PSA Outcomes and Clinical Surveillance Among Patients With Nonmetastatic Castration-Resistant Prostate Cancer Treated With a Next-Generation Androgen Receptor Inhibitor in Urology Practices With or Without In-Office Dispensing

J Clin Pathways. 2024;10(3):33-41. doi:10.25270/jcp.2024.05.01

With the increasing use of oral therapies in urology and oncology settings, in-office dispensing (IOD), a practice that allows clinicians to provide medications directly to patients through an integrated specialty pharmacy rather than outside standalone pharmacies, has become more common.¹ By enabling dispensing at the point of care, patients can receive medications more readily alongside any neces­sary therapy-related education.¹ Importantly, IOD can improve patient care and coordination of cancer treatment by enabling comprehensive disease management and enhancing continuity of care.^1,2 Prior literature has shown a positive impact of clinician-managed oral anticancer therapy programs, including integrated care with in-office specialty pharmacy services, among a wide range of oncology pa­tient populations.^3-11

Nonmetastatic castration-resistant prostate cancer (nmCRPC) is characterized by rising levels of prostate-specific antigen (PSA) despite castrate testosterone levels with androgen deprivation therapy (ADT) or orchiectomy, without evidence of metastasis.^12,13 For patients with nmCRPC, oral next-gener­ation androgen receptor inhibitors (ARIs), such as apalutamide, darolutamide, and enzalutamide, combined with ADT have been shown to decrease PSA, delay metastasis development, and prolong survival compared to treatment with ADT alone.^14-17

Studies have shown that Black patients diagnosed with pros­tate cancer (PC) have higher rates of PC-related mortality rela­tive to non-Black patients.^18,19 This gap in prognosis can be ex­plained by the disparity in the standards of care being provided to Black patients, as they are less likely than non-Black patients to be prescribed new hormonal therapies.^20,21 However, when given access to the same treatments, Black patients were shown to have similar survival outcomes, further highlighting the im­portance of removing barriers to appropriate treatment.²² IOD has been suggested as a potential option to improve access to next-generation ARIs among Black patients.²³

Though oral ARIs can be administered through IOD, there is limited real-world data on treatment and surveillance pat­terns or clinical outcomes among patients with nmCRPC pre­scribed these medications in practices with and without IOD services. In patients with advanced PC, attainment of a rapid PSA response has been shown to serve as an early indicator of treatment effectiveness and significantly better long-term progression-free and survival outcomes.^24-28 Consequently, this study aimed to describe PSA outcomes and clinical surveillance practices in patients with nmCRPC treated in US community-based urology practices with or without IOD services.

Data source

Clinical electronic medical record (EMR) data was collected by Precision Point Specialty (PPS) Analytics from February 1, 2017, to September 17, 2021, as part of routine clinical care from 95 community-based urology practices in the US. PPS data included patient demographics, clinical information, labo­ratory assessments, prescriptions, and, for a subset of partici­pating practices, IOD information. Data were deidentified and complied with the patient requirements of the Health Insur­ance Portability and Accountability Act of 1996; therefore, no review by an institutional review board was required per Title 45 of the Code of Federal Regulations, Part 46.101(b)(4).

Study design and sample selection

A retrospective, longitudinal analysis of adults with nmCRPC who were initially prescribed apalutamide, darolutamide, or enzalutamide was conducted. Three mutually exclusive cohorts were analyzed. Patients with IOD access who had an IOD fill observed within 14 days following the first prescrip­tion date were assigned to the IOD+ cohort. Patients with IOD access who did not have an IOD fill observed within 14 days following the first prescription date were assigned to the IOD- cohort. Patients with a prescription from clinics without IOD access were assigned to the non-IOD cohort. Within each of the three cohorts, the subset of Black patients was also identified. The baseline period was defined as the 12-month period preceding the first prescription date, and the observation period spanned from the index date (ie, 14 days following the first prescription date) until the earliest among a prescription for a new next-generation antiandro­gen, advanced PC therapy, end of clinical activity (including death), or end of data availability (Figure 1).

Measures and outcomes

Demographics and clinical characteristics were assessed during the 12-month baseline period for patients in the IOD+, IOD-, and non-IOD cohorts. Frequency of PSA testing and PSA response was assessed during the observation period. PSA response was defined as a decline ≥50% (PSA50) or ≥90% (PSA90) from the baseline PSA value measured within 13 weeks prior to and including the first prescription date. PSA progression was defined as a ≥25% increase from baseline PSA and was assessed among those with ≥1 follow-up PSA. Addition­ally, for each type of imaging test (ie, bone scan, computed to­mography [CT] scan, prostate-specific magnetic resonance im­aging [MRI], and next-generation imaging), the proportion of patients who underwent ≥1 postindex imaging test, including the proportion who underwent testing within 6 months and 12 months after the index date, the mean number of postindex imaging tests, and the time from index date to the most recent imaging test, were assessed during the observation period.

Statistical analyses

Categorical variables were reported using counts and propor­tions, and continuous variables were reported using means, medians, and standard deviations. PSA50 and PSA90 response were assessed using Kaplan-Meier (KM) analyses and further evaluated separately for Black patients. Results were reported separately by study cohort and all analyses were descriptive (ie, no statistical significance testing performed).

Baseline characteristics

A total of 3300 patients with nmCRPC who were treated with a next-generation ARI were included in the study. Among them, 615 patients (18.6%; mean age 79.3 years, 20.3% Black) were included in the IOD+ cohort, 2474 patients (75.0%; mean age 78.2 years, 20.2% Black) were included in the IOD-cohort, and 211 patients (6.4%; mean age 79.9 years, 12.3% Black) were included in the non-IOD cohort (Figure 1). Prior use of a first-generation ARI was observed in 16.3% (IOD+), 15.9% (IOD-), and 24.2% (non-IOD) of patients. A greater proportion of patients who received treatment from sites with IOD services used bone antiresorptive therapy (IOD+: 23.9%; IOD-: 21.0%) than those who did not (non-IOD: 11.4%). A smaller proportion of patients with access to IOD services (IOD+: 73.2%; IOD-: 69.8%) had a PSA doubling time of ≤ 10 months relative to those without IOD access (non-IOD 80.7%; Table 1).

PSA outcomes

Overall, 86.2% of patients in the IOD+ cohort, 78.4% in the IOD- cohort, and 79.1% in the non-IOD cohort had a PSA test within 6 months following the index date (Supplementary Table 1). PSA testing was conducted every 6 months on aver­age for 75.9% (IOD+), 68.8% (IOD-), and 71.1% (non-IOD) of patients. In the KM analysis, a PSA50 response was observed by 6 months in 80.0% (IOD+), 63.8% (IOD-), and 62.5% (non- IOD) of patients (Figure 2). By 12 months, a PSA50 response was observed in 83.8% (IOD+), 72.3% (IOD-), and 69.1% (non-IOD) of patients. The median time to a PSA50 response was 2.3 months in the IOD+ cohort, 3.6 months in the IOD-cohort, and 3.0 months in the non-IOD cohort. A similar trend in PSA90 response was also observed across the cohorts (Supplementary Figure 1). PSA progression was observed in 9.7% (IOD+), 16.1% (IOD-), and 17.2% (non-IOD) of patients (Supplementary Figure 2).

Among the subgroup of Black patients with nmCRPC who were treated with next-generation ARIs (IOD+: n = 108; IOD-: n = 413; non-IOD: n = 22), PSA50 response was observed by 6 months in 79.1% (IOD+), 64.9% (IOD-), and 71.9% (non-IOD) of patients. At 12 months, a PSA50 response was observed in 87.3% (IOD+), 76.4% (IOD-), and 78.9% (non-IOD) of patients. The median time to a PSA50 response was 2.5 months (IOD+), 3.6 months (IOD-), and 2.9 months (non-IOD) (data not shown).

Surveillance outcomes

A similar proportion of patients in the IOD+, IOD-, and non- IOD cohorts had ≥1 CT scan, bone scan, MRI scan, or next-generation imaging test (Figure 3). On average, patients in the IOD+ cohort underwent fewer bone scans, CT scans, and next-generation imaging than patients in the IOD- and non-IOD cohorts. The mean times from the index date to the most recent bone scan, CT scan, and next-generation imaging were also de­scriptively longer in the IOD+ cohort relative to the IOD- and non-IOD cohorts (Table 2).

In this retrospective, longitudinal cohort study, EMR data was used to describe the demographic and clinical characteristics of patients with nmCRPC treated with next-generation ARIs in US community-based urology practices with or without IOD services, and to assess their PSA outcomes and clinical surveil­lance patterns. Relative to patients who received treatment from sites with IOD services, higher proportions of patients in the non-IOD cohort had previously used first-generation ARIs and had a PSA50 of ≤ 10 months, while a lower propor­tion used bone antiresorptive therapies. Patients in the IOD+ cohort were more likely to experience a PSA50 response and had a shorter median time to PSA50 response relative to those in the IOD- and non-IOD cohorts. The proportion of pa­tients in the IOD+ cohort with PSA progression was nearly half that of IOD- and non-IOD cohorts. The relative differ­ence in PSA50 response between patients in the IOD+ cohort versus those in the other cohorts was consistent when the analysis was restricted to Black patients. Differences were not­ed in clinical surveillance patterns within IOD and non-IOD clinics whereby patients in the non-IOD cohort tended to undergo more frequent monitoring, particularly using bone, CT, and next-generation imaging. Overall, these results high­light improved clinical outcomes associated with receiving a fill in an IOD setting within 14 days of prescription among patients with nmCRPC treated with next-generation ARIs in community-based urology practices.

The general findings of this study are consistent with pri­or analyses on the benefit of interventional and/or monitor­ing programs in other PC populations. A retrospective cohort study among patients with metastatic castration-resistant PC receiving ARIs in an outpatient clinic that compared patients before and after enrollment in a pharmacist-led oral chemo­therapy-monitoring program found that patients who were en­rolled had a significant increase in the average number of inter­ventions made by the oncology pharmacist per patient as well as improved adherence to laboratory monitoring up to 2 years postprogram initiation compared to patients without pharma­cist intervention.⁸ In a pre vs post study conducted among pa­tients with advanced PC who received pharmacist intervention in an outpatient community pharmacy, PSA levels decreased significantly by 6 and 12 months,11 underscoring the impact of comprehensive disease management strategies on improving markers of disease progression.

Our findings are also consistent with a modest body of lit­erature showing a positive impact of clinician-managed oral anticancer therapy programs, which include IODs, on clini­cal outcomes among a wide range of oncology patient popu­lations.^3-11,29 Based on a systematic review of the literature on patient-centered best practices for the delivery of oral antican­cer and supportive care drugs, interventions, many of them including an integrated in-office oncology pharmacist com­ponent, were found to improve patient outcomes and led to higher rates of prescription initiation, adherence to treatment and laboratory monitoring, and reduced symptom severity fol­lowing the intervention.³ One study assessed the impact of an integrated, closed-loop, pharmacy-led oral chemotherapy man­agement program on patients with chronic myeloid leukemia and found that the program led to a higher major molecular response rate compared with published clinical trials, as well as improved treatment adherence rates that exceeded nationally established thresholds.⁴ Although results from this study are not directly comparable to studies with formal pharmacist-led in­terventions, they are in agreement with findings of improved outcomes based on the principle of patient integration within a structured care network. Though many studies have evalu­ated the impact of clinician-managed oral anticancer therapy programs, they are often limited by small sample sizes, hetero­geneous oncology patient populations, single-center settings, short follow-up time, or the absence of a non-IOD arm. Most studies also focused on the impact of interventional programs on medication adherence, as evidence suggests that nonadherence to oral anticancer therapy can be associated with worse clini­cal outcomes (ie, disease progression and death), including among patients with PC.^30-33 This study fills an important gap in knowledge by characterizing patients receiving care in community-based oncology practices with and without IOD services and describing their PSA outcomes, which serve as a marker of disease progression. Our findings demonstrated that patients without IOD access were nearly twice as likely to fail to achieve a PSA50 response compared to those who had IOD access and received a fill. Furthermore, to discern the potential impact that IOD access may have on reducing health dispari­ties between non-Black and Black patients, the latter of which may receive diagnosis at later disease stages,34 KM analyses of PSA response were replicated among Black patients. Trends observed in the subgroup analysis were consistent with those of the overall analysis, with a greater proportion of Black pa­tients who had IOD access and received a fill achieving PSA50 than those who did not. Given that cancer therapies have be­come more complex, this study suggests that continuity of care and the utilization of IOD services for patients receiving ARIs may be critical to maximizing the benefits of oral anti­cancer therapy, thereby leading to improved clinical outcomes.

To our knowledge, this is the first study to describe clinical surveillance patterns among patients with nmCRPC with or without access to IOD services. Based on a large, representative study population of 95 community-based urology practices in the US, findings show that although a similar proportion of patients in practices with and without IOD services under­went postindex imaging, non-IOD patients had more frequent postindex bone, CT, and next-generation imaging per patient than those who used IOD services. Likewise, time to imaging was shorter for patients in non-IOD practices than those who used IOD services. Additional resource utilization among pa­tients with nmCRPC without IOD access may have financial implications and may not reflect an increased quality of care given the potential for unnecessary or excessive monitoring.³⁵

Limitations

The results of this study are subject to certain limitations. Analyses from this retrospective study were descriptive and no adjustment was made for observable confounders between the cohorts. Therefore, further studies with adjustment for potential confounders are necessary to draw any comparative conclusions. The present study relied on clinical data that may contain inaccuracies or omissions (eg, diagnosis dates, treat­ment start dates) and does not capture any diagnoses, medical services, or services received outside of the included urology sites (ie, from pharmacies, oncologist services, treatment, test­ing, and imaging completed outside of PPS network). Addi­tionally, it is not possible to assess adherence to the index med­ication or determine whether patients in the non-IOD cohort filled their prescription, as medications received elsewhere (eg, local or specialty pharmacies) would not have been captured in the clinical EMR data. Our study also did not capture dispens­ing in the IOD- cohort post-14 days. Although patients who received an initial fill within the IOD may have improved ad­herence to ARIs, this study examined the impact of accessing oral PC medications through an IOD on real-world clinical outcomes. As the database does not contain patient enrollment data and prescription claims for non-IOD patients, an intent-to-treat approach was used whereby patients were followed until the earliest of their last record within the EMR, a treat­ment addition, or data availability. Furthermore, information on geographic regions, socioeconomic disparities, or insurance coverage that may explain differences in access to ARIs were not available in the EMR data. Though overall survival and metastasis-free survival are important clinical outcomes to consider, the follow-up period was not sufficient in duration to observe those outcomes. Instead, PSA outcomes, which represent an important factor in defining nmCRPC prognosis and reflect the more immediate and measurable effectiveness of the prescribed next-generation ARI treatments in this patient population,^36,37 were evaluated. Finally, the database represents the community urology perspective and may not be represen­tative of the entire US population of patients with nmCRPC treated with next-generation ARIs.

Among patients with nmCRPC treated with next-generation ARIs in US community-based urology practices with or with­out access to IOD services, greater proportions of patients using IOD services achieved a PSA50 response by 6 and 12 months of prescription relative to those who did not use or did not have access to IOD services. This response was attained earlier among patients that used IOD services, who had less oc­currence of PSA progression and less frequent PSA and imag­ing surveillance than those who did not have access to IOD ser­vices. The descriptive improvement in PSA was also observed when analyses were restricted to Black patients. Differences were noted in patient characteristics within IOD and non-IOD practices. Additional studies controlling for differences in base­line characteristics and examining longer-term outcomes are needed to understand how IOD services impact quality of on­cology patient care.

This article has supplementary material, which can be accessed here.

Authors: 

Sabree C. Burbage, PharmD¹; Dexter Waters, MSPH¹; Carmine Rossi, PhD²; Frederic Kinkead, MA²; Erik Muser, PharmD¹; Lorie Ellis, PhD¹; Patrick Lefebvre, MA²; Dominic Pilon, MA²

Affiliations: 

¹Janssen Scientific Affairs, LLC, Horsham, PA; ²Analysis Group, Inc, Montréal, Québec, Canada

Address correspondence to: 

Carmine Rossi, PhD

Manager, Analysis Group, Inc

1190 Av. des Canadiens-de-Montréal

Suite 1500

Montreal, QC H3B 0G7, Canada

Phone: 514-871-4233

Fax: 514-394-4461

Email: carmine.rossi@analysisgroup.com

Disclosures: 

S.C.B., D.W., E.M., and L.E. reported being employees of Janssen Scientific Affairs, LLC, and owning Johnson & Johnson stock. C.R., F.K., P.L., and D.P. reported being employees of Analysis Group, Inc.