Abstract
Aim:
This study examined the clinical outcomes of patients in the US with low/intermediate-risk localized prostate cancer (LIR-LPC) and high-risk localized prostate cancer (HR-LPC) who received radical prostatectomy (RP) as initial treatment.
Materials & methods:
This is a retrospective analysis of the SEER-Medicare database. Patients newly diagnosed with LPC at age of ≥65 years during 2012–2019 who underwent RP as initial definitive treatment and had continuous Medicare Fee-For-Service for ≥12 months prior to RP were included. Eligible patients were stratified into LIR-LPC and HR-LPC cohorts. Overall survival, metastatic free survival and time to advanced prostate cancer treatment (TTAT) were described and compared using the Kaplan–Meier method and Cox proportional-hazards model.
Results:
The LIR-LPC cohort (n = 4120) and the HR-LPC cohort (n = 5359) had comparable socio-demographic characteristics, with a mean age of approximately 70 years. Survival analysis showed that HR-LPC was associated with significantly shorter overall survival, metastatic free survival and TTAT than LIR-LPC (log rank p < 0.001). After adjusting for comprehensive socio-demographic and baseline clinical characteristics, patients with HR-LPC had an approximately 70% increased risk for all-cause death (hazard ratio [HR]: 1.72; confidence interval [CI]:1.39–2.12), 2.5-fold increased risk for metastasis or death (HR: 2.57; CI: 2.14–3.09), and ninefold increased risk for initiating advanced treatments (HR: 9.06; CI: 6.22–13.18) compared with patients with LIR-LPC.
Conclusion:
In patients with LPC who received RP as initial definitive treatment, high risk is strongly associated with suboptimal clinical outcomes. Novel therapeutic approaches are needed to enhance the management and improve the outcomes for this patient population.
Keywords: clinical outcomes, HR-LPC, LIR-LPC, prostate cancer, radical prostatectomy, real-world evidence
Plain language summary: Real-world clinical outcomes after radical prostatectomy for localized prostate cancer
What is this article about?
Prostate cancer is the second most common cancer among men in the US. Most new cases are diagnosed at an early stage, when the cancer is still limited to the prostate, known as localized prostate cancer (LPC). Radical prostatectomy (RP) is one of the most often used initial treatments. However, the clinical outcomes are not expected to be the same depending on the risk classifications. This study examined long-term clinical outcomes among more than 9000 Medicare beneficiaries who were newly diagnosed with LPC between 2012 and 2019 and received RP as initial definitive treatment. Patients were divided into two cohorts of low/intermediate-risk (LIR-LPC) and high-risk localized prostate cancer (HR-LPC), respectively. Their clinical outcomes were compared with regards to overall survival, metastasis-free survival and time to advanced prostate cancer treatment (TTAT).
What were the results?
The study found that patients with HR-LPC had significantly worse outcomes than patients with LIR-LPC even after adjusting for comprehensive socio-demographic and clinical characteristics. High risk is associated with a 70% increased risk for all-cause death, 2.5-fold increased risk for metastasis or death and ninefold increased risk for initiating advanced treatments.
What do the results mean?
These findings suggest that current treatment strategies are inadequate for patients with HR-LPC. New treatments and enhanced clinical management are needed to improve care for HR patients.
Background
Prostate cancer (PC) is the second most common cancer among American men with an annual increase in incidence rate by 3% since 2014 [1,2]. The increase in PC cases has led to an increased disease burden and healthcare costs in recent years. The majority of newly diagnosed cases are localized prostate cancer (LPC) [3], and the primary treatment options include radical prostatectomy (RP), often used in conjunction with androgen deprivation therapy (ADT), external beam radiation therapy, focal therapy and active surveillance [4,5].
Clinical guidelines classify LPC into three risk levels as low, intermediate and high. The risk classification is important in determining initial treatment options and predicting clinical outcomes [4–7]. Although high risk is known to be associated with worse clinical outcomes, there are limited real-world studies quantifying the elevated risks in patients with LPC who received RP as their initial definite treatment.
This study investigated the clinical outcomes of elderly patients in the US with low/intermediate risk LPC (LIR-LPC) and high risk LPC (HR-LPC) after RP was received as initial definitive treatment. Overall survival (OS), metastasis-free survival (MFS) and time to initiation of advanced prostate cancer treatment (TTAT) were compared between the two cohorts. This study also described these outcomes by race and ethnicity, considering the significant racial and ethnic disparities documented in prostate cancer incidence and survival [8–10].
Materials & methods
Data source
This study used the Surveillance, Epidemiology, and End Results (SEER)-Medicare database, which was created by linking the SEER data of cancer patients eligible for Medicare with their claims for covered healthcare services under Medicare. The SEER program of the National Cancer Institute (NCI) is an authoritative source of information on cancer incidence and survival in the US from 18 population-based cancer registries covering approximately 26% of the US population [11]. The registries collect clinical and demographic information as well as data on primary tumor site, tumor morphology, stage of cancer at time of diagnosis and follow-up for vital status [11]. Use of the SEER-Medicare data for this study complied with patient requirements of the Health Insurance Portability and Accountability Act (HIPAA).
Study design
This retrospective study included Medicare beneficiaries newly diagnosed with LPC and received RP as their initial definitive treatment (Figure 1). The date of the initial LPC diagnosis was identified from the SEER CANCER data [11]. The recipient of RP was identified from the Medicare claims. Patients with LPC were stratified into two cohorts according to the National Comprehensive Cancer Network (NCCN) criteria based on tumor-node-metastasis (TNM) staging, Gleason score and prostate-specific antigen (PSA) level at diagnosis (Supplementary Table 1): LIR-LPC and HR-LPC. Patients were followed from the first Medicare claims of RP (index date) until the earliest event of death, disenrollment from Medicare Fee-For-Service (FFS) Part A/B or Part D, or the end of the study period (31 December 2020).
Figure 1. . Study design.
*Where RP treatment was initiated within 6 months after LPC diagnosis as the patient’s initial definitive therapy.
FFS: Fee-For-Service; LPC: Localized prostate cancer; RP: Radical prostatectomy.
Study population & sample selection
The study patient sample selection is summarized in Figure 2. Patients were included in this analysis if they met the following inclusion criteria: Medicare beneficiaries newly diagnosed (at age 65 or older) with LPC between 2012 and 2019, who received RP as initial definitive treatment within 6 months of initial diagnosis and had continuously coverage of Medicare FFS Parts A, B, and prescription drug coverage (Part D) for at least 12 months prior to the index date. Exclusion criteria included a diagnosis of other primary cancers as documented in the SEER data, and indeterminate NCCN risk stratification based on TNM staging, Gleason score, and PSA level.
Figure 2. . Patient selection criteria.
Percentages were calculated using the sample size from the preceding step as the denominator.
FFS: Fee-For-Service; HR-LPC: High-risk localized prostate cancer; LIR-LPC: Low/intermediate-risk localized prostate cancer; PC: Prostate cancer; RP: Radical prostatectomy.
Patient socio-demographic & baseline clinical characteristics
Patient socio-demographic characteristics were obtained from the SEER CANCER data, including age, race/ethnicity, marital status, urban/rural residence, geographic location, household income and education level at the census tract level. TNM staging, PSA level and Gleason score were also obtained from the SEER CANCER data to derive LPC risk classification. Medicare claims were used to estimate the NCI comorbidity score, ADT use, healthcare resource utilization (HCRU), medical and pharmacy spending during the 12-month baseline period before the index date.
Clinical outcomes
The clinical outcomes assessed in the current study include OS, MFS and TTAT. Metastasis events were defined as secondary cancer using International Classification of Diseases (ICD) diagnosis codes (ICD 9: 196x–198x; ICD10: C77x–C80x) with at least one imaging procedure code within 60 days prior. Advanced PC treatments were identified from both medical and pharmacy claims, including chemotherapy, androgen receptor signaling inhibitors, radiopharmaceuticals, PARP inhibitors and immunotherapy. All outcomes were also examined across four race/ethnicity subgroups defined as non-Hispanic (NH) White, NH Black, NH Asian and Hispanic.
Statistical analysis
Socio-demographics and baseline characteristics for the LIR-LPC and HR-LPC cohorts were summarized using descriptive statistics, including the number and proportion of patients (%) for categorical variables as well as mean, standard deviation (SD), and/or median and interquartile range (IQR) for continuous variables. OS, MFS and TTAT were initially compared between the LIR-LPC and HR-LPC cohorts using the Kaplan–Meier (KM) method. Hazard ratios (HRs) were then obtained using Cox proportional hazard (PH) regression models, adjusting for comprehensive socio-demographic and baseline clinical characteristics, including age, race/ethnicity, marital status, urban/rural residence, region, year of initial diagnosis, household income and education level at the census-tract level, NCI comorbidity score, ADT use prior to the index date and healthcare resource utilization and costs during the 1-year baseline period.
Results
Baseline patient characteristics & treatment patterns
Among the 9479 patients who underwent RP within 6 months of LPC diagnosis, 43.5% (n = 4120) were classified as LIR-LPC and 56.5% (n = 5359) as HR-LPC (Table 1). The socio-demographic characteristics for both cohorts were comparable, with a mean age at the index date of 69.6 (SD: 3.3) years for LIR-LPC and 70.4 (SD: 3.7) years for HR-LPC. Additionally, both cohorts had nearly 88% of patients being White, 93% patients being non-Hispanic, 75% patients being married and 85% of patients resided in urban areas.
Table 1. . Demographic and clinical characteristics of patients with low/intermediate-risk localized prostate cancer and high-risk localized prostate cancer.
| LIR-LPC (n = 4120) | HR-LPC (n = 5359) | |
|---|---|---|
| Age at diagnosis, mean (SD) | 69.6 (3.3) | 70.4 (3.7) |
| Age categories (years) | ||
| 65–69 | 2305 (56.0%) | 2550 (47.6%) |
| 70–74 | 1460 (35.4%) | 2117 (39.5%) |
| 75–79 | 314 (7.6%) | 593 (11.1%) |
| 80+ | 41 (1.0%) | 99 (1.9%) |
| Race | ||
| White | 3633 (88.2%) | 4700 (87.7%) |
| Black | 243 (5.9%) | 282 (5.3%) |
| Asian/Pacific Islander/American–Indian/Alaskan Native | 206 (5.0%) | 333 (6.2%) |
| Unknown | 38 (0.9%) | 44 (0.8%) |
| Ethnicity | ||
| Non-Hispanic | 3853 (93.5%) | 4995 (93.2%) |
| Hispanic | 267 (6.5%) | 364 (6.8%) |
| Geographic location | ||
| Northeast | 660 (16.0%) | 811 (15.1%) |
| North Central | 591 (14.3%) | 666 (12.4%) |
| South | 932 (22.6%) | 1058 (19.7%) |
| West | 1937 (47.0%) | 2824 (52.7%) |
| T stage | ||
| T0/T1/Tx/unknown | 161 (3.9%) | 91 (1.70%) |
| T2 | 3959 (96.1%) | 1408 (26.3%) |
| T3 | 0 (0%) | 3788 (70.7%) |
| T4 | 0 (0%) | 72 (1.34%) |
| N stage | ||
| N0 | 4120 (100%) | 4814 (89.8%) |
| N1 | 0 (0%) | 545 (10.2%) |
| PSA level at diagnosis | ||
| <10 ng/ml | 3543 (86.0%) | 3369 (62.9%) |
| 10–20 ng/ml | 577 (14%) | 1074 (20.0%) |
| >20 ng/ml | 0 (0%) | 595 (11.1%) |
| Unknown | 0 (0%) | 321 (5.99%) |
| Gleason score at diagnosis | ||
| 2–6 | 596 (14.5%) | 56 (1.0%) |
| 7 | 3524 (85.5%) | 2023 (37.8%) |
| 8–10 | 0 (0%) | 3266 (60.9%) |
| Missing | 0 (0%) | 14 (0.3%) |
| NCI comorbidity score†, mean (SD) | 1.1 (1.4) | 1.3 (1.5) |
| Selected individual comorbidities‡ | ||
| Hypertension | 2930 (71.1%) | 3907 (72.9%) |
| Diabetes | 1019 (24.7%) | 1442 (26.9%) |
| Chronic Obstructive Pulmonary Disease | 627 (15.2%) | 904 (16.9%) |
| Peripheral Vascular Disease | 518 (12.6%) | 823 (15.4%) |
| Obesity | 580 (14.1%) | 850 (15.9%) |
| Liver Disease | 362 (8.8%) | 636 (11.9%) |
| Baseline ADT use | 71 (1.7%) | 276 (5.2%) |
| Baseline emergency room visit | ||
| Yes | 828 (20.1%) | 1071 (20.0%) |
| No | 3292 (79.9%) | 4288 (80.0%) |
| Baseline hospital admission | ||
| Yes | 338 (8.2%) | 488 (9.1%) |
| No | 3782 (91.8%) | 4871 (90.9%) |
| Baseline physician office visit | ||
| Mean (SD) | 11.6 (6) | 11.8 (6.4) |
| Median (IQR) | 10 (7, 15) | 11 (7, 15) |
| Baseline outpatient encounter | ||
| Mean (SD) | 5.4 (5.4) | 6.1 (5.5) |
| Median (IQR) | 4 (2, 7) | 5 (2, 8) |
| Baseline medical spending§ | ||
| Mean (SD) | $8566 ($11,489) | $9338 ($12,817) |
| Median (IQR) | $5601 ($3,702, $8,863) | $5848 ($4021, $9366) |
| Baseline pharmacy spending§ | ||
| Mean (SD) | $1230 ($4,652) | $1395 ($5078) |
| Median (IQR) | $185 ($0, $1441) | $228 ($0, $1588) |
Included comorbidities that were present at >10%.
All spending amounts were inflated to 2022 values based on Consumer Price Index.
ADT: Androgen deprivation therapy; HR-LPC: High-risk localized prostate cancer; IQR: Interquartile range; LIR-LPC: Low/intermediate-risk localized prostate cancer; NCI: National Cancer Institute; PSA: Prostate-specific antigen; RP: Radical prostatectomy; SD: Standard deviation.
Compared with the LIR-LPC patients, the HR-LPC patients had a slightly higher mean NCI comorbidity score (1.1 vs 1.3; Table 1). The most prevalent comorbidities in both cohorts included hypertension (LIR-LPC: 71.1%, HR-LPC: 72.9%), diabetes (LIR-LPC: 24.7%, HR-LPC: 26.9%), chronic obstructive pulmonary disease (LIR-LPC: 15.2%, HR-LPC: 16.9%), obesity (LIR-LPC: 14.1%, HR-LPC: 15.9%) and peripheral vascular disease (LIR-LPC: 12.6%, HR-LPC: 15.4%). Furthermore, HR-LPC patients had a greater baseline use of ADT compared with LIR-LPC patients (1.7% vs 5.2%).
HR-LPC patients had slightly greater baseline HCRU and healthcare spending compared with LIR-LPC patients. Approximately 20% of both LIR-LPC and HR-LPC patients visited the emergency room, with hospital admission rates of 8.2% and 9.1%, respectively. The median number of physician office visits and outpatient encounters were 10 versus 11 and 5 versus 6 for LIR-LPC and HR-LPC patients, respectively. Median baseline medical spending and pharmacy spending were $5601 versus $5848 and $185 versus $228 for LIR-LPC and HR-LPC patients, respectively.
Clinical outcomes during follow-up period
Although the HR-LPC cohort had a shorter follow-up time than the LIR-LPC cohort (43.8 vs 47.7 months), they experienced more deaths and metastasis events. KM analyses (Figures 3A–C) showed significantly shorter OS, MFS and TTAT in HR-LPC patients compared with the LIR-LPC patients (log rank p < 0.001). After adjusting for comprehensive socio-demographic and baseline clinical characteristics by using Cox PH models, the elevated risks in HR-LPC patients remained substantial. HR-LPC patients had an approximately 2.5-fold increased risk for metastasis or death (HR: 2.57; CI: 2.14–3.09), a 70% increased risk for death (HR: 1.72; CI:1.39–2.12), and ninefold increased risk for initiating advanced treatments (HR: 9.06; CI: 6.22–13.18) compared with LIR-LPC patients (Table 2).
Figure 3. . Kaplan–Meier analyses.
(A) Overall survival in patients with LIR-LPC and HR-LPC. (B) Metastasis-free survival in patients with LIR-LPC and HR-LPC. (C) Time to advanced prostate cancer treatment in patients with LIR-LPC and HR-LPC.
HR-LPC: High-risk localized prostate cancer; LIR-LPC: Low/intermediate-risk localized prostate cancer.
Table 2. . Cox proportional hazard analysis of clinical outcomes in patients with low/intermediate-risk localized prostate cancer and high-risk localized prostate cancer.
| Outcomes | LIR-LPC patients with events (n = 4120), n (%) | HR-LPC patients with events (n = 5358), n (%) | Adjusted hazard ratio‡ (95% CI) HR-LPC vs LIR-LPC patients |
p-value |
|---|---|---|---|---|
| Metastasis or death | 154 (3.7%) | 489 (9.1%) | 2.57 (2.14, 3.09) | <0.0001 |
| Death | 131 (3.2%) | 289 (5.4%) | 1.72 (1.39, 2.12) | <0.0001 |
| Advanced prostate cancer treatment† | 30 (0.7%) | 339 (6.3%) | 9.06 (6.22, 13.18) | <0.0001 |
Initiation of the earliest treatments such as chemotherapy, PARP inhibitors, radiopharmaceuticals, related immunotherapies or advanced androgen signaling inhibitors.
After adjusting for age, race/ethnicity, marital status, urban/rural residence, region, year of initial diagnosis, household income and education level at census tract level, National Cancer Institute comorbidity score, ADT use prior to index date, HRU and cost during the 1-year baseline period.
HR-LPC: High-risk localized prostate cancer; LIR-LPC: Low/intermediate-risk localized prostate cancer.
Clinical outcomes stratified by race & ethnicity
Clinical outcomes and survival probabilities over a 5-year period were stratified by racial/ethnic groups and summarized in Table 3. The proportion of patients who died or had metastatic PC ranged from 4% to 7% for LIR-LPC patients and 8% to 14% for HR-LPC patients across race and ethnic groups. The 5-year MFS analysis indicated a lower probability of survival for patients with HR-LPC relative to LIR-LPC across all racial and ethnic groups, with the largest difference for NH Black patients (87% vs 94%). NH Black patients also exhibited the highest all-cause mortality rate (11.2% among the HR-LPC cohort).
Table 3. . Clinical outcomes and survival probability for 5 years in patients with low/intermediate-risk localized prostate cancer and high-risk localized prostate cancer stratified by race and ethnicity.
| Clinical outcomes | Non-Hispanic White (n = 7739) | Non-Hispanic Black (n = 517) | Non-Hispanic Asian (n = 514) | Hispanic (n = 631) | ||||
|---|---|---|---|---|---|---|---|---|
| LIR-LPC (n = 3382) | HR-LPC (n = 4357) | LIR-LPC (n = 239) | HR-LPC (n = 278) | LIR-LPC (n = 198) | HR-LPC (n = 316) | LIR-LPC (n = 267) | HR-LPC (n = 364) | |
| Metastasis or death, n (%) | 120 (3.6%) | 390 (9.0%) | 16 (6.7%) | 39 (14.0%) | † | 24 (7.6%) | 14 (5.2%) | 30 (8.2%) |
| Death, n (%) | 102 (3.0%) | 222 (5.1%) | 14 (5.9%) | 31 (11.2%) | † | 13 (4.1%) | 13 (4.9%) | 19 (5.2%) |
| Use of advanced PC treatment, n (%) | 24 (0.7%) | 286 (6.6%) | † | 13 (4.7%) | † | 18 (5.7%) | † | 19 (5.2%) |
| 5-year survival probability (95% CI)‡ | ||||||||
| Metastasis free survival | 96.0% (95.0%, 96.8%) | 88.4% (87.1%, 89.6%) | 89.7% (81.7%, 94.3%) | 77.7% (69.3%, 84.1%) | 96.7% (88.4%, 99.1%) | 91.2% (86.2%, 94.4%) | 91.5% (84.5%, 95.4%) | 89.8% (84.3%, 93.4%) |
| Overall survival | 96.9% (96.1%, 97.5%) | 94.1% (93.1%, 95%) | 93.9% (88.7%, 96.8%) | 86.7% (80.8%, 90.9%) | 98.5% (89.9%, 99.8%) | 96.3% (92.9%, 98.1%) | 94.4% (89.3%, 97.1%) | 94.6% (90.5%, 96.9%) |
| No advanced PC treatment | 99.2% (98.7%, 99.5%) | 91.1% (89.9%, 92.2%) | 97.4% (90.5%, 99.3%) | 92.3% (86.3%, 95.7%) | 0% (–, –) | 91.2% (85.5%, 94.7%) | 98.5% (95.2%, 99.5%) | 93.3% (89.2%, 95.9%) |
Patients with unknown race/ethnicity were excluded from this analysis.
Counts less than 11 were suppressed in compliance with CMS cell suppression policies for the SEER-Medicare data.
The survival probabilities were based on KM methods.
CI: Confidence interval; HR-LPC: High-risk localized prostate cancer; KM: Kaplan–Meier; LIR-LPC: Low/intermediate-risk localized prostate cancer; PC: Prostate cancer.
Discussion
In this real-world study of Medicare beneficiaries diagnosed with LPC who received RP as their initial definitive treatment, patients with HR-LPC experienced significantly worse clinical outcomes than patients with LIR-LPC. The KM analyses showed that HR-LPC patients were much more likely to progress to metastasis and/or death and were much more likely to receive advanced therapy than LIR-LPC patients. Furthermore, the Cox PH models indicated that the elevated risks associated with HR-LPC persisted after adjusting for comprehensive socio-demographic and clinical characteristics. In the subgroup analysis, HR-LPC patients consistently appeared to have worse post-RP clinical outcomes than LIR-LPC patients across all the race/ethnic subgroups.
RP has been demonstrated to improve survival rates and reduce the risk of disease progression in LPC patients relative to deferred treatment [12]. Consistent with our findings, several real-world studies show that the effectiveness of RP is reduced in patients with HR-LPC. In a study by Carver et al. assessing long-term outcomes following RP in 176 patients with high-risk PC, 48% of patients exhibited disease recurrence during the mean follow-up period of 6.4 years [13]. In a study by Ward et al., 5652 patients with LPC who underwent RP at a single institution were included [14]. Patients with locally advanced stage of clinical T3 had worse long-term clinical outcomes than those with clinical T2 disease. The 5-year and 10-year death rates were 10% versus 5% and 25% versus 16%, respectively. The biochemical recurrence rates were 42% versus 26% and 57% versus 39%, respectively. Our study provides additional insights by quantifying the elevated risks in real-world clinical outcomes among Medicare beneficiaries. Despite receiving the same initial definite treatment of RP, patients with HR-LPC experienced a substantially worse clinical outcomes compared with patients with LIR-LPC, with a 2.5-fold increased risk for metastasis or death and a ninefold increased risk for initiating advanced treatments. Another key observation of our study is the low utilization of advanced treatment. During a mean follow-up time of nearly 5 years, 6.3% of patients with HR-LPC received advanced treatment after RP. This rate is relatively low considering that more than 30% of high-risk patients are expected to receive post-RP salvage therapies [15–17]. Potential contributing factors to the under use of advanced treatment include evolving clinical evidence and guidelines, a preference for observation with early salvage to avoid overtreatment, variations in physician referral patterns, as well as patient- and access-related factors [15,18]. These findings together suggest room to improve the current management for patients with HR-LPC and highlight the need for novel treatment approaches.
Strengths
One highlight of the current study is the inclusion of a large sample size of more than 9000 patients and a long follow-up time of approximately 7 years on average. The target population of Medicare beneficiaries is relevant for PC research given that the median PC onset age is 67 years old in the US [1]. The SEER-Medicare linkage is another highlight of the current study, as it is a population-based, high-quality database for real-world cancer research. SEER provides validated tumor-level data, which improves case ascertainment and clinical risk classification. The included vital status makes SEER one of the most often used databases for cancer survival studies. The availability of race/ethnicity and census track level family income and education data help mitigate the potential confounding due to socio-economic disparities. Medicare FFS and Part D data record comprehensive inpatient and outpatient encounters, procedures, and pharmaceutical treatments, enabling analyses of broad comorbid conditions, treatment history and outcomes. Lastly, concerns over confounding are mitigated in our study by using Cox PH models that account for comprehensive socio-demographic and clinical characteristics.
Limitations
The findings from the current study should be interpreted within the context of several limitations. First, there are inherent biases associated with claims data, such as miscoding and misclassification errors, which may have led to misclassifications. Particularly, it may not be accurate to determine metastatic progression by using a Medicare claims-based algorithm consisting of diagnosis and procedure codes. Second, we cannot exclude the possibility that the observed association between LPC risk classification and clinical outcomes were due to any unmeasured confounding factors, such as patients' lifestyle and genetic testing information, which may impact the clinical outcomes. Nevertheless, we do not expect any unmeasured confounding factors to be strong enough to change the overall conclusion given the substantial elevated risks found in the HR-LPC cohort after accounting for comprehensive socio-demographics and clinical characteristics. Third, the current study did not explore the pattern and impact of post-RP adjuvant or salvage therapies of either radiation or ADT. It is unclear how post-RP clinical management contributed to the differential outcomes associated with risk classifications. However, the main purpose of the study is to illustrate the independent association between LPC risk classification at diagnosis and post-RP clinical outcomes. Although post-RP clinical management is important, it falls outside of the scope of this study. Lastly, the current study population is restricted to US Medicare FFS beneficiaries. The findings may not be generalizable to patients with LPC in other countries, or US patients primarily covered by other types of health insurance, such as Medicare Advantage, Commercial or Medicaid.
Conclusion
Despite receiving RP as initial definitive treatment, patients with HR-LPC were substantially more likely to experience disease progression and death than those with LIR-LPC. The elevated risks were independent of sociodemographic factors and pre-existing comorbidities, and were consistent across race/ethnicity groups. Together these findings underscore the need for novel therapeutic approaches and enhanced clinical management to improve long-term clinical outcomes in patients with HR-LPC.
Summary points
This study used Surveillance, Epidemiology, and End Results (SEER)-Medicare data, a population-based high-quality database for real-world cancer research among Medicare patients in the US.
The study included patients aged 65 and older, diagnosed between 2012 and 2019, with continuous Medicare coverage for a year before surgery.
All selected patients were stratified into low/intermediate-risk localized prostate cancer (LPC) and high-risk LPC cohorts and compared regarding overall survival, metastasis-free survival and time to advanced prostate cancer treatment (TTAT) using the Kaplan–Meier method and Cox proportional hazard models.
Among the 9479 patients who underwent radical prostatectomy within 6 months of LPC diagnosis, 43.5% (n = 4120) were classified as low/intermediate risk LPC and 56.5% (n = 5359) as high-risk LPC, with similar baseline medical and pharmacy spending.
Patients with high-risk LPC exhibited a statistically significant 70% increased risk for all-cause death, a 2.5-fold increased risk for metastasis or death, and a ninefold increased risk for initiating advanced treatments.
These trends were also reflected across race/ethnicity subgroups, with non-Hispanic Black patients with high-risk LPC exhibited the highest mortality rate among all race/ethnicity subgroups.
Supplementary Material
Acknowledgments
This study used the linked SEER-Medicare database. The interpretation and reporting of these data are the sole responsibility of the authors. The authors acknowledge the efforts of the National Cancer Institute; Information Management Services (IMS), Inc.; and the Surveillance, Epidemiology, and End Results (SEER) Program tumor registries in the creation of the SEER-Medicare database. The collection of cancer incidence data used in this study was supported by the California Department of Public Health pursuant to California Health and Safety Code Section 103885; Centers for Disease Control and Prevention’s (CDC) National Program of Cancer Registries, under cooperative agreement 1NU58DP007156; the National Cancer Institute's Surveillance, Epidemiology and End Results Program under contract HHSN261201800032I awarded to the University of California, San Francisco, contract HHSN261201800015I awarded to the University of Southern California, and contract HHSN261201800009I awarded to the Public Health Institute. The ideas and opinions expressed herein are those of the author(s) and do not necessarily reflect the opinions of the State of California, Department of Public Health, the National Cancer Institute, and the CDC or their Contractors and Subcontractors. Programming support was provided by Dengzhi Wang.
Footnotes
Supplementary data
To view the supplementary data that accompany this paper please visit the journal website at: https://becarispublishing.com/doi/epdf/10.57264/cer-2025-0004
Financial disclosure
This study was funded by Johnson & Johnson (PA, USA).
Competing interests disclosure
L Karsh is an employee of The Urology Center of Colorado and has received consulting fees from Johnson & Johnson. S Du and J He are employees of Johnson & Johnson and stockholders of Johnson & Johnson. N Shore is an employee of Carolina Urologic Research Center and has received consulting fees from Johnson & Johnson. The authors have no other competing interests or relevant affiliations with any organization or entity with the subject matter or materials discussed in the manuscript apart from those disclosed.
Writing disclosure
Medical writing and editorial support were provided by Y Wang and H Tian of Cobbs Creek Healthcare, LLC and funded by Johnson & Johnson.
Ethical conduct of research
Use of the SEER-Medicare data for this study complies with patient requirements of the Health Insurance Portability and Accountability Act (HIPAA).
References
Papers of special note have been highlighted as: • of interest
- 1.Key Statistics for Prostate Cancer | Prostate Cancer Facts. American Cancer Society. (Accessed: 29 March 2024). https://www.cancer.org/cancer/types/prostate-cancer/about/key-statistics.html [Google Scholar]
- 2.Siegel RL, Miller KD, Wagle NS, Jemal A. Cancer statistics, 2023. CA Cancer J. Clin. 73(1), 17–48 (2023). [DOI] [PubMed] [Google Scholar]; • By highlighting the continued decline in cancer mortality rate this article continues to explain how targeted treatment and immunotherapy have enhanced patient survival rate.
- 3.Murthy V, Aggarwal R, Koo PJ. The emerging role of next-generation imaging in prostate cancer. Curr. Oncol. Rep. 24(1), 33–42 (2022). [DOI] [PubMed] [Google Scholar]; • Understanding recent advances in molecular imaging and therapeutics and their expanding role in imaging and treating prostate cancer.
- 4.Prostate Cancer, Version 4.2023, NCCN Clinical Practice Guidelines in Oncology. (Accessed: 20 May 2024). https://pubmed.ncbi.nlm.nih.gov/37856213/ [DOI] [PubMed]; • Provides framework for basing decisions regarding the workup of patients with prostate cancer, risk stratification and management of localized disease, post treatment monitoring and treatment of recurrence and advanced disease.
- 5.Eastham JA, Auffenberg GB, Barocas DA et al. Clinically localized prostate cancer: AUA/ASTRO Guideline, Part I: Introduction, risk assessment, staging, and risk-based management. J. Urol. 208(1), 10–18 (2022). [DOI] [PubMed] [Google Scholar]
- 6.Eastham JA, Auffenberg GB, Barocas D et al. Clinically localized prostate cancer: AUA/ASTRO Guideline, Part II: Principles of active surveillance, principles of surgery, and follow-up. J. Urol. 208(1), 19–25 (2022). [DOI] [PubMed] [Google Scholar]
- 7.Eastham JA, Auffenberg GB, Barocas DA et al. Clinically localized prostate cancer: AUA/ASTRO Guideline. Part III: Principles of radiation and future directions. J. Urol. 208(1), 26–33 (2022). [DOI] [PubMed] [Google Scholar]
- 8.Siegel DA, O'Neil ME, Richards TB, Dowling NF, Weir HK. Prostate cancer incidence and survival, by stage and race/ethnicity – United States, 2001–2017. MMWR Morb. Mortal. Wkly Rep. 69(41), 1473–1480 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]; • Understanding incidence and survival differences by stage, race/ethnicity and age can guide public health planning related to screening, treatment and survivor care. Future research into differences by stage, race/ethnicity and age could inform interventions aimed at improving disparities in outcomes.
- 9.Yamoah K, Lee KM, Awasthi S et al. Racial and ethnic disparities in prostate cancer outcomes in the Veterans Affairs Health Care System. JAMA Netw. Open 5(1), e2144027 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]; • This cohort analysis found significant disparities in the incidence of localized and metastatic PCa between African American veterans and White veterans. This increased incidence is a major factor associated with the residual disparity in PCa metastasis observed in African American veterans compared with White veterans despite their nearly equal response to treatment.
- 10.Moses KA, Orom H, Brasel A, Gaddy J, Underwood W 3rd. Racial/ethnic disparity in treatment for prostate cancer: does cancer severity matter? Urology 99, 76–83 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]; • Shows a significant disparity in the receipt of treatment for PCa among AA and Hispanic men compared to White men. The variations in receipt of treatment reveal an area of opportunity to develop risk-stratified approaches to treatment regardless of ethnic identity, which may address the poorer PCa-related outcomes in these populations.
- 11.Overview of the SEER Program. National Cancer Institute. (Accessed: 29 March 2024). https://seer.cancer.gov/about/overview.html [Google Scholar]
- 12.Vernooij RW, Lancee M, Cleves A et al. Radical prostatectomy versus deferred treatment for localised prostate cancer. Cochrane Database Syst. Rev. 6(6), CD006590 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Carver BS, Bianco FJ Jr, Scardino PT, Eastham JA. Long-term outcome following radical prostatectomy in men with clinical stage T3 prostate cancer. J. Urol. 176(2), 564–568 (2006). [DOI] [PubMed] [Google Scholar]
- 14.Ward JF, Slezak JM, Blute ML, Bergstralh EJ, Zincke H. Radical prostatectomy for clinically advanced (cT3) prostate cancer since the advent of prostate-specific antigen testing: 15-year outcome. BJU Int. 95(6), 751–756 (2005). [DOI] [PubMed] [Google Scholar]
- 15.Hochberg AR, Ho AH, Thompson RAM et al. Real-world management of high-risk prostate cancer post-radical prostatectomy: insights from a regional quality collaborative. Cancers (Basel) 17(10), 1600 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Kalbasi A, Swisher-McClure S, Mitra N et al. Low rates of adjuvant radiation in patients with nonmetastatic prostate cancer with high-risk pathologic features. Cancer 120(19), 3089–3096 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Morgan TM, Hawken SR, Ghani KR et al. Variation in the use of postoperative radiotherapy among high-risk patients following radical prostatectomy. Prostate Cancer Prostatic Dis. 19(2), 216–221 (2016). [DOI] [PubMed] [Google Scholar]
- 18.Parker CC, Clarke NW, Cook AD et al. Timing of radiotherapy after radical prostatectomy (RADICALS-RT): a randomised, controlled Phase III trial. Lancet 396(10260), 1413–1421 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.