Real-World PSA Response and Overall Survival Among Men with mCSPC Receiving Apalutamide Versus Darolutamide (Both Without Docetaxel) in the US

Abstract

Introduction

To date, no head-to-head comparisons of apalutamide versus darolutamide have been reported for metastatic castration-sensitive prostate cancer (mCSPC). This study compared prostate-specific antigen decline ≥ 90% (PSA90) and overall survival (OS) between apalutamide and darolutamide, both without docetaxel, among patients with mCSPC in real-world clinical practice in the USA.

Methods

Men diagnosed with mCSPC who initiated apalutamide or darolutamide between 2022 and 2025 were identified from linked electronic medical records and insurance claims. The apalutamide and darolutamide cohorts were balanced using inverse probability of treatment weighting. PSA90 response was assessed on-treatment. The proportions of patients achieving a PSA90 response and OS through a maximum of 6 months and 24 months post-treatment initiation, respectively, were compared between the two cohorts using weighted Kaplan-Meier and weighted Cox proportional hazards models.

Results

For PSA90 analyses, weighted characteristics were well balanced between the apalutamide (n = 714; mean age 73.9 years, 59.6% White, 21.2% Black, 13.7% other, 5.5% unknown race) and darolutamide cohorts (n = 145; mean age 74.3 years, 58.8% White, 21.6% Black, 15.0% other, 4.7% unknown race). PSA90 response rates through 6 months were 49% higher for apalutamide than for darolutamide (weighted hazard ratio [HR]: 1.49 [95% confidence interval (CI) 1.07, 2.07]; p = 0.017). For OS analyses, weighted characteristics were also well balanced between apalutamide (n = 1460; mean age 73.5 years, 59.6% White, 21.5% Black, 13.5% other, 5.4% unknown race) and darolutamide (n = 287; mean age 73.7 years, 60.0% White, 22.4% Black, 12.4% other, 5.2% unknown race). Apalutamide was associated with a 51% lower rate of mortality relative to darolutamide through 24 months (weighted HR: 0.49 [95% CI 0.30, 0.83]; p = 0.007).

Conclusion

Among patients with mCSPC treated without docetaxel, apalutamide was associated with higher PSA90 response rates and lower mortality than darolutamide. These findings indicate a potential difference in disease control and survival between the two androgen receptor-targeted agents in routine clinical practice.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s12325-026-03707-z.

Plain Language Summary

Apalutamide and darolutamide are medications that block hormone activity that fuels prostate cancer growth. They are currently approved to treat metastatic castration-sensitive prostate cancer (mCSPC) with standard hormone-lowering combination therapy (androgen-deprivation therapy, ADT).

This study used medical and insurance data from US community urology practices to compare outcomes between patients with mCSPC who started apalutamide (1460 patients) or darolutamide (287 patients), both without docetaxel. Apalutamide was FDA approved for this use throughout the study period. Darolutamide without docetaxel was FDA approved only near the end of the study period (June 2025) but was still widely used in US urology practices during this time. Statistical methods were used to standardize patient features between the two groups by weighting them for specific measured differences in patient and disease characteristics.

Overall, 74.6% of patients treated with apalutamide and 56.1% of patients receiving darolutamide achieved ≥ 90% drop in prostate-specific antigen (PSA, a prostate cancer marker) levels among those tested within 6 months of starting treatment, a 49% increase in the rate of PSA response for apalutamide.

Additionally, 24 months after starting treatment, 92.1% of patients who received apalutamide and 85.8% of patients who received darolutamide remained alive, a 51% lower rate of death with apalutamide.

While these findings are subject to limitations, discussed in the full manuscript, they suggest that in clinical practice and without chemotherapy intensification, apalutamide may provide better disease control and longer overall survival than darolutamide in patients with mCSPC. Notably, no prior studies have directly compared these two treatments for these outcomes.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s12325-026-03707-z.

Key Summary Points

Why carry out this study?

Apalutamide and darolutamide are two therapeutic agents among the class of androgen receptor pathway inhibitors approved for the treatment of metastatic castration (hormone)-sensitive prostate cancer (mCSPC)

In combination with androgen deprivation therapy (ADT), the pivotal phase 3 TITAN clinical trial for apalutamide has shown a statistically significant difference in overall survival (OS) relative to placebo + ADT, while no statistically significant difference in OS was observed for darolutamide + ADT versus placebo + ADT in the ARANOTE trial

To date, no real-world comparative effectiveness studies have assessed differences in prostate-specific antigen (PSA) response and OS between patients with mCSPC treated with apalutamide or darolutamide, both without docetaxel

What was learned from this study?

Rates of PSA response ≥ 90% from pre-treatment levels were 49% higher for patients with mCSPC initiating apalutamide without docetaxel using up to 6-month follow-up data compared with those initiated on darolutamide without docetaxel (PSA90 response: apalutamide 74.6%, darolutamide 56.1%)

Patients with mCSPC initiating apalutamide without docetaxel experienced a 51% reduction in the rate of death using up to 24-month follow-up data compared with those initiated on darolutamide without docetaxel (OS: apalutamide 92.1%, darolutamide 85.8%)

Introduction

Prostate cancer (PC) is the most common non-dermatologic malignancy and the second leading cause of cancer-related mortality in the USA [1, 2]. Although patients with localized PC generally have a good prognosis, those with metastatic PC have a 5-year survival rate of only 37.9% [1, 3], signaling a need for improved management in later disease stages.

Apalutamide, an androgen receptor pathway inhibitor (ARPI), was approved by the US Food and Drug Administration (FDA) for use in combination with androgen deprivation therapy (ADT) to treat metastatic castration-sensitive prostate cancer (mCSPC) in September 2019 based on the results of the phase 3 TITAN clinical trial, which demonstrated significantly longer radiographic progression-free survival (hazard ratio [HR]: 0.48 [95% confidence interval (CI) 0.39, 0.60]; p < 0.001) and overall survival (OS; HR: 0.67 [95% CI 0.51, 0.89]; p = 0.005) with apalutamide than placebo [4–6]. Subsequently, the ARPI darolutamide received FDA approval for use in combination with ADT and docetaxel in August 2022 [7], based on superior OS relative to placebo (HR: 0.68 [95% CI 0.57, 0.80]; p < 0.001) in the phase 3 ARASENS trial [8]. More recently, darolutamide received an additional mCSPC indication for use in combination with ADT alone in June 2025 [9]. This most recent approval was based on the phase 3 ARANOTE trial, in which darolutamide was associated with a lower rate of radiologic progression or death than placebo (HR: 0.54 [95% CI 0.41, 0.71]; p < 0.001) [10]. However, unlike the TITAN trial for apalutamide [10], no OS benefit for darolutamide was observed in the ARANOTE trial [10].

Beyond survival outcomes, post hoc analyses of the TITAN and ARANOTE trials have since demonstrated that deep and durable prostate-specific antigen (PSA) responses can also be achieved with apalutamide and darolutamide, respectively. Importantly, such responses were associated with improved clinical outcomes, including long-term survival, demonstrating the prognostic importance of early PSA measurement in the management of mCSPC [11, 12].

In recent years, real-world studies of patients with mCSPC have found that apalutamide is significantly associated with a higher likelihood of achieving an early PSA decline of ≥ 90% (PSA90) and a reduced mortality compared with enzalutamide or abiraterone acetate [13–16]. However, there are currently no assessments of the comparative effectiveness of apalutamide without docetaxel versus darolutamide without docetaxel in real-world clinical practice in the US.

As such, given the association between early PSA response and OS, and the failure to reach the secondary OS endpoint in the ARANOTE trial, this study was conducted to compare PSA response and OS between apalutamide without docetaxel and darolutamide without docetaxel among patients with mCSPC in real-world clinical practice.

Methods

Data Source

This study used linked patient-level data from the following two unique sources: (1) electronic medical records (EMRs) from private, community-based urology practices located across all US regions, provided by Precision Point Specialty Analytics (PPS), and (2) insurance claims data from Komodo Research Database (KRD). KRD reports monthly mortality data aggregated from multiple third-party sources (i.e., not dependent on healthcare provider reporting), capturing > 90% of all oncology deaths identified by the US Centers for Disease Control in 2018–2023 based on an external validation assessment [17]. Data from the two databases were linked using Datavant’s patent-pending, machine-learning-validated tokenization technology, which de-identifies patient information using encrypted and non-reversible tokens.

Ethical Approval

The study was considered exempt research under 45 CFR § 46.104(d)(4) as it involved only the secondary use of data that were de-identified in compliance with the Health Insurance Portability and Accountability Act, specifically, 45 CFR § 164.514. The study was performed in accordance with the Helsinki Declaration of 1964 and its later amendments.

Study Design

A retrospective, longitudinal, cohort study design was used to evaluate PSA outcomes (study period: January 1, 2016–April 30, 2025) and OS (study period: January 1, 2016–June 30, 2025) among patients treated with apalutamide without docetaxel or darolutamide without docetaxel (Figures S1 and S2 in the Supplementary Material). The index date was defined as the date of first dispensation or pharmacy claim of either apalutamide or darolutamide on or after the FDA approval date for darolutamide plus docetaxel and ADT for mCSPC (i.e., August 5, 2022) [5, 7], with patients classified into the corresponding, mutually exclusive apalutamide or darolutamide cohorts. Given the end of the study period was prior (PSA outcomes: April 30, 2025) or immediately following (OS outcome: June 30, 2025) FDA approval for darolutamide without docetaxel (i.e., June 3, 2025) [9], patients included in the darolutamide cohort were all treated off-label for the PSA analysis and almost entirely treated off-label for the OS analysis.

The 12-month period before the index date was defined as the baseline period. For the PSA analysis, the observation period spanned from the index date until the earliest of index ARPI discontinuation (based on a > 90-day gap), ARPI switch, initiation of a radiopharmaceutical, docetaxel initiation > 30 days post-index, end of clinical activity in PPS, or end of data availability. For the OS analysis, the observation period spanned from the index date until the earliest of end of open insurance claim activity in Komodo or end of data availability.

Study Population

For both the PSA and OS analyses, patients were selected for study inclusion if they met the following criteria: (1) had ≥ 1 dispensation or paid claim for apalutamide or darolutamide as a first ARPI on or after August 5, 2022; (2) had no dispensations or paid claims for other, non-index ARPIs on or before the index date; (3) had no prescriptions for other, non-index ARPIs from the beginning of PPS data up to and including the index date; (4) had evidence of metastasis (based on diagnosis observed in medical claims or clinical indicator in the EMR) in the absence of castration resistance before or on the index date (Table S1 in the Supplementary Material) [18]; (5) had ≥ 12 months of clinical activity in PPS or claims activity in KRD before the index date; (6) were ≥ 18 years of age on the index date; (7) had no prior diagnoses of other primary cancers during the baseline period; (8) had no use of immunotherapy, poly (ADP-ribose) polymerase inhibitors, radiopharmaceutical therapy, or chemotherapy any time before or on the index date; and (9) had no docetaxel use within 90 days before or 30 days following the index date. As part of a sensitivity analysis, patients were required to have had no docetaxel use within 90 days before or 90 days following the index date. Additionally, for the PSA analysis, patients were required to have ≥ 1 PSA measurement during the 13 weeks before and including the index date, with the most recent measurement being detectable PSA (i.e., > 0.2 ng/ml). Concurrent ADT use was not required for inclusion in either the apalutamide or darolutamide cohort.

Study Outcomes

The proportion of patients achieving a PSA response was evaluated to a maximum of 6 months following the index date. PSA90 response was defined as the first PSA measurement showing a decline of ≥ 90% from the most recent detectable baseline level, while patients remained on their index ARPI (i.e., on-treatment approach). As an exploratory analysis, undetectable PSA response (i.e., PSA ≤ 0.2 ng/ml) while patients remained on their index ARPI was also assessed. Post-index PSA testing patterns, including frequency of testing, were assessed in the two cohorts during the on-treatment period.

Treatment adherence was assessed using the medication possession ratio (MPR). The MPR was calculated as the sum of the days of supply divided by the total exposure to therapy; the exposure was defined as the number of days between the dates of the first and last paid pharmacy claims or dispensations plus the number of days of supply of the last refill. MPR was calculated among patients with ≥ 2 paid pharmacy claims or dispensations.

OS was assessed as the proportion of patients surviving to a maximum of 24 months following the index date, using an intention-to-treat approach.

Statistical Analyses

Baseline variables were balanced between the apalutamide and darolutamide cohorts using inverse probability of treatment weighting (IPTW), where propensity scores (PS) were generated using probability estimates from a logistic regression model based on dependent (i.e., apalutamide initiation) and independent variables [19]. The following independent variables were included in the PS model: age, race, geographic region, payer type, year of initiation of index ARPI, time between metastasis and ARPI initiation, time between PC diagnosis and ARPI initiation, de novo mCSPC, ADT claim observed 180 days before or after index date, prior use of first-generation ARPI, prior use of bone antiresorptive therapy, most recent PSA level, earliest Gleason score, sites of metastases (i.e., bone, nodal, visceral, and metastasis in multiple sites), and Quan-Charlson Comorbidity Index score. Patients in the apalutamide and darolutamide cohorts were assigned weights of 1/PS and 1/1 − PS, respectively, which were normalized to the mean weight of each cohort and truncated at the 95th percentile to mitigate bias from extreme weights. This weighting approach allowed for the preservation of the full sample size in both cohorts. Baseline characteristics that were considered well balanced between the cohorts had a standardized difference of < 10% [20]. Given different sample sizes, IPTW was performed separately for patients included in the PSA and OS analyses. This method for balancing patient characteristics has been used in prior analyses, including one that assessed hemodialysis treatment modalities and OS [21]. The PS overlap and IPTW distribution are included in Figures S3 to S6 in the Supplementary Material.

Weighted Kaplan–Meier analyses were used to assess separately the proportions of patients achieving a PSA90 response, an undetectable PSA response, and OS in the apalutamide and darolutamide cohorts. The PSA response rates through a maximum follow-up of 6 months and OS rates through a maximum follow-up of 24 months were compared between the two cohorts using weighted Cox proportional hazards models, with HRs and their 95% CIs reported. The proportional hazards assumption was evaluated by including a treatment-by-time interaction term in the Cox model, and statistical significance was assessed using a Wald test.

Power calculations were performed before data analyses to determine the minimum sample size needed to detect a significant difference in each study outcome between the cohorts, using two-sided hypothesis testing and a 0.05 type I error rate. For PSA90, assuming 85% of the sample was treated with apalutamide, a 6-month 74% PSA90 response rate, and an estimated HR of 1.40, an overall sample of approximately 750 patients would be needed to have 80% power to detect this difference. For OS, assuming 84% of the sample was treated with apalutamide, a 24-month 10% mortality rate, and an estimated HR of 0.50, an overall sample of approximately 1330 patients would be needed to have 80% power to detect this difference.

Results

Patient Characteristics

For the PSA analysis, 714 patients were included in the apalutamide cohort and 145 in the darolutamide cohort (Fig. 1). Following weighting, baseline characteristics were well balanced between the two cohorts (Table 1 and Figure S7 in the Supplementary Material). The mean (median) age was 73.9 (75.0) years in the apalutamide cohort and 74.3 (74.0) years in the darolutamide cohort. More than half of all patients were White (apalutamide: 59.6%; darolutamide: 58.8%), while 21.2% and 21.6% were Black, respectively. Patients had a median of 1.9 months (apalutamide cohort) and 2.0 months (darolutamide cohort) between the onset of metastasis to the index date. Less than one-quarter of patients had baseline PSA ≥ 20 ng/ml (apalutamide: 23.6%; darolutamide 23.5%).

Fig. 1.

Sample selection of patients with mCSPC treated with apalutamide without docetaxel and darolutamide without docetaxel—PSA analysis. ARPI, androgen receptor pathway inhibitor; mCSPC, metastatic castration-sensitive prostate cancer; PARP, poly ADP-ribose polymerase; PPS, Precision Point Specialty Analytics; PSA, prostate-specific antigen

Table 1.

Patient characteristics—PSA analysis

	Unweighted population			Weighted populationa
	Apalutamide	Darolutamide	Standardized difference, %	Apalutamide	Darolutamide	Standardized difference, %
	N = 714	N = 145		N = 714	N = 145
Age, years, mean ± SD [median]	74.0 ± 8.2 [75.0]	72.6 ± 9.6 [73.0]	16.5	73.9 ± 8.2 [75.0]	74.3 ± 8.8 [74.0]	4.9
Race,b n (%)
White	425 (59.5)	87 (60.0)	1.0	425 (59.6)	85 (58.8)	1.7
Black	149 (20.9)	33 (22.8)	4.6	151 (21.2)	31 (21.6)	0.9
Otherc	98 (13.7)	20 (13.8)	0.2	98 (13.7)	22 (15.0)	3.7
Unknown	42 (5.9)	5 (3.4)	11.6	40 (5.5)	7 (4.7)	3.8
Geographic region, n (%)
South	436 (61.1)	72 (49.7)	23.1	426 (59.7)	91 (62.7)	6.3
Midwest	169 (23.7)	34 (23.4)	0.5	169 (23.6)	32 (22.1)	3.8
Northeast	62 (8.7)	19 (13.1)	14.2	66 (9.2)	12 (8.4)	2.8
West	47 (6.6)	20 (13.8)	24.0	53 (7.5)	10 (6.8)	2.7
Payer, n (%)
Medicare	575 (80.5)	106 (73.1)	17.7	569 (79.7)	118 (81.5)	4.5
Commercial	114 (16.0)	33 (22.8)	17.3	120 (16.8)	24 (16.4)	1.2
Medicaid	21 (2.9)	5 (3.4)	2.9	21 (3.0)	3 (2.0)	6.5
Unknown	4 (0.6)	1 (0.7)	1.6	4 (0.5)	0 (0.2)	5.1
Year of index date, n (%)
2022	109 (15.3)	26 (17.9)	7.2	112 (15.6)	23 (15.7)	0.2
2023	325 (45.5)	67 (46.2)	1.4	324 (45.4)	61 (41.9)	7.1
2024	212 (29.7)	38 (26.2)	7.8	209 (29.3)	48 (33.1)	8.2
2025	68 (9.5)	14 (9.7)	0.5	69 (9.6)	13 (9.3)	1.3
Time between metastasis and index date, months, median [IQR]	1.9 [0.7, 6.3]	1.8 [0.7, 4.2]	6.1	1.9 [0.7, 6.1]	2.0 [0.5, 5.5]	1.3
Time between PC diagnosis and index date, months, median [IQR]	54.0 [6.1, 98.6]	41.6 [3.1, 92.3]	15.0	51.5 [5.8, 97.6]	62.3 [5.5, 94.6]	0.1
Types of metastasis,d n (%)
Visceral metastasis	98 (13.7)	38 (26.2)	31.6	109 (15.3)	23 (15.5)	0.8
Bone metastasis	354 (49.6)	100 (69.0)	40.3	373 (52.2)	73 (50.5)	3.4
Nodal metastasis	372 (52.1)	66 (45.5)	13.2	364 (51.0)	71 (49.2)	3.7
Metastasis in multiple sites	185 (25.9)	65 (44.8)	40.4	202 (28.3)	41 (28.5)	0.3
Quan-CCI, mean ± SD [median]	10.1 ± 3.4 [10.0]	10.0 ± 3.2 [10.0]	2.9	10.1 ± 3.4 [10.0]	9.9 ± 3.3 [10.0]	3.8
De novo mCSPC,e n (%)	269 (37.7)	64 (44.1)	13.2	274 (38.3)	52 (35.6)	5.8
Concurrent use of ADT with index ARPI,f n (%)	702 (98.3)	140 (96.6)	11.2	700 (98.1)	141 (97.4)	4.3
Cumulative duration of ADT use, months,g mean ± SD [median]	9.1 ± 11.7 [4.3]	7.8 ± 11.3 [2.5]	11.8	9.0 ± 11.6 [4.2]	9.0 ± 11.7 [3.8]	0.6
Prior use of first-generation ARPIs,h n (%)	75 (10.5)	16 (11.0)	1.7	75 (10.5)	15 (10.7)	0.4
Most recent PSA level,i ng/ml, n (%)
> 0.2 to ≤ 2	246 (34.5)	46 (31.7)	5.8	243 (34.1)	46 (31.7)	5.1
> 2 to ≤ 10	237 (33.2)	39 (26.9)	13.8	232 (32.5)	51 (35.2)	5.8
> 10 to < 20	68 (9.5)	17 (11.7)	7.2	70 (9.8)	14 (9.6)	0.7
≥ 20	163 (22.8)	43 (29.7)	15.6	169 (23.6)	34 (23.5)	0.4
Initial Gleason score,j n (%)
≤ 6	52 (7.3)	8 (5.5)	7.2	51 (7.1)	12 (8.0)	3.1
7	186 (26.1)	36 (24.8)	2.8	187 (26.1)	44 (30.5)	9.8
8	100 (14.0)	18 (12.4)	4.7	98 (13.8)	18 (12.7)	3.2
9	134 (18.8)	34 (23.4)	11.5	139 (19.4)	28 (19.3)	0.4
10	17 (2.4)	3 (2.1)	2.1	17 (2.3)	2 (1.5)	6.1
Unknown	225 (31.5)	46 (31.7)	0.5	223 (31.2)	41 (28.0)	7.0

ADT, androgen deprivation therapy; ARPI, androgen receptor pathway inhibitor; CCI, Charlson Comorbidity Index; IQR, interquartile range; KRD, Komodo Research Database; mCSPC, metastatic castration-sensitive prostate cancer; PC, prostate cancer; PSA, prostate-specific antigen; SD, standard deviation

(a) Of note, the number of patients reported in this weighted population represents the sum of weights for the corresponding non-weighted patients, rounded to the nearest integer. The proportions displayed were calculated before the rounding and may be slightly different than if they had been calculated based on rounded numbers

(b) Race is self-reported and augmented with modeling by the KRD, yielding high concordance with US Census data

(c) Other race included Hispanic or Latino, Asian or Pacific Islander, or other not specified

(d) Types of metastases were defined at any time before (and including) the index date. Types of metastases were not mutually exclusive

(e) De novo mCSPC was defined as ≤ 180 days between first observed PC diagnosis and date of metastasis

(f) Concurrent use was defined as having a record for an ADT agent from 180 days before or after the index date

(g) Cumulative duration of ADT was evaluated among 607 apalutamide and 117 darolutamide patients with ≥ 90 days of continuous ADT use at any time before the index date

(h) Prior use of first-generation ARPI was defined as any prescription for bicalutamide, nilutamide, or flutamide in the 12 months preceding the index date

(i) Most recent PSA level was assessed within 13 weeks before or on the index date

(j) Gleason score was evaluated at any time before and including the index date

For the OS analysis, 1460 patients were included in the apalutamide cohort and 287 in the darolutamide cohort (Fig. 2). Following weighting, baseline characteristics were well balanced between the two cohorts (Table 2 and Figure S8 in the Supplementary Material). The mean (median) age was 73.5 (74.0) years in the apalutamide cohort and 73.7 (74.0) years in the darolutamide cohort. More than half of all patients were White (apalutamide: 59.6%; darolutamide: 60.0%), while 21.5% and 22.4% were Black, respectively. Patients had a median of 4.1 months (apalutamide cohort) and 2.9 months (darolutamide cohort) between the onset of metastasis and the index date. Most patients in their respective cohorts were treated with standard dosing (apalutamide 240 mg total daily dose: 98.9%; darolutamide 1200 mg total daily dose: 96.4%).

Fig. 2.

Sample selection of patients with mCSPC treated with apalutamide without docetaxel and darolutamide without docetaxel—OS analysis. ARPI, androgen receptor pathway inhibitor; mCSPC, metastatic castration-sensitive prostate cancer; OS, overall survival; PARP, poly ADP-ribose polymerase; PPS, Precision Point Specialty Analytics

Table 2.

Patient characteristics—OS analysis

	Unweighted population			Weighted populationa
	Apalutamide	Darolutamide	Standardized difference, %	Apalutamide	Darolutamide	Standardized difference, %
	N = 1460	N = 287		N = 1460	N = 287
Age, years, mean ± SD [median]	73.8 ± 8.1 [74.0]	71.4 ± 9.2 [72.0]	27.3	73.5 ± 8.2 [74.0]	73.7 ± 8.5 [74.0]	2.6
Race,b n (%)
White	873 (59.8)	167 (58.2)	3.3	870 (59.6)	172 (60.0)	0.8
Black	312 (21.4)	61 (21.3)	0.3	313 (21.5)	64 (22.4)	2.4
Otherc	198 (13.6)	39 (13.6)	0.1	197 (13.5)	36 (12.4)	3.4
Unknown	77 (5.3)	20 (7.0)	7.1	79 (5.4)	15 (5.2)	1.2
Geographic region, n (%)
South	887 (60.8)	146 (50.9)	20.0	870 (59.6)	174 (60.6)	2.0
Midwest	351 (24.0)	58 (20.2)	9.2	344 (23.6)	66 (23.0)	1.3
Northeast	130 (8.9)	50 (17.4)	25.4	143 (9.8)	27 (9.5)	1.1
West	92 (6.3)	33 (11.5)	18.3	103 (7.0)	20 (6.9)	0.5
Payer, n (%)
Medicare	1179 (80.8)	209 (72.8)	18.9	1166 (79.8)	234 (81.4)	4.1
Commercial	223 (15.3)	63 (22.0)	17.2	235 (16.1)	44 (15.2)	2.4
Medicaid	35 (2.4)	13 (4.5)	11.7	38 (2.6)	6 (2.1)	3.2
Unknown	23 (1.6)	2 (0.7)	8.3	21 (1.5)	4 (1.2)	2.0
Year of index date, n (%)
2022	167 (11.4)	43 (15.0)	10.5	174 (11.9)	34 (11.8)	0.3
2023	574 (39.3)	111 (38.7)	1.3	571 (39.1)	112 (39.0)	0.3
2024	552 (37.8)	87 (30.3)	15.9	538 (36.8)	107 (37.3)	1.0
2025	167 (11.4)	46 (16.0)	13.4	177 (12.1)	34 (11.9)	0.7
Time between metastasis and index date, months, median [IQR]	4.3 [1.1, 17.3]	2.6 [1.0, 13.7]	8.8	4.1 [1.1, 17.1]	2.9 [1.0, 16.3]	3.4
Time between PC diagnosis and index date, months, median [IQR]	54.3 [12.7, 97.5]	40.9 [4.2, 89.3]	18.9	51.7 [11.5, 95.9]	62.6 [8.7, 94.8]	3.2
Types of metastasis,d n (%)
Visceral metastasis	239 (16.4)	76 (26.5)	24.8	258 (17.6)	53 (18.5)	2.1
Bone metastasis	748 (51.2)	197 (68.6)	36.1	783 (53.6)	152 (52.9)	1.4
Nodal metastasis	745 (51.0)	128 (44.6)	12.9	731 (50.1)	137 (47.8)	4.6
Metastasis in multiple sites	386 (26.4)	119 (41.5)	32.1	415 (28.4)	83 (29.0)	1.2
Quan-CCI, mean ± SD [median]	10.1 ± 3.4 [10.0]	10.2 ± 3.3 [10.0]	1.2	10.1 ± 3.4 [10.0]	10.0 ± 3.4 [9.0]	2.9
De novo mCSPC,e n (%)	553 (37.9)	135 (47.0)	18.6	570 (39.1)	106 (37.0)	4.3
Concurrent use of ADT with index ARPI,f n (%)	1410 (96.6)	273 (95.1)	7.3	1407 (96.4)	274 (95.6)	3.9
Cumulative duration of ADT use,g months, mean ± SD [median]	13.1 ± 16.4 [7.2]	10.6 ± 14.8 [5.0]	16.3	12.9 ± 16.2 [7.0]	13.2 ± 18.0 [6.9]	1.4
Prior use of first-generation ARPIs,h n (%)	146 (10.0)	30 (10.5)	1.5	146 (10.0)	25 (8.9)	3.9
Most recent PSA level,i ng/ml, n (%)
≤ 0.2	374 (25.6)	47 (16.4)	22.8	355 (24.3)	62 (21.6)	6.4
> 0.2 to ≤ 2	263 (18.0)	49 (17.1)	2.5	261 (17.9)	51 (17.6)	0.8
> 2 to ≤ 10	250 (17.1)	44 (15.3)	4.9	247 (16.9)	52 (18.0)	2.8
> 10 to < 20	68 (4.7)	18 (6.3)	7.1	72 (4.9)	15 (5.4)	1.9
≥ 20	168 (11.5)	45 (15.7)	12.2	177 (12.1)	37 (12.8)	2.0
Unknown	337 (23.1)	84 (29.3)	14.1	348 (23.8)	71 (24.6)	2.0
Initial Gleason score,j n (%)
≤ 6	100 (6.8)	22 (7.7)	3.2	102 (7.0)	21 (7.5)	1.8
7	382 (26.2)	63 (22.0)	9.9	376 (25.8)	84 (29.2)	7.6
8	206 (14.1)	38 (13.2)	2.5	205 (14.0)	41 (14.2)	0.6
9	281 (19.2)	64 (22.3)	7.5	286 (19.6)	52 (18.0)	4.1
10	31 (2.1)	7 (2.4)	2.1	31 (2.2)	5 (1.9)	2.1
Unknown	460 (31.5)	93 (32.4)	1.9	459 (31.4)	84 (29.3)	4.7

ADT, androgen deprivation therapy; ARPI, androgen receptor pathway inhibitor; CCI, Charlson Comorbidity Index; IQR, interquartile range; KRD, Komodo Research Database; mCSPC, metastatic castration-sensitive prostate cancer; PC, prostate cancer; PSA, prostate-specific antigen; OS, overall survival; SD, standard deviation

(a) Of note, the number of patients reported in this weighted population represents the sum of weights for the corresponding non-weighted patients, rounded to the nearest integer. The proportions displayed were calculated before the rounding and may be slightly different than if they had been calculated based on rounded numbers

(b) Race is self-reported and augmented with modeling by the KRD, yielding high concordance with US Census data

(c) Other race included Hispanic or Latino, Asian or Pacific Islander, or other not specified

(d) Types of metastases were defined at any time before (and including) the index date. Types of metastases were not mutually exclusive

(e) De novo mCSPC was defined as ≤ 180 days between first observed PC diagnosis and date of metastasis

(f) Concurrent use was defined as having a record for an ADT agent from 180 days before or after the index date

(g) Cumulative duration of ADT was evaluated among 1250 apalutamide and 226 darolutamide patients with ≥ 90 days of continuous ADT use at any time before the index date

(h) Prior use of first-generation ARPI was defined as any prescription for bicalutamide, nilutamide, or flutamide in the 12 months preceding the index date

(i) Most recent PSA level was assessed within 13 weeks before or on the index date

(j) Gleason score was evaluated at any time before and including the index date

PSA Analysis

Through a maximum of 6 months following the index date, 74.6% of patients in the apalutamide cohort and 56.1% of patients in the darolutamide cohort achieved a PSA90 response while on-treatment (Fig. 3). Apalutamide patients without docetaxel had a statistically significant 49% increase in the rate of PSA90 response compared with patients initiated on darolutamide without docetaxel (weighted HR: 1.49 [95% CI 1.07, 2.07]; p = 0.017) using 6-month maximum follow-up data.

Fig. 3.

Comparison of time-to-PSA90 response between patients with mCSPC treated with apalutamide without docetaxel and darolutamide without docetaxel. ADT, androgen deprivation therapy; ARPI, androgen receptor pathway inhibitor; CCI, Charlson Comorbidity Index; CI, confidence interval; HR, hazard ratio; mCSPC, metastatic castration-sensitive prostate cancer; PC, prostate cancer; PSA, prostate-specific antigen. *Significant at the 5% level. Notes: (a) Propensity scores were generated using probability estimates from a logistic regression model using the following predictors: age, race, geographic region, payer, year of index date, time between metastasis and index date, time between PC diagnosis and index date, de novo mCSPC, ADT claim observed 180 days before or after index date, prior use of first-generation ARPI, bone antiresorptive therapy, most recent PSA level, earliest Gleason score, types of metastases (bone, nodal, visceral, and metastasis in multiple sites), and Quan-CCI score. Each patient was attributed an inverse probability of treatment weight defined as follows: 1/(propensity score) for the apalutamide cohort and 1/(1-propensity score) for the darolutamide cohort. Normalized inverse probability of treatment weights were truncated at the 95th percentiles. (b) A HR > 1 indicates that the apalutamide cohort had a higher PSA90 response rate compared to the darolutamide cohort. (c) Of note, the number of patients reported in this weighted population represents the sum of weights for the corresponding non-weighted patients, rounded to the nearest integer. The proportions displayed were calculated before the rounding and may be slightly different than if they had been calculated based on rounded numbers

Similarly, 71.0% of patients in the apalutamide cohort and 55.2% of patients in the darolutamide cohort achieved an undetectable PSA response while on-treatment through a maximum of 6 months (Fig. 4). Consistent with the PSA90 analysis, patients initiated on apalutamide without docetaxel had a statistically significant 44% increase in the rate of undetectable PSA response compared with patients initiated on darolutamide without docetaxel (weighted HR: 1.44 [95% CI 1.03, 2.02]; p = 0.032) using 6-month maximum follow-up data. No violation of the proportional hazards assumption was observed for either PSA outcome (p > 0.05).

Fig. 4.

Comparison of time-to-undetectable PSA response between patients with mCSPC treated with apalutamide without docetaxel and darolutamide without docetaxel. ADT, androgen deprivation therapy; ARPI, androgen receptor pathway inhibitor; CCI, Charlson Comorbidity Index; CI, confidence interval; HR, hazard ratio; mCSPC, metastatic castration-sensitive prostate cancer; PC, prostate cancer; PSA, prostate-specific antigen. *Significant at the 5% level. Notes: (a) Propensity scores were generated using probability estimates from a logistic regression model using the following predictors: age, race, geographic region, payer, year of index date, time between metastasis and index date, time between PC diagnosis and index date, de novo mCSPC, ADT claim observed 180 days before or after index date, prior use of first-generation ARPI, bone antiresorptive therapy, most recent PSA level, earliest Gleason score, types of metastases (bone, nodal, visceral, and metastasis in multiple sites), and Quan-CCI score. Each patient was attributed an inverse probability of treatment weight that was defined as follows: 1/(propensity score) for the apalutamide cohort and 1/(1-propensity score) for the darolutamide cohort. Normalized inverse probability of treatment weights were truncated at the 95th percentiles. (b) A HR > 1 indicates that the apalutamide cohort had a higher undetectable PSA response rate compared to the darolutamide cohort. (c) Of note, the number of patients reported in this weighted population represents the sum of weights for the corresponding non-weighted patients, rounded to the nearest integer. The proportions displayed were calculated before the rounding and may be slightly different than if they had been calculated based on rounded numbers

In addition, 6-month follow-up results were consistent for the sensitivity analysis excluding patients with docetaxel use between 90 days before or 90 days post-index (weighted HR: 1.52 [95% CI 1.08, 2.13]; p = 0.016; Figure S9 in the Supplementary Material).

Patients were treated with their index ARPI over a mean (median) of 9.8 (7.0) months in the apalutamide cohort and 9.2 (5.7) months in the darolutamide cohort. During the on-treatment period, 79.9% of patients in the apalutamide cohort and 74.7% in the darolutamide cohort had ≥ 1 PSA measurement. The mean (median) number of PSA measurements per year was 6.0 (4.3) and 6.0 (4.4), respectively.

OS Analysis

The mean (median) observation period was 17.7 (17.5) months in patients who initiated apalutamide and 17.3 (17.5) months in those who initiated darolutamide. Patients were treated with their index ARPI continuously without a gap > 90 days of treatment over a mean (median) of 10.3 (8.4) months in the apalutamide cohort and 10.7 (7.9) months in the darolutamide cohort. The mean (median) MPR was 91.9% (98.4%) among patients who initiated apalutamide and 90.2% (96.4%) among those who initiated darolutamide.

Through a maximum of 24 months following the index date, 92.1% of patients in the apalutamide cohort and 85.8% of patients in the darolutamide cohort remained alive (Fig. 5). Apalutamide patients without docetaxel had a statistically significant 51% reduction in the rate of death compared with patients initiated on darolutamide without docetaxel (weighted HR: 0.49 [95% CI 0.30, 0.83]; p = 0.007) using 24-month maximum follow-up data. Using all available follow-up, apalutamide patients without docetaxel had a 49% reduction in the rate of death compared with patients initiated on darolutamide without docetaxel (weighted HR: 0.51 [95% CI 0.32, 0.82]). No violation of the proportional hazards assumption was observed (p > 0.05).

Fig. 5.

Comparison of OS between patients with mCSPC initiated on apalutamide without docetaxel and darolutamide without docetaxel. ADT, androgen deprivation therapy; ARPI, androgen receptor pathway inhibitor; CCI, Charlson Comorbidity Index; CI, confidence interval; HR, hazard ratio; OS, overall survival; mCSPC, metastatic castration-sensitive prostate cancer; PC, prostate cancer; PSA, prostate-specific antigen. *Significant at the 5% level. Notes: (a) Propensity scores were generated using probability estimates from a logistic regression model using the following predictors: age, race, geographic region, payer, year of index date, time between metastasis and index date, time between PC diagnosis and index date, de novo mCSPC, ADT claim observed 180 days before or after index date, prior use of first-generation ARPI, bone antiresorptive therapy, most recent PSA level, earliest Gleason score, types of metastases (bone, nodal, visceral, and metastasis in multiple sites), and Quan-CCI score. Each patient was attributed an inverse probability of treatment weight that was defined as follows: 1/(propensity score) for the apalutamide cohort and 1/(1-propensity score) for the darolutamide cohort. Normalized inverse probability of treatment weights were truncated at the 95th percentiles. (b) A HR < 1 indicates that the apalutamide cohort had a lower mortality rate compared to the darolutamide cohort. (c) Of note, the number of patients reported in this weighted population represents the sum of weights for the corresponding non-weighted patients, rounded to the nearest integer. The proportions displayed were calculated before the rounding and may be slightly different than if they had been calculated based on rounded numbers

In addition, 24-month follow-up results were consistent for the sensitivity analysis excluding patients with docetaxel use between 90 days before and 90 days post-index (weighted HR: 0.49 [95% CI 0.29, 0.84]; p = 0.009; Figure S10 in the Supplementary Material).

Discussion

This study is the first to compare the real-world effectiveness of apalutamide and darolutamide, without docetaxel, in the mCSPC population. Over a maximum of 6 months following treatment initiation, patients receiving apalutamide had a 49% higher rate of achieving a deep PSA90 reduction than those receiving darolutamide, despite similar annual testing rates. Additionally, the mortality rate was 51% lower with apalutamide than with darolutamide through a maximum of 24 months post-treatment initiation. Differences in survival between apalutamide and darolutamide patients who initiated treatment without docetaxel remained consistent using all available follow-up data. In the absence of prior comparative data between these two ARPIs in the mCSPC setting, these results support the corresponding pivotal trial readouts and expand the generalizability of the data to real-world clinical practice.

The PSA90 and OS rates of apalutamide and darolutamide in this study are generally comparable to those reported in the TITAN and ARANOTE trials [4, 10]. In the post hoc analysis of TITAN, patients treated with apalutamide had a PSA90 response rate of 68% by 6 months, and the 24-month OS rate in the main analysis was 82.4% [4, 11]. Meanwhile, in the ARANOTE trial, darolutamide was associated with a 24-month OS rate of 79.8% (PSA90 response was not evaluated) [10]. The OS endpoint, however, was not achieved in the ARANOTE trial for darolutamide [10]. These rates from the clinical trials are slightly lower than those from the current study (apalutamide: PSA90 of 74.6% and OS of 92.1%; darolutamide: OS of 85.8%), which may be explained by the inherent differences between rigid trial conditions and real-world practice, as well as the much larger proportion of patients with de novo metastasis in the trials (71.1–78.3%) relative to the present analysis (35.6–40.1%) [4, 10], given de novo mCSPC is associated with a poorer prognosis [22]. Relatedly, patients treated with apalutamide and darolutamide enrolled in the TITAN and ARANOTE trials, respectively, had elevated Gleason score > 7 more frequently (66.9–69.7%) than in the current study (46.2–52.4%, among those with non-missing initial Gleason score) and had a large majority of patients with high-volume metastasis, as assessed per CHAARTED criteria (61.9–70.6%) [4, 10], both prognostic of poorer clinical outcomes in patients with mCSPC [23]. High-volume metastasis was not assessed in the current study, but few patients had visceral metastasis (15.3–18.5%) or multiple-site metastasis (28.3–29.0%), both indicators of high-volume metastatic disease.

As there are currently no other comparisons of PSA response and OS between apalutamide and darolutamide in the mCSPC literature, a review of real-world analyses assessing the two ARPIs separately may help to contextualize the current findings. In one study of 242 patients with mCSPC receiving apalutamide across 24 European urology/oncology centers, Boegemann et al. reported a PSA90 response rate of 77.6% by 6 months following apalutamide initiation [24], similar to the 74.6% observed in the present study. In a separate analysis of 432 patients receiving apalutamide in Spain, approximately 86.2–94.9% achieved a PSA90 response by 6 months; this range is higher than our estimate, potentially due to differences in patient selection and characteristics, as the Spanish study selected patients based solely on the diagnosis of mCSPC and reported limited baseline characteristics in the publication [25]. Regarding OS, 24-month rates for apalutamide in real-world studies have ranged from 66 to 92% [26–28], with our estimate of 92.3% observed in this study similar to the higher end of this range. Given the observation period of the current study began in August 2022, following darolutamide approval with docetaxel for mCSPC, higher OS rates for apalutamide may be explained by shorter available follow-up and additional administrative censoring than earlier studies that may have identified apalutamide patients as early as in 2019.

In contrast to apalutamide, real-world studies evaluating clinical outcomes with darolutamide in mCSPC without docetaxel are more limited [29, 30]. Two small studies of patients with mCSPC receiving darolutamide in China reported PSA90 response rates of 91.8% and 81.5% by 6 months [29, 30]. These rates are higher than those of the current analysis (56.1%), potentially because the Chinese studies specifically restricted the small cohorts (n < 100) to patients with good performance status, and only 40.6% to 45.1% of patients treated with darolutamide were qualified as having comorbidities [29, 30]. Given the average Quan-Charlson Comorbidity Index score was 9.9 in the darolutamide cohort of the current study (with ≥ 4 or 5 typically considered as a severe/high-risk comorbidity profile) [31, 32], these patients may have been less fit overall than those included in the Chinese studies, which may explain the better responses seen in the prior analyses.

Although not directly assessed in the present study, there is a wealth of literature characterizing the relationship between deep, early PSA response and improved survival outcomes [11, 12, 33–36]. Indeed, post hoc analyses of both the TITAN and ARANOTE clinical trials have shown that achievement of deep PSA declines (defined as PSA90 response or PSA ≤ 0.2 ng/ml in TITAN and PSA < 0.2 ng/ml or < 0.02 ng/ml in ARANOTE) with apalutamide or darolutamide, respectively, was significantly associated with improved OS [11, 12]. Similar trends have been observed in real-world clinical practice [33–36]. For instance, in a recent analysis using the same linked PPS and KRD data, patients with mCSPC achieving a PSA90 response by 6 months of ARPI initiation had a significantly higher 36-month OS rate than those without an early PSA90 response (weighted HR: 0.40 [95% CI 0.31, 0.50]; p < 0.001) [33]. Taken together, in light of the now well-established synergistic relationship between PSA response and survival, the present study confirms that the significant improvements seen with apalutamide in both clinical outcomes in the TITAN trial also translate to real-world effectiveness in this first comparison with darolutamide.

Although longer treatment duration with ARPIs has been associated with improved clinical outcomes among patients with advanced PC [37], the median duration of treatment was similar for apalutamide (8.4 months) and darolutamide (7.9 months) patients in the intention-to-treat OS analysis in the current study, and thus not likely to explain differences in survival between the treatment cohorts. The median MPR, as a measure of treatment adherence, was also similar between the apalutamide (98.4%) and darolutamide (96.4%) cohorts. The observed real-world duration of treatment was shorter than those reported in TITAN for apalutamide (median 39.3 months) [6] and ARANOTE for darolutamide (median 24.2 months at data cutoff) [10], however, given administrative right-censoring in the current study and longer follow-up available in the clinical trial settings. Although not directly assessed in this study, additional differences in dosing, as well as the greater bioavailability and longer half-life of apalutamide relative to darolutamide [38, 39], may have contributed to the observed improvements in PSA response and longer OS among patients treated with apalutamide.

To achieve sufficient follow-up to detect a meaningful difference in OS between apalutamide and darolutamide, patients in both treatment cohorts were included from the date of FDA approval for darolutamide with docetaxel in mCSPC on August 5, 2022, rather than the approval date of darolutamide without docetaxel on June 3, 2025. Although patients initiated on darolutamide without docetaxel in the current study were almost all treated before FDA approval, recent real-world data from community-based urology practices in the US have reported that among 420 patients treated with darolutamide after initial FDA approval in 2022, more than half (59.3%) were treated without docetaxel combination [40]. Given the earlier mCSPC approval for apalutamide (September 2019) and higher category of National Comprehensive Cancer Clinical Practice Guidelines in Oncology (NCCN Guidelines®) recommendations (category 1) for the treatment of mCSPC without docetaxel, it was expected that more patients would be identified as apalutamide initiators relative to darolutamide initiators using this real-world data source [41]. Furthermore, this study was adequately powered with probability > 80% to detect a true difference in both PSA90 response and OS outcomes, if present in the data.

This study adds to the growing body of real-world evidence comparing apalutamide with other ARPIs in the mCSPC setting using linked EMR and insurance claims data [13–16]. The use of both data sources enables comprehensive capture of ARPI initiation from paid claims and dispensations and identification of castration resistance across multiple indicators, ensuring an accurately defined mCSPC target population. Additionally, the linked data are ideal for evaluating both PSA response (with detailed PSA test results from the EMRs) and survival (with comprehensive mortality data from the claims, benchmarked against Centers for Disease Control estimates [17]). Based on these prior studies, apalutamide has been shown to be significantly associated with a higher likelihood of achieving an early PSA90 response and a lower mortality rate compared with enzalutamide or abiraterone acetate among patients with mCSPC [13–16]. As the treatment landscape for mCSPC continues to expand, the current study provides a first real-world comparison between apalutamide and darolutamide and, together with the existing analyses, further reinforces the clinical benefit of apalutamide relative to other ARPIs in the mCSPC setting.

The study findings are subject to some limitations. As with all EMR- or claims-based analyses, any miscoding of data may have resulted in selection or information biases. The apalutamide and darolutamide cohorts were balanced using IPTW based on measured potential confounders. Any remaining unmeasured confounders may have affected the analyses, although it is not possible to know the directionality of any hypothetical bias resulting from unmeasured confounding. More specifically, disease volume was not available in the data and could not be balanced using IPTW. However, visceral metastasis and metastasis in multiple sites, which are related to high-volume disease, were balanced between cohorts. Given the data available for analyses at the time the study was conducted ended in the first half of 2025, all darolutamide patients in the PSA analysis and nearly all darolutamide patients in the OS analysis were treated off-label before FDA approval for darolutamide treatment without docetaxel. Potential measured differences between the cohorts, such as prior treatments, metastasis duration, patient comorbidity profile, and Gleason score, were balanced with IPTW; however, given the nature of observational data, it is impossible to rule out the possibility of unmeasured clinical factors that are associated with both the treatment prescribed and study outcomes. Although unmeasured confounding cannot be ruled out, resulting biases are expected to be limited. Given the use of open-source claims and EMR data, combination treatment with docetaxel may not be completely captured if administered outside the PPS network or hospitals that do not provide data to KRD. For the PSA analysis, PSA testing conducted outside the PPS network was not captured, which may have led to an underestimation of PSA response rates; however, this would have affected both cohorts equally and thus would not have changed the conclusions of the study. Similarly, for the OS analysis, a small portion of all oncology-related deaths reported to the Centers for Disease Control were not captured by KRD (< 10%), which may have led to an underestimation of mortality; this also would have affected both cohorts equally. Although patient-reported outcomes, including quality of life, are important therapeutic considerations for the management of patients with mCSPC, this information was not available in the linked EMR and administrative claims data source and was not an endpoint of this study. Further research is needed to understand real-world patient-reported outcomes among patients treated with apalutamide or darolutamide for mCSPC. Lastly, given that MPR is a pharmacy claims-based measure of adherence, this metric reflects the quantity of ARPI pills received and may not reflect the actual amount of medication used by the patient, and adherence may be overestimated.

Conclusions

This real-world study of patients with mCSPC demonstrated a significant increase in the rate of PSA response and a significantly lower rate of mortality for patients initiating apalutamide without docetaxel relative to those initiating darolutamide without docetaxel, highlighting both the short- and long-term benefits of apalutamide in the mCSPC setting, even without treatment intensification with docetaxel.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contribution

Mehmet A. Bilen, Gordon Brown, Mukul Singhal, Courtney D. Longfield, Carmine Rossi, Dominic Pilon, and Benjamin Lowentritt made substantial contributions to the conception or design of the study, or the acquisition, analysis, or interpretation of data, drafting the manuscript and revising it critically for important intellectual content, and provided final approval of this version to be published and agree to be accountable for all aspects of the work.

Funding

This study and the journal’s Rapid Service and Open Access fees were funded by Johnson & Johnson. The study sponsor was involved in several aspects of the research, including the study design, interpretation of data, and writing of the manuscript.

Data Availability

The data that support the findings of this study are available from Komodo Health Solutions and PPS Analytics. Restrictions apply to the availability of these data, which were used under license for this study. As such, the authors had full permission to access and use the two databases.

Declarations

Conflict of Interest

Mehmet A. Bilen is an employee of Winship Cancer Institute of Emory University and has received consulting fees from Johnson & Johnson. Gordon Brown is an employee of New Jersey Urology and has received consulting fees from Johnson & Johnson. Mukul Singhal and Courtney D. Longfield are employees and stockholders of Johnson & Johnson. Carmine Rossi and Dominic Pilon are employees of Analysis Group, Inc., a consulting company that has provided paid consulting services to Johnson & Johnson, which funded the development and conduct of this study and manuscript. Benjamin Lowentritt is an employee of Chesapeake Urology and has received consulting fees from Johnson & Johnson.

Ethical Approval

The study was considered exempt research under 45 CFR § 46.104(d)(4) as it involved only the secondary use of data that were de-identified in compliance with the Health Insurance Portability and Accountability Act (HIPAA), specifically, 45 CFR § 164.514. The study was performed in accordance with the Declaration of Helsinki of 1964 and its later amendments.