Early prostate-specific antigen response among Black and non-Black patients with advanced prostate cancer treated with apalutamide

In 2021, an estimated 248,000 new cases of prostate cancer (PC) were diagnosed and over 34,000 people experienced PC-related mortality in the USA [1]. Among the factors that affect the risk of developing PC (e.g., advancing age and genetic abnormalities) [2], race and ethnicity have an important role [3]. Although one in eight men will be diagnosed with PC during their lifetime, Black men have a 76% higher incidence rate of PC compared with non-Hispanic White men [4]. In addition to a higher incidence rate, additional studies have reported higher age-adjusted PC-related mortality rates for Black men relative to non-Black men in the USA [5,6].

Current guidelines indicate that men with a diagnosis of nonmetastatic castration-resistant PC (nmCRPC) or metastatic castration-sensitive PC (mCSPC) should be treated with a next-generation androgen receptor inhibitor (i.e., apalutamide, enzalutamide or darolutamide) or an androgen biosynthesis inhibitor (i.e., abiraterone acetate) in combination androgen deprivation therapy (ADT), all which have been proven to be superior to ADT alone [7]. Furthermore, the addition of docetaxel to approved androgen receptor/biosynthesis inhibitor + ADT regimens has also demonstrated survival benefit among patients with mCSPC with high-volume disease [8]. Apalutamide, an important next-generation androgen receptor inhibitor therapeutic option for patients with advanced PC, was approved by the US FDA for patients with nmCRPC in February 2018 [9] and for patients with mCSPC in September 2019 [10]. FDA approval of apalutamide was based on results from two phase III trials: SPARTAN (NCT01946204) and TITAN (NCT02489318). Results from the SPARTAN trial showed that among patients with nmCRPC, apalutamide and concomitant ADT administration was associated with >70% reduction in metastasis progression or death compared with treatment with ADT alone, resulting in a median increase of 24.3 months of metastasis-free survival for patients treated with apalutamide and ADT [11]. Among patients with mCSPC enrolled in the TITAN trial, treatment with apalutamide and ADT was associated with >50% reduction in radiographic progression or death compared with treatment with ADT alone [5].

Although the effectiveness of apalutamide has been demonstrated in clinical trials, there is a paucity of real-world data describing treatment responses among patients initiating apalutamide, particularly among minority populations, who tend to be underrepresented in clinical trials [12]. This study describes real-world attainment of early outcomes (i.e., changes in PSA and PSA response) as observed in Black or African–American (hereafter ‘Black’) and non-Black patients diagnosed with nmCRPC or mCSPC and treated with apalutamide in urology practices throughout the USA.

Patients & methods

Data source

This study used clinical data (Precision Point Specialty [PPS], Cleveland, OH, USA) that were collected between 1 February 2017 and 5 March 2021 during routine care from 69 community-based urology practices throughout the USA providing in-office dispensation information. PPS data are extracts from electronic medical records and contain information on patient demographics and clinical variables, including dates of metastasis, indicators for castration resistance, prescriptions and laboratory assessments. All data were deidentified and comply with the requirements of the Health Insurance Portability and Accountability Act of 1996. As a result, approval from an institutional review board was not required to conduct this study.

Patient selection criteria

Male patients ≥18 years of age were included in the current study if they were identified as having nmCRPC or mCSPC and had at least one dispensation for apalutamide within the study period (14 February 2017 to 5 March 2021). The index date was defined as the date of the first dispensation for apalutamide. Inclusion in the nmCRPC cohort required evidence of castration resistance, as defined in the section below, prior to or on the index date, in the absence of documented metastases. Inclusion in the mCSPC cohort required evidence of metastatic disease prior to or on the index date, in the absence of castration resistance. Metastatic disease was defined based on a PPS-provided indicator variable for the presence of bone, nodal or visceral metastasis based on diagnoses or data entry. Patients were excluded if they had use of any other next-generation androgen receptor or biosynthesis inhibitor prior to or on the index date, initiated apalutamide prior to the FDA approval for the patient’s specific indication (i.e., 14 February 2018 for nmCRPC or 17 September 2019 for mCSPC [9,10]), had <12 months of clinical activity before the index date or had evidence of prior use of radiopharmaceuticals.

Identification of castration resistance

Identification of castration resistance was based on the combination of a previously published algorithm by Freedland and colleagues [13] and the use of a PPS-provided indicator variable for castration resistance. From the published algorithm [13], the three criteria that were considered evidence of castration resistance were: one or more International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) codes prior to or on the index date indicating hormone resistance (i.e., Z19.2); rising PSA (i.e., one or more claims for surgical castration and two or more PSA test results that include one nadir and one post-nadir in addition to one or more rises in PSA of ≥25% with an absolute increase of ≥2 ng/ml after nadir) after surgical castration; and rising PSA (two or more PSA test results including one nadir and one post-nadir within the same episode of continuous ADT and one or more rises in PSA of ≥25% with an absolute increase of ≥2 ng/ml after nadir while within the same episode of continuous ADT) after medical castration (i.e., one or more episodes of ≥90 days of continuous ADT during the study period) [13]. In the absence of evidence of castration resistance, patients were assumed to be castration sensitive.

Study design

A retrospective, longitudinal study among patients with nmCRPC or mCSPC who initiated apalutamide was used. Baseline patient characteristics were evaluated in the 12 months preceding the index date. The follow-up period spanned from the index date to the earliest date of apalutamide discontinuation (using a 90-day treatment gap to define discontinuation), initiation of a new next-generation androgen receptor or biosynthesis inhibitor or the use of radiopharmaceuticals, end of clinical activity (including death), or end of data availability (5 March 2021), whichever occurred first.

Study outcomes

The following PSA response outcomes were assessed during the follow-up period among patients who had one or more PSA tests within the 13 weeks before and including the index date (baseline) and one or more PSA tests at least 8 weeks after the index date (follow-up): proportions of patients who achieved ≥50% reduction in PSA (PSA50) and ≥90% reduction in PSA (PSA90) from the most recent baseline PSA value. In addition, among those who had a most recent baseline PSA value >0.2 ng/ml, the proportion of patients who achieved a decline to ≤0.2 ng/ml was also assessed (PSA ≤0.2). Time to PSA50, PSA90 and PSA ≤0.2 response observed any time after the index date was also assessed among patients who had one or more PSA tests within the 13 weeks before and including the index date and was not limited to those with a PSA test at least 8 weeks after index. Time to PSA ≤0.2 response was similarly restricted, but only to patients with a most recent baseline PSA value >0.2 ng/ml.

Statistical analysis

Baseline characteristics (i.e., age, prior use of ADT, PSA testing, PSA doubling time [14], testosterone and Gleason score assessments) were reported using means, standard deviations and medians for continuous variables, and frequencies and percentages for categorical variables. The median time to PSA50, PSA90 and PSA ≤0.2 response was assessed using Kaplan–Meier analysis. Baseline characteristics and PSA response outcomes were reported separately for patients with nmCRPC and patients with mCSPC, with stratifications by race (i.e., overall, Black and non-Black) for both cohorts. All analyses were descriptive with no statistical comparisons conducted between study cohorts or race stratifications. All analyses were performed using SAS^® Enterprise Guide 7.1 (SAS Institute, Inc., NC, USA).

Results

Baseline characteristics

A total of 313 patients with nmCRPC and 260 patients with mCSPC initiated apalutamide and met all the study inclusion criteria (Figure 1). Patient baseline characteristics are summarized in Table 1. Of the 313 patients in the nmCRPC cohort, 56 were Black (17.9%), 233 were non-Black (74.4%), and the remainder had an unknown or missing ethnicity. Of the 260 patients in the mCSPC cohort, 44 were Black (16.9%), 193 were non-Black (74.2%), and the remainder had an unknown or missing ethnicity. Overall, patients in the nmCRPC cohort were slightly older than patients in the mCSPC cohort (mean age: 79.0 vs 76.1 years). Most patients in both cohorts had prior use of ADT (nmCRPC: 97.4%; mCSPC: 92.3%). Similarly, most patients in both cohorts had a PSA test during the 12-month baseline period (nmCRPC: 98.1%; mCSPC: 90%). In the nmCRPC cohort, the baseline mean PSA level taken most recently before initiating apalutamide was 9.9 ng/ml, with a numerically higher mean PSA level reported for Black patients as compared with non-Black patients (15.0 vs 8.6 ng/ml). The mean most recent PSA level in the baseline period for the mCSPC cohort was 18.0 ng/ml, also with a numerically higher mean level reported for Black patients compared with non-Black patients (32.9 vs 15.7 ng/ml).

PSA doubling time using two or more PSA tests during the baseline period could be calculated for 34.8% of patients in the nmCRPC cohort and 9.2% of patients in the mCSPC cohort. For the nmCRPC cohort, the mean PSA doubling time was similar for both Black patients (10.2 months) and non-Black patients (10.2 months). For the mCSPC cohort, the mean PSA doubling time was numerically longer among Black patients (12.4 months) relative to non-Black patients (5.7 months). Over half of all patients in the nmCRPC cohort (72.8%) and mCSPC cohort (58.1%) had baseline testosterone tests. The most recent mean testosterone level for the nmCRPC cohort was 19.6 ng/dl, with numerically lower values observed for Black patients (17.9 ng/dl) relative to non-Black patients (20.5 ng/dl). The most recent mean testosterone level for the mCSPC cohort was 103.2 ng/dl, with similar values observed for Black (104.7 ng/dl) and non-Black patients (103.9 ng/dl).

Continuous ADT use

Most patients in the nmCRPC cohort (78.6%) and mCSPC cohort (73.8%) concomitantly used ADT during continuous apalutamide treatment. Similar proportions of Black (78.6%) and non-Black (79.8%) patients with nmCRPC and Black (81.8%) and non-Black (71.5%) patients with mCSPC concomitantly used ADT during continuous apalutamide treatment.

PSA outcomes

During the follow-up period, on average a PSA test occurred every 14.4 weeks among patients with nmCRPC (Black: every 15.8 weeks; non-Black: every 13.7 weeks) and every 11.8 weeks among patients with mCSPC (both Black and non-Black: every 11.8 weeks). Among the subset of patients with a baseline PSA measured within the 13 weeks before and including the index date and with one or more follow-up PSA tests at least 8 weeks after apalutamide initiation, PSA50 response was achieved by 82.3% of patients in the nmCRPC cohort and 86.7% of patients in the mCSPC cohort. The proportions of patients achieving PSA50 response were numerically similar between Black and non-Black patients in both cohorts (Figure 2). PSA90 response following apalutamide initiation was achieved by 66% of patients in the nmCRPC cohort and 73.5% of patients in the mCSPC cohort. In the nmCRPC cohort, PSA90 response was achieved by a numerically higher proportion of Black patients relative to non-Black patients. However, in the mCSPC cohort, PSA90 response rates were numerically similar between Black and non-Black patients treated with apalutamide (Figure 2). PSA ≤0.2 was achieved by 63% of patients in the nmCRPC cohort and 77.7% of patients in the mCSPC cohort. The proportions of patients achieving PSA ≤0.2 response were numerically similar between Black and non-Black patients (Figure 2).

In the Kaplan–Meier analysis, the median time to achieving a PSA50 response after apalutamide initiation was 2.4 months for patients with nmCRPC (Figure 3A) and 1.8 months for patients with mCSPC (Figure 3B). The median times to PSA50 responses were numerically similar between Black and non-Black patients. The median time to achieving a PSA90 response after apalutamide initiation was 4.9 months for patients with nmCRPC (Figure 4A) and 2.7 months for patients with mCSPC (Figure 4B). In the nmCRPC cohort, Black patients treated with apalutamide tended to achieve a PSA90 response faster than non-Black patients (Figure 4A). The median time to PSA90 response was numerically similar between Black and non-Black patients with mCSPC (Figure 4B). The median time to achieving a PSA ≤0.2 after apalutamide initiation was 6.7 months for patients with nmCRPC (Figure 5A) and 3.5 months for patients with mCSPC (Figure 5B). For both cohorts, the median times to achieve PSA ≤0.2 were numerically similar between Black and non-Black patients.

Discussion

While the safety and efficacy of apalutamide have been demonstrated in the clinical trial setting [11,15,16], this is among the first studies demonstrating early clinical response with apalutamide in routine clinical practice. Results from this real-world study have shown that the effectiveness of apalutamide on PSA reduction is consistent with that observed in phase III clinical trials (SPARTAN and TITAN) [15,16]. Additionally, given reports of potential differences in patient clinical characteristics and clinical responses to treatment by patients with advanced PC with different ethnicities [3,17], this study provides information about early clinical response to apalutamide in Black patients, a population which is typically underrepresented in clinical trials (6% in SPARTAN; 2% in TITAN) [10,11]. Black patients who initiate apalutamide may be diagnosed with more advanced disease (e.g., higher Gleason score) and may have additional prognostic factors which may decrease the effectiveness of advanced PC therapy, although this was not observed in this current study [3]. The disproportionate impact of PC among Black patients suggests that there could be a higher clinical burden in this demographic [4]. Results from an analysis of data from the Surveillance, Epidemiology and End Results program (Veterans Affairs and National Cancer Institute) have demonstrated that when Black men with PC have similar access to care and standardized treatment to that received by non-Black men, the stage-for-stage PC-specific mortality is similar, highlighting the importance of treatment and elimination of disparities for this population [5].

Results from the current study have demonstrated similar early PSA response for apalutamide among Black and non-Black patients; therefore, improved access to treatment for nmCRPC and mCSPC may reduce overall outcome disparities between these two populations [18]. Indeed, a recent analysis of electronic health records in the USA has found that Medicaid expansion was associated with a 6.9% increase in the proportion of African–American patients with advanced or metastatic cancer who initiated treatment within 30 days of diagnosis [19]. Expansion of insurance coverage, specifically Medicaid, may increase access to screening and shorten the time to diagnosis and receipt of treatment for patients with advanced PC, particularly among those who may be from historically underprivileged communities [20,21]. Policy-makers should continue to work on improving access to urology and oncology care for Black and other underprivileged patient populations with nmCRPC and mCSPC.

Findings from the present study clearly support that apalutamide provides consistent outcomes in the clinical management of PC between Black and non-Black patients, with regard to the attainment of early PSA response in the real-world and clinical trial settings. More specifically, in the current study, real-world PSA50 response rates assessed at least 8 weeks after apalutamide initiation were 82.3% among patients with nmCRPC and 86.7% among patients with mCSPC; they appeared generally similar to results from post hoc analyses of PSA responses among patients initiated on apalutamide in the SPARTAN (nmCRPC) and TITAN (mCSPC) trials, which both reported 90% overall PSA50 response rates [15,16]. Further, PSA90 response rates were also comparable, with 66% of patients with nmCRPC and 73.5% of patients with mCSPC achieving the response in the real-world setting, compared with 62% of patients with nmCRPC in SPARTAN and 72% of patients with mCSPC in TITAN achieving this response [15,16]. Additional studies assessing real-world attainment of longer-term outcomes are equally important, but will require sufficient time since the introduction of apalutamide into clinical practice in the USA to complete. Given that survival benefits from clinical trials of apalutamide were observed after a median of 41 months of follow-up in men with nmCRPC [11] and after a median of 23 months of follow-up in men with mCSPC [10], the real-world impact of apalutamide on longer-term outcomes in advanced PC may not be evaluable until after 2–4 years of apalutamide availability in the USA (i.e., late 2021), assuming large samples of patients can be identified after treatment becomes available.

Numerous studies have demonstrated that rapid and deep PSA reduction after receiving an oral next-generation androgen receptor inhibitor or androgen biosynthesis inhibitor is associated with a greater likelihood of achieving long-term clinical success [16,22–25]. For example, a post hoc analysis of the SPARTAN trial in patients with nmCRPC who initiated apalutamide + ADT demonstrated that achieving a PSA90 response by 6 months was associated with a 59% lower risk of developing metastasis or mortality and that developing undetectable PSA (i.e., PSA ≤0.2) was associated with a 70% lower risk of developing metastasis or mortality [24]. Similarly, a post hoc analysis of patients with mCSPC who initiated apalutamide + ADT in the TITAN trial demonstrated that achieving a PSA90 response by 6 months resulted in a 54% lower risk of disease progression or mortality, with that risk decreasing to 81% if patients developed undetectable PSA by 6 months [16]. Both post hoc analyses also reported a correlation between an increased depth of PSA response by 6 months and reductions in disease progression and mortality. In SPARTAN, for example, patients who achieved a PSA90 response had a median metastasis-free survival time of 40.5 months, compared with 29.1 months for those who achieved a response between ≥50 and <90% [24]. Other studies in advanced PC have also shown that early, deep PSA response may be a prognostic indicator for better long-term outcomes in men with advanced PC [22,23,25].

The results from this study should be considered in the context of several factors. First, while all Black patients treated with apalutamide in the 69 practices were studied, few Black patients overall were treated in these practices (56 Black patients with nmCRPC and 44 Black patients with mCSPC), highlighting the potential disparities in treatment between Black and non-Black patients. Second, this analysis included data from patients treated in an extensive network of community-based urology practices with in-office dispensations, but may not be generalizable to patients receiving care in academic-based practices or large tertiary care centers. Third, PSA testing frequency in the real world varies compared with that observed in clinical trials, and may be a factor that influences time to PSA response; however, in this study, PSA tests in general occurred every 14 weeks for patients with nmCRPC and every 12 weeks for patients with mCSPC. Fourth, clinical data present in electronic medical records are subject to inherent limitations related to coding inaccuracies, underreporting of diagnoses and the inability to confirm that the filled prescriptions were taken as prescribed. Fifth, the clinical data used in this study did not capture diagnoses, medical services or prescription fills that may have been obtained outside of the community-based urology practices included in PPS Analytics. Lastly, the algorithm to identify nonmetastatic and castration-sensitive patients relied on the absence of information for metastasis and castration resistance, respectively, and thus patients with missing information on metastasis and castration resistance may have been misclassified. To mitigate this source of bias, information from physician notes abstracted by PPS was used to supplement the study algorithm, which resulted in a more conservative approach to identifying castration status.

Conclusion

This real-world study of patients with nmCRPC or mCSPC demonstrated that early reduction of PSA occurs in the majority of patients treated in the real-world setting and to a similar extent as was observed in phase III trials of apalutamide (SPARTAN and TITAN). Moreover, this study increases knowledge about the use of apalutamide in Black men and demonstrates that apalutamide appears to produce early PSA responses in Black patients with nmCRPC or mCSPC that are similar to those attained in non-Black patients.