highest in patients > 70 years of age (11.7%). Among patients aged 60–69 years, 9.8% used free combination therapy. The average duration of treatment was similar for patients treated with free combination therapy and each monotherapy across all age groups (Additional file 4). The risk of any AE and SAE within each age category was similar to that for the overall population, with no meaningful differences between treatment groups (Additional file 5).
Overall risk of specific AEs and SAEs
The most common AEs throughout the treatment observation period among free combination therapy, dutasteride 0.5 mg monotherapy, and tamsulosin 0.4 mg monotherapy were constipation (26.2%, 19.1%, 24.3%, respectively), depressed mood (14.3%, 11.9%, 15.6%, respectively), urticaria (14.1%, 12.4%, 15.6%, respectively), dizziness (13.2%, 12.5%, 13.5%, respectively), and prostate cancer (12.6%, 9.4%, 12.0%, respectively) (Table 3). The 95% CI of the RR for most specific AEs overlapped with 1, indicating no meaningful differences between treatment groups.
The risk of specific SAEs was low (< 1% each) in all treatment groups. The most commonly reported SAEs in the free combination therapy, dutasteride monotherapy and tamsulosin monotherapy groups, respectively, were dizziness (0.5%, 0.3%, 0.4%), prostate cancer (0.4%, 0.2%, 0.4%), cardiac failure (0.3%, 0.1%, 0.2%), arrhythmia (0.2%, 0.2%, 0.2%), and vertigo (0.1%, 0.2%, 0.2%).
Table 3
Risk of specific AEs with free combination therapy compared with dutasteride or tamsulosin monotherapy.
|
|
Free combination therapy vs
dutasteride monotherapy
|
Free combination therapy vs
tamsulosin monotherapy
|
|
Free combination of dutasteride plus tamsulosin therapy
(n = 1529)
|
Dutasteride monotherapy
(n = 6600)
|
Adjusted RR
(95% CI)*
|
Tamsulosin
monotherapy
(n = 6566)
|
Adjusted RR
(95% CI)*
|
Specific AE, n (%)
|
|
|
|
|
|
Constipation
|
401 (26.2)
|
1271 (19.1)
|
1.31 (1.17, 1.45)
|
1595 (24.3)
|
1.05 (0.94, 1.16)
|
Depressed mood
|
219 (14.3)
|
794 (11.9)
|
1.05 (0.90, 1.22)
|
1025 (15.6)
|
0.91 (0.78, 1.05)
|
Urticaria
|
216 (14.1)
|
828 (12.4)
|
1.02 (0.87, 1.18)
|
1024 (15.6)
|
0.89 (0.76 ,1.03)
|
Dizziness
|
202 (13.2)
|
833 (12.5)
|
1.01 (0.87, 1.19)
|
889 (13.5)
|
0.98 (0.84, 1.14)
|
Prostate cancer
|
193 (12.6)
|
625 (9.4)
|
1.40 (1.19, 1.65)
|
786 (12.0)
|
1.11 (0.95, 1.30)
|
Arrhythmia
|
188 (12.3)
|
608 (9.1)
|
1.33 (1.13, 1.57)
|
638 (9.7)
|
1.24 (1.05, 1.46)
|
Vertigo
|
176 (11.5)
|
643 (9.7)
|
1.20 (1.01, 1.42)
|
735 (11.2)
|
1.05 (0.89, 1.24)
|
Diarrhea
|
166 (10.9)
|
702 (10.5)
|
0.99 (0.83, 1.17)
|
765 (11.7)
|
0.94 (0.80, 1.12)
|
Pruritus
|
146 (9.5)
|
624 (9.4)
|
0.98 (0.81, 1.18)
|
757 (11.5)
|
0.83 (0.69, 1.00)
|
Cardiac failure
|
137 (9.0)
|
400 (6.0)
|
1.37 (1.12, 1.68)
|
442 (6.7)
|
1.23 (1.01, 1.49)
|
Vomiting
|
117 (7.7)
|
374 (5.6)
|
1.29 (1.03, 1.62)
|
513 (7.8)
|
0.99 (0.80, 1.22)
|
Rhinitis
|
103 (6.7)
|
422 (6.3)
|
1.11 (0.89, 1.39)
|
454 (6.9)
|
0.97 (0.78, 1.22)
|
Dyspnea
|
74 (4.8)
|
266 (4.0)
|
1.10 (0.84, 1.44)
|
288 (4.4)
|
1.00 (0.77, 1.30)
|
Asthenia
|
51 (3.3)
|
196 (2.9)
|
1.06 (0.77, 1.47)
|
204 (3.1)
|
1.06 (0.75, 1.50)
|
Localized edema
|
30 (2.0)
|
130 (2.0)
|
0.84 (0.55, 1.29)
|
149 (2.3)
|
0.80 (0.52, 1.21)
|
Impotence
|
21 (1.4)
|
88 (1.3)
|
1.15 (0.69, 1.93)
|
108 (1.6)
|
1.09 (0.66, 1.79)
|
Epistaxis
|
19 (1.2)
|
84 (1.3)
|
1.10 (0.65, 1.87)
|
87 (1.3)
|
1.08 (0.60, 1.95)
|
Syncope orthostatic
|
13 (0.9)
|
62 (0.9)
|
0.83 (0.45, 1.53)
|
70 (1.1)
|
0.74 (0.41, 1.34)
|
Hypotension
|
11 (0.7)
|
35 (0.5)
|
1.34 (0.67, 2.70)
|
34 (0.5)
|
1.24 (0.62, 2.46)
|
Rash
|
9 (0.6)
|
22 (0.3)
|
1.45 (0.65, 3.26)
|
23 (0.4)
|
1.33 (0.57, 3.09)
|
Alopecia
|
5 (0.3)
|
34 (0.5)
|
0.87 (0.33, 2.28)
|
28 (0.4)
|
1.09 (0.41, 2.91)
|
Breast disorder
|
5 (0.3)
|
25 (0.4)
|
1.12 (0.39, 3.19)
|
18 (0.3)
|
1.53 (0.47, 4.99)
|
Dry mouth
|
5 (0.3)
|
17 (0.3)
|
0.98 (0.34, 2.83)
|
23 (0.4)
|
0.83 (0.30, 2.31)
|
Visual impairment
|
2 (0.1)
|
16 (0.2)
|
0.44 (0.10, 1.97)
|
17 (0.3)
|
0.41 (0.10, 1.79)
|
Other specified disorders of male genital organ
|
1 (0.1)
|
8 (0.1)
|
0.92 (0.12, 7.37)
|
10 (0.2)
|
0.54 (0.07, 4.25)
|
Vision blurred
|
0 (0.0)
|
5 (0.1)
|
—
|
7 (0.1)
|
—
|
Angioedema
|
0 (0.0)
|
4 (0.1)
|
—
|
2 (0.0)
|
—
|
Erythema multiforme
|
0 (0.0)
|
2 (0.0)
|
—
|
4 (0.1)
|
—
|
Premature ejaculation
|
0 (0.0)
|
2 (0.0)
|
—
|
1 (0.0)
|
—
|
Breast cancer
|
0 (0.0)
|
1 (0.0)
|
—
|
1 (0.0)
|
—
|
Hypertrichosis
|
0 (0.0)
|
1 (0.0)
|
—
|
0 (0.0)
|
—
|
Loss of libido
|
0 (0.0)
|
1 (0.0)
|
—
|
0 (0.0)
|
—
|
Dermatitis exfoliative
|
0 (0.0)
|
0 (0.0)
|
—
|
3 (0.0)
|
—
|
Priapism
|
0 (0.0)
|
0 (0.0)
|
—
|
1 (0.0)
|
—
|
AE, adverse event; CI, confidence interval; RR, risk ratio. |
*RRs of any or specific AEs among patients receiving free combination therapy compared to each monotherapy were estimated using logbinomial regression models adjusted for inverse probability of treatment weights. A robust variance estimator was used to derive the 95% CIs. |