Recent Advances in the Management of High-Risk Localized Prostate Cancer: Local Therapy, Systemic Therapy, and Biomarkers to Guide Treatment Decisions

Abstract

High-risk prostate cancer accounts for approximately 15% of all prostate cancer diagnoses. Patients with high-risk disease have an increased risk of developing biochemical recurrence, metastases, and death from prostate cancer. As the optimal management of high-risk disease in patients with prostate cancer continues to evolve, the contemporary treatment paradigm is moving toward a multidisciplinary integrated approach of systemic and local therapy for patients with high-risk disease. The strategies for definitive, adjuvant, and salvage local treatment, including radical prostatectomy or radiation therapy, serve as the backboneof therapy for patients with localized disease. Systemic therapy decisions regarding use in combination with surgery, choice of therapy (hormone therapy, chemotherapy), and treatment duration continue to be refined. As more effective hormonal agents populate the treatment landscape for advanced prostate cancer, including abiraterone and next-generation antiandrogens, an opportunity is provided to explore these treatments in patients with localized disease in the hope of improving the long-term outcome for patients. Integration of innovative blood and tissue-based biomarkers to guide therapy selection for patients with high-risk disease is an area of active research. Contemporary studies are using such biomarkers to stratify patients and select therapies. In this review, we summarize contemporary evidence for local treatment strategies, systemic therapy options, and biomarkers in development for the management of high-risk prostate cancer in patients.

INTRODUCTION

Prostate cancer is the second most common malignancy among men worldwide, and most patients in Western countries (greater than 80%) present with localized disease (T1–4N0M0).¹ Approximately 15% of individuals with localized disease are identified as at high risk for disease recurrence.^2,3 Treatment options have historically included surgery, radiation therapy, and/or androgen deprivation therapy (ADT). However, despite treatment, a subset of men with localized disease develop recurrent lethal prostate cancer. Given the potential for cure of patients with localized disease, effective treatment strategies are crucial to improving long-term outcomes for patients at this critical stage. Multimodal treatment strategies that integrate local and systemic treatments have the potential to decrease recurrence rates and prolong survival. Such approaches that integrate neoadjuvant and/or adjuvant systemic therapy, including chemotherapy and/or hormonal therapy, with localized treatment, including surgery and/or radiation therapy, are standard across several solid tumor malignancies. However, consensus regarding the optimal treatment of patients with high-risk localized prostate cancer has not been established. Additionally, biomarkers, beyond clinicopathologic features, have not been routinely used in practice to guide clinical decision-making and are still under development. In this article, we aim to summarize the current evidence for (1) local treatment strategies of patients with high-risk localized disease; (2) use of neoadjuvant and adjuvant systemic therapy, including a discussion on novel therapeutics for disease control; and (3) integration of innovative blood- and tissue-based biomarkers to guide therapy selection for patients with high-risk disease.

DEFINING HIGH-RISK DISEASE

There are multiple definitions used to categorize individuals with high-risk prostate cancer. Pretreatment parameters, including clinical stage, prostate-specific antigen (PSA), and Gleason score, are established predictors of disease recurrence and have historically been used in high-risk disease classifications.⁴ In 1998, using an endpoint of PSA recurrence, D’Amico et al² defined high-risk disease as a clinical T stage of at least cT2c, a Gleason score of at least 8, or a PSA greater than 20 ng/mL. This definition is widely used, given its simplicity and ease of use, and has been adopted by the American Urologic Association,⁵ European Association of Urology,⁶ and United Kingdom National Institute for Health and Clinical Excellence.⁷ The Radiation Therapy Oncology Group (RTOG) developed the first classification system using factors associated with overall survival and prostate cancer–specific survival as opposed to PSA recurrence. The RTOG definition of high-risk disease is (1) PSA of 20 to 100 ng/mL, biopsy Gleason score of at least 7, and any clinical T stage or (2) PSA less than 100 ng/mL, Gleason score 8 to 10, and clinical T stage cT2c.⁸ The National Comprehensive Cancer Network (NCCN) defines high risk as clinical T stage cT3a, Gleason score of at least 8, or PSA of at least 20 ng/mL, and it defines very high risk as T3b or T4 disease.⁹ Although these definitions do not integrate disease extent within the gland, the Cancer of the Prostate Risk Assessment score considers the percent of biopsy cores positive for cancer in risk stratification, along with age, clinical T stage, and Gleason score, to predict prostate cancer–specific mortality independent of treatment.¹⁰ Despite the introduction of this additional classification system, given variability in reliably reproducing estimates of percentage of cores involved with tumor, traditional classification systems are more commonly used for risk stratification.

In addition to tiered classifications systems, nomograms that incorporate several interacting continuous and categoric clinical variables have been established to stratify patients by risk. The Kattan preoperative nomogram uses a multivariable model that combines clinical T stage, Gleason score, and PSA to generate an estimate of the risk of treatment failure after radical prostatectomy on a continuous scale.¹¹ Application of the nomogram requires a multistep paper tool or computer program for use.

Although the development of instruments to aid in risk stratification has been helpful, there is considerable heterogeneity in the outcomes observed within the high-risk groups defined by each stratification tool. Additionally, the stratification tools to date define T stage clinically based on digital rectal examination. Interobserver variability and underand overstaging are challenges with digital rectal examination staging.¹² The American Joint Committee on Cancer staging recognizes the use of MRI in clinical T stage categorization. Incorporation of multiparametric prostate MRI can be useful to evaluate for the presence of extraprostatic disease, seminal vesicle invasion, and pelvic node involvement. However, challenges also arise with results interpretation, given differences in MRI quality (magnet strength, use of coils), reader variability, prostate factors (size of gland, size and grade of tumor), and patient factors (motion artifact, metal artifact, biopsy-related hemorrhage).¹³ Molecular biomarker tests (e.g., Decipher, Oncotype DX Prostate, Prolaris, or ProMark) have further improved risk stratification algorithms. Although previous guidelines panels have recommended consideration of the use of any of these four assays for initial risk stratification in patients (with life expectancy greater than 10 years) with low or favorable-intermediate risk of prostate cancer, the most recent 2020 NCCN guidelines expand the use of biomarkers into more aggressive patient populations, stating that “Men with unfavorable intermediate- and high-risk disease and life expectancy ≥ 10 years may consider use of Decipher and Prolaris tumor-based molecular assays.”

LOCAL TREATMENT OF PATIENTS WITH HIGH-RISK DISEASE

Radical Prostatectomy

Historically, high-risk prostate cancer in men, especially if clinically advanced, was managed with systemic monotherapy using ADT, radiation therapy, or a combination of both. Radical prostatectomy was discouraged because of concerns regarding positive surgical margins, inadequate disease control, and morbidity.¹⁴ However, studies have investigated men with locally advanced prostate cancer and found that excellent long-term survival outcomes are achieved. Carver and colleagues¹⁵ showed cancer-specific survival was 85% at 10 years and 76% at 15 years. Although there is concern for micrometastatic disease in patients at high risk, patients undergoing radical prostatectomy benefit from local control, with rates of local recurrence approximating 10%.¹⁶ The important local control and debulking not only improve the efficacy of sequential therapy with either radiation therapy or ADT aimed at micrometastatic and locoregional disease control but also prevent clinical complications, such as hematuria and obstruction.¹⁷ Furthermore, radical prostatectomy allows for accurate staging that can guide selection of patients who may benefit from adjuvant therapies. There are also data showing that downgrading to a lower Gleason score is not infrequent after radical prostatectomy for high-risk prostate cancer.¹⁸

Role of Pelvic Lymph Node Dissection

The current NCCN prostate cancer guidelines recommend performing extended pelvic lymph node dissection (ePLND) in patients with 2% or greater predicted probability of nodal metastases by nomograms. The ePLND includes lymph node removal of all node-bearing tissue from a predefined anatomic region; however, the decision to perform PLND and the extent have historically been left to the surgeon’s discretion. As a result, most studies looking at the risks and benefits of PLND are retrospective, with associated confounding variables.¹⁷ A systemic review analyzing comparative studies found the data to be mixed regarding oncologic outcomes, and, overall, there was no solid quality evidence indicating that any type of PLND improves biochemical recurrence, distant metastasis, or survival compared with no PLND for high-risk prostate cancer.¹⁹ Furthermore, PLND was associated with a high rate of lymphocele but no statistical significance in urinary continence and erectile function recovery.^20,21 Although the benefits of PLND are not clear, this may be because of a lack of adequately powered randomized controlled trials. The rationale for PLND still holds true in that it provides the most accurate staging to determine node-positive disease, which may impact therapeutic treatment decisions. In addition, PLND may potentially be curative for patients with limited nodal involvement at time of PLND.

Adjuvant Versus Early Salvage Radiotherapy

The optimal timing of postoperative radiation for patients with adverse pathologic findings, such as seminal vesicle invasion, extraprostatic extension, and/or positive surgical margins, has been widely studied.²² Historic trials, including SWOG-8794,^22,23 ARO-96–02,²⁴ EORTC-22911,^25,26 and FinnProstate,²⁷ investigated the role of adjuvant radiotherapy versus observation in patients with high-risk features. Overall, these studies demonstrated improvements in biochemical progression-free survival, but only SWOG 8794 showed improvements in metastasis-free survival and overall survival. Despite these studies, the use of adjuvant radiotherapy in patients at high risk is low, given the concern for overtreatment of patients in whom the cancer is never destined to recur after surgery.

Contemporary prospective studies have compared adjuvant versus early salvage radiation therapy. The RAVES trial, recently presented at the 2019 American Society for Radiation Oncology (ASTRO) meeting, was the first randomized study comparing adjuvant and early salvage radiotherapy. Patients were randomly selected to either adjuvant radiotherapy or an early salvage strategy in which radiotherapy was started at PSA of 0.2 ng/mL or greater. ADT was not used concurrently with radiotherapy. At a follow-up median of 6.1 years, there was no evidence of improved biochemical progression-free survival among patients who received adjuvant radiotherapy. Furthermore, patients in the adjuvant arm were more likely to experience clinically relevant genitourinary toxicity. In addition, the RADICALS-RT trial, which used a trigger for early salvage radiotherapy of PSA greater than 0.1 ng/mL, was recently presented at the 2019 European Society of Medical Oncology meeting, showing that between patients who received adjuvant versus early salvage radiation, there was no difference in biochemical progression-free survival. A meta-analysis of these trials (as well as the GETUG-AFU 17 trial, which has not been independently presented or published), termed the ARTISTIC collaboration, found that, despite some difference in patient population and study design, the results were remarkably similar; there was no significant improvement in biochemical event-free survival for patients receiving adjuvant radiotherapy compared with early salvage radiotherapy. Thus, early salvage radiotherapy is an accepted approach for patients with high-risk disease undergoing initial surgical treatment, although additional follow-up data are warranted to assess long-term outcomes.

Radiation Therapy Paradigms

Conformal techniques, particularly intensity-modulated radiation therapy and image-guided radiation therapy, are the contemporary standard of care for treatment of high-risk localized prostate cancer for patients electing radiation therapy. These conformal techniques allow higher doses to the target while minimizing toxicity to normal tissues, compared with older approaches. The role of moderate hypofractionation continues to evolve, and the optimal regimen for hypofractionation for high-risk prostate cancer has not yet been established. In the Dutch HYPRO trial, 820 patients, including more than 73% with high-risk prostate cancer, were randomly assigned to radiation therapy with conventional fractionation (39 fractions of 2 Gy over 8 weeks) or hypofractionation (19 fractions of 3.4 Gy in 6.5 weeks).²⁸ At 60 months, there was no difference in 5-year relapse-free survival; however, gastrointestinal toxicity was more common in the hypofractionation group.

For men with localized high-risk prostate cancer without clinical pelvic adenopathy, the role of whole-pelvis radiation therapy remains a controversial area. Whole-pelvis radiation therapy can be considered in men with an estimated risk of nodal involvement exceeding 15% on the basis of Partin tables or other tools. Two randomized trials (RTOG 9413 and GETUG-01) investigated the role of whole-pelvis radiation therapy and did not demonstrate a clear benefit of whole-pelvis radiation therapy compared with prostate-only radiation therapy.^29,30 RTOG 0924 (NCT01368588), which has patients with high-risk or locally advanced prostate cancer receiving ADT in conjunction with either prostate-only or whole-pelvis radiation therapy, will provide additional information on the extent of radiation therapy for localized disease.

The use of brachytherapy with external beam radiotherapy has been studied extensively in patients with both intermediate- and high-risk disease. The ASCENDE-RT trial demonstrated that the addition of brachytherapy boost to external beam radiotherapy and ADT in men with intermediate- and high-risk disease was associated with improved biochemical control and comparable overall survival.³¹ However, brachytherapy boost was associated with increased genitourinary toxicity but no difference in gastrointestinal toxicity. Furthermore, among patient-reported outcomes, brachytherapy boost was associated with worse overall health, sexual function, and urinary function.

SYSTEMIC THERAPY FOR PATIENTS WITH HIGH-RISK DISEASE

Systemic Therapy in Combination With Radical Prostatectomy

Neoadjuvant therapy

Several clinical trials have evaluated neoadjuvant therapy before radical prostatectomy. Initial studies were conducted in the early 1990s with the primary intent of improving pathologic surgical outcomes, mainly the rate of positive surgical margins.^32,33 Most studies included luteinizing hormone-releasing hormone agonists or antagonists with or without first-generation antiandrogens. In aggregate, these studies included a small number of patients, most of whom had lower-risk prostate cancer, and did not robustly evaluate pathology responses and long-term outcomes. Gleave and colleagues³⁴ evaluated 8 versus 3 months of leuprolide plus flutamide in patients with T1b-T2 tumors. This study enrolled 547 patients and demonstrated a lower positive surgical margin rate (12% vs. 23% with 8 vs. 3 months) and improved pathologic response rate (9.3% vs. 5.1% with 8 vs. 3 months) with longer therapy; however, disease-free survival and overall survival were not reported.

The introduction of more effective hormonal agents for patients with advanced disease has revived interest in this approach for patients with localized prostate cancer at high risk for recurrence. Abiraterone—a CYP17 inhibitor that decreases testosterone production from the adrenal gland, testicles, and prostate cancer cells—has demonstrated improved survival for patients with metastatic castration-resistant prostate cancer (CRPC)^35,36 and also advanced hormone-sensitive disease.^37,38 Additionally, three next-generation antiandrogens, including enzalutamide (for nonmetastatic³⁹ and metastatic^40,41 CRPC and metastatic hormone-sensitive disease⁴²), apalutamide (for nonmetastatic CRPC⁴³ and metastatic hormone-sensitive disease⁴⁴), and darolutamide (for nonmetastatic CRPC⁴⁵), are currently used for patients with advanced prostate cancer. Unlike first-generation antiandrogens, in addition to blocking androgens from binding to the androgen receptor, these agents prevent nuclear translocation of androgen receptors and binding of androgen receptors to androgen response elements on DNA.⁴⁶

Several contemporary studies have investigated neoadjuvant abiraterone, enzalutamide, and apalutamide in patients with localized prostate cancer undergoing radical prostatectomy (Table 1).^47–50 Although the systemic therapy regimens differed among these studies (leuprolide × 6 months + abiraterone × 3 vs. 6 months [NCT00924469]⁵⁰; enzalutamide vs. enzalutamide + dutasteride + leuprolide 6 months [NCT01547299]⁴⁸; enzalutamide + leuprolide with or without abiraterone [NCT02268175]⁴⁷), eligible patients were required to have unfavorable intermediate- or high-risk prostate cancer. Additionally, the studies integrated central pathology review with predefined criteria for measurement and reporting of pathologic outcomes. Overall, combination therapy resulted in pathologic complete responses in 4% to 10% of patients and minimum residual disease in 17% to 30% of patients. A critical component of neoadjuvant therapy is that pathologic response correlates with long-term outcomes. A meta-analysis of pooled contemporary clinical trials of more intense ADT demonstrated that with a median follow-up of 3.4 years, 3-year biochemical-free survival was 70% and 3-year metastasis-free survival was 98%.⁵¹ All patients who experienced a pathologic response (defined as a pathologic complete response or minimum residual disease no more than 5 mm of residual tumor) were still disease free at the time of last follow-up. In light of the results of these phase II studies, an international, phase III, randomized, placebo-controlled study is underway evaluating leuprolide with apalutamide versus placebo for 6 months before and 6 months after radical prostatectomy (NCT03767244). The coprimary endpoints are the pathologic complete response rate and metastasis-free survival. The results will inform evaluation of pathologic complete response as a surrogate for metastasis-free survival in prostate cancer.⁵²

TABLE 1.

Summary of PSA and Pathologic Outcomes of Contemporary Clinical Trials of Intense Neoadjuvant ADT Before RP

	NeoAbi50		NeoEnza48		NeoAbiEnza47
Clinical Outcomes	12wAA (n = 27)	24wAA (n = 29)	Enza (n = 25)	EDL (n = 23)	ELAP (n = 50)	EL (n = 25)
Median PSA pre-RP visit, ng/mL	0.06	0.04	0.51	0.04	0.03	0.02
≥ ypT3	59%	48%	72%	61%	50%	56%
	n = 16	n = 14	n = 18	n = 14	n = 25	n = 14
Positive nodes	11%	24%	4%	26%	10%	12%
	n = 3	n = 7	n = 1	n = 6	n = 5	n = 3
Positive margins	19%	10%	16%	22%	18%	12%
	n = 5	n = 3	n = 17	n = 13	n = 9	n = 3
pCR (%)	4%	10%	0%	4%	10%	8%
	n = 1	n = 3	n = 0	n = 1	n = 5	n = 2
MRD (largest cross-sectional dimension ≤ 5 mm)	0%	14%	—	—	20%	8%
	n = 0	n = 4			n = 10	n = 2
MRD (largest cross-sectional dimension ≤ 3 mm)	—	—	0%	13%	—	—
			n = 0	n = 3
pCR or MRD (largest cross-sectional dimension ≤ 3 or ≤ 5 mm)	4%	24%	0%	17%	30%	16%
	n = 1	n = 7	n = 0	n = 4	n = 15	n = 4
RCB ≤ 0.25 cm 3	44%	52%	36%	74%	—	—
	n = 12	n = 15	n = 9	n = 17

Abbreviations: PSA, prostate-specific antigen; ADT, androgen deprivation therapy; RP, radical prostatectomy; 12wAA, 12-week abiraterone acetate; 24wAA, 24-week abiraterone acetate; AA, abiraterone acetate; Enza, enzalutamide; leuprolide; EDL, enzalutamide, dutasteride, leuprolide; ELAP, enzalutamide, leuprolide, abiraterone, prednisone; EL, enzalutamide; pCR, pathologic complete response; MRD, minimum residual disease; RCB, residual cancer burden.

Studies have evaluated the impact of neoadjuvant chemotherapy or chemohormonal therapy before radical prostatectomy with heterogeneous results. Docetaxel, a taxane chemotherapy that prevents microtubule depolymerization, has substantial activity in patients with advanced prostate cancer, including those with CRPC⁵³ and metastatic hormone-sensitive disease.^54,55 To date, the most robust data for use of neoadjuvant docetaxel with leuprolide come from the PUNCH trial (NCT00430183).⁵⁶ The trial enrolled 750 patients with clinically localized high-risk prostate cancer (defined as a Kattan preoperative nomogram probability of less than 60% biochemical progression-free survival at 5 years after radical prostatectomy or biopsy Gleason score of 8–10). Patients were randomly assigned 1:1 to docetaxel plus leuprolide versus surgery alone. Although there was no difference in the trial primary endpoint of 3-year biochemical progression-free survival, with 8 years of follow-up, rates of biochemical progression-free survival were greatly improved in patients receiving docetaxel and leuprolide. Additionally, although the study was not powered to detect differences in overall survival, there was a strong signal of improved overall survival in patients receiving systemic therapy (HR, 0.66; 95% CI, 0.42–1.03).⁵⁶

Adjuvant therapy

The role of adjuvant hormonal therapy after radical prostatectomy is uncertain, given the lack of prospective clinical trials of this approach without radiation therapy after radical prostatectomy. The addition of high-dose bicalutamide to watchful waiting, radical prostatectomy, and radiation therapy was investigated in three complementary, double-blind, placebo-controlled trials enrolling 8,113 patients.⁵⁷ Treatment was for 2 years in Trial 23 and until disease progression in Trials 24 and 25. At a median follow-up of 10 years, bicalutamide resulted in an improvement in progression-free survival in men with extraprostatic disease (T3–4), regardless of choice of local therapy, although no overall survival benefit was observed. For patients with lymph node–positive disease at the time of radical prostatectomy, immediate ADT until disease progression versus delayed treatment was investigated in a study of 98 patients.⁵⁸ Although immediate ADT improved overall survival and prostate cancer–specific survival and reduced the risk of recurrence in patients with node-positive prostate cancer, this single-institution, relatively small-scale study did not test the use of ADT for PSA recurrence and did not investigate the long-term quality-of-life impact of long-term ADT.⁵⁹ At the present time, use of adjuvant hormonal therapy after radical prostatectomy remains controversial. The phase III AFU-GETUG-20 trial (NCT01442246) will evaluate adjuvant leuprolide for 2 years after radical prostatectomy. The primary endpoint is 10-year metastasis-free survival. The trial has completed accrual of 700 patients, and results, which are expected in 2027, will inform the role of adjuvant hormonal therapy after radical prostatectomy. Additionally, the ERADICATE trial will investigate the role of adjuvant ADT with darolutamide after prostatectomy in patients with high-risk prostate cancer. Several studies have investigated the role of adjuvant chemohormonal therapy after radical prostatectomy. Collectively, these trials have largely been unsuccessful, and this approach is not used in clinical practice after surgery.

Systemic Therapy in Combination With Radiation Therapy

Hormone therapy

Hormonal therapy with radiation therapy is a standard of care for patients with high-risk localized prostate cancer (Table 2). Several clinical trials have established the role of ADT with radiation therapy, given improvements in cancer-specific and overall survival. The landmark EORTC 22863 trial evaluated radiation therapy with or without 3 years of ADT.⁶⁰ At a median follow-up of 9 years, the 10-year disease-free survival (48% vs. 23%; HR, 0.42; 95% CI, 0.33–0.53) and overall survival (58% vs. 40%; HR, 0.60; 95% CI, 0.45–0.80) were improved with combination therapy compared with radiation alone. Additionally, prostate cancer mortality was decreased (10% vs. 30%; HR, 0.38; 95% CI, 0.24–0.60).

TABLE 2.

Phase III Randomized Trials of ADT Use in Combination With Radiation Therapy for Localized Prostate Cancer

Trial	Year	Stage	No. of Patients	Arms	RT	Primary Endpoint	OS
RTOG 85-31 61	2005	T3N0-1 (15% RP)	977	RT ± LHRHa/orchiectomy until progression	65–70 Gy		Improved survival (10-year survival 49% vs. 39%)
RTOG 86-10 62	2008	T2-4N0-1	456	RT ± LHRHa + flutamide 2 months before and concurrent vs. no ADT	65–70 Gy	10-year OS	No difference (10-year OS 43% vs. 34%)
RTOG 92-02 63	2008	T2c-4N0-1M0	1,554	LHRHa + flutamide × 4 months + RT ± 24-month adjuvant LHRHa	65–70 Gy	10-year DFS	Improved OS in Gleason 8–10 subgroup (45% vs. 32%)
EORTC 22961 64	2009	T1c-2abN1M0 or T2c-4N0-1M0	970	LHRHa + antiandrogen × 6 months + RT ± 30-month adjuvant LHRHa	70 Gy	OS (noninferior)	Improved OS (5-year mortality 15% vs. 19%)
EORTC 22836 60	2010	T1-2 WHO grade 3 or T3-4	415	RT ± LHRHa × 3 years	70 Gy	DFS	Improved OS (10-year OS 58% vs. 40%)
TROG 96-01 65	2011	T2b-4N0M0	802	Neoadjuvant LHRHa + flutamide (0 vs. 3 vs. 6 months) + RT	66 Gy	Time to local failure; PCSM	Improved PCSM (10-year PCSM 11% vs. 22% for 6 mos vs. no ADT)
DART 01/05 GICOR 66	2015	T1c-T3bN0M0	255	LHRHa + antiandrogen × 4 months + RT ± 24-month adjuvant LHRHa	76 Gy	BDFS	Improved OS (5-year OS 95% vs. 86%)
Nabid 67	2018	T3-4 or PSA > 20 ng/mL or Gleason > 7	630	LHRHa + antiandrogen × 4 months + RT + 36- vs. 18-month adjuvant LHRHa	70 Gy	OS	No difference (5-year OS 91% vs. 86%)
TROG 03-04 68	2019	T2b-N0M0 or T2a + PSA ≥ 10 ng/mL and Gleason ≥ 7	1,071	Neoadjuvant LHRHa x 6 months + RT ± 12-month adjuvant LHRHa ± 18-month zoledronic acid	66–74 Gy (or 46 Gy + HDB boost 19.5 Gy)	PCSM	Improved PCSM (10% vs. 13% for 18 vs. 6 months of ADT)

Abbreviations: ADT, androgen deprivation therapy; RT, radiation therapy; OS, overall survival; RP, radical prostatectomy; LHRHa, lutenizing hormone releasing hormone agonist; DFS, disease-free survival; PCSM, prostate cancer–specific mortality; BDFS, biochemical disease-free survival; PSA, prostate-specific antigen; HDB, high-dose brachytherapy boost.

Efforts to decrease the toxicity associated with prolonged ADT have led to several clinical trials comparing short- and long-course ADT. The optimal duration of therapy has not been firmly established. RTOG 92–02⁶³ and DART 01/05 GICOR,⁶⁶ which investigated 4 versus 28 months of ADT, and EORTC 22961,⁶⁴ which investigated 6 versus 36 months of ADT, demonstrated improvements in overall survival with prolonged ADT. Nabid and colleagues⁶⁷ investigated 18 versus 36 months of ADT. Although 36 months was not superior to 18 months of ADT, the trial was not designed as a noninferiority trial, and therefore equivalency between the two arms cannot be established from the results of this study. Guidelines from several panels, including ASCO, American Urologic Association, ASTRO, and the Society of Urologic Oncology, generally recommend prolonged ADT for men with high-risk localized prostate cancer (24–36 months) rather than a shorter duration of therapy.^5,6,9 The use of a luteinizing hormone-releasing hormone agonist with or without an antiandrogen has not been formally addressed in a randomized trial. Although data from observational studies support combination therapy, these studies did not systematically evaluate longer durations (greater than 4 months) of combination therapy.⁶⁹

The addition of abiraterone to ADT for patients with newly diagnosed hormone-sensitive prostate cancer was investigated in the STAMPEDE trial, a multiarm, multistage, randomized controlled trial.³⁸ This study included a heterogeneous group of patients, including 27% with newly diagnosed node-negative, nonmetastatic prostate cancer. These patients received abiraterone plus ADT versus ADT alone for 2 years and were mandated to receive definitive radiation therapy. In subgroup analyses by metastasis status, nodal status, and planned radiation therapy, overall survival favored abiraterone plus ADT compared with ADT alone (M0: HR, 0.75; 95% CI, 0.48–1.18; N0: HR, 0.69; 95% CI, 0.49–0.96; planned radiotherapy: HR, 0.64; 95% CI, 0.38–1.08). The use of abiraterone in conjunction with ADT and radiation therapy is controversial and requires a thoughtful discussion between patients and clinicians on the risks and benefits of this approach. Two phase III studies are investigating the addition of apalutamide (ATLAS, NCT02531516) and enzalutamide (ENZARAD, NCT02446444) combined with ADT for patients with high-risk prostate cancer undergoing primary radiation therapy. Both studies have completed accrual. The primary endpoint for the ATLAS study is metastasis-free survival, and the primary endpoint for the ENZARAD study is overall survival.

Chemotherapy

The addition of chemotherapy to radiation therapy for patients with localized prostate cancer has not been established as a standard practice. The phase III RTOG 05–21 trial investigated the benefit of adjuvant docetaxel for six cycles to 24 months of ADT and radiotherapy in patients with high-risk prostate cancer.⁷⁰ At a median follow-up of 5.7 years, disease-free survival and overall survival were improved with adjuvant chemotherapy (4-year overall survival, 93% vs. 89%; HR, 0.69; 95% CI, 0.49–0.97). In addition, evidence supporting the potential utility of chemotherapy for patients with localized prostate cancer stem from the STAMPEDE trial.⁵⁴ In the stage of the study testing docetaxel for hormone-sensitive disease, 22% of patients has localized disease. Radiotherapy was initially encouraged and then later mandated for patients with N0M0 disease. Overall survival in patients with disease staged as M0 and N0 favored docetaxel (M0: HR, 0.95; 95% CI, 0.62–1.47; N0: HR, 0.58; 95% CI, 0.41–0.81); however, in patients with planned radiation therapy, overall survival favored ADT alone (HR, 1.11; 95% CI, 0.67–1.85). In a meta-analysis of three trials evaluating docetaxel of M0 prostate cancer (GETUG-12, RTOG 05–21, and STAMPEDE), no overall survival benefit was observed with the addition of docetaxel chemotherapy.⁷¹ Additional data are needed to evaluate the effects of docetaxel for patients with M0 disease. The PEACE2 trial will investigate the role of cabazitaxel for patients with localized high-risk prostate cancer (NCT01952223). The trial, which has a primary endpoint for progression-free survival, is actively enrolling patients and is expected to report results in December 2025.

BIOMARKERS TO GUIDE THERAPY SELECTION FOR LOCALIZED PROSTATE CANCER

The use of biomarkers to guide treatment approaches in high-risk prostate cancer is evolving, as evidenced by recent changes in NCCN guidelines. In the 2019 guidelines, biomarkers were not recommended for initial risk stratification within high-risk populations, but the most recent 2020 guidelines advocate for consideration of the Decipher and Prolaris tumor-based molecular assays in men with high-risk disease and life expectancy of at least 10 years. To date, Decipher is the most extensively studied biomarker assay within the high-risk population. Large studies have demonstrated that the addition of Decipher to clinicopathologic variables improves the estimation of risk of distant metastases compared with risk stratification by NCCN groups alone.⁷² In addition, in studies focused specifically within patients with high-risk prostate cancer, Decipher is an independent predictor of metastatic progression even when accounting for clinicopathologic features in patients treated with surgery or with radiotherapy.^73,74 Recent data also support the prognostic value of Prolaris in predicting metastatic progression in patients with NCCN intermediate- or high-risk prostate cancer.⁷⁵

As a consequence of these retrospective studies demonstrating the benefit of biomarkers, a large, prospective, biomarker-driven phase III clinical trial (the PREDICT-RT trial, NRG GU009) is being initiated for patients with high-risk prostate cancer. In this study, patients with high-risk disease, by NCCN criteria, will undergo biomarker testing of their tumor sample with the Decipher assay. Those with high Decipher scores will be enrolled in a phase III randomized trial of treatment intensification, assessing the addition of intensified androgen ablation (with abiraterone and apalutamide) to the standard of care (radiation and 24 months of ADT). Those with low or intermediate Decipher scores will be enrolled in a phase III randomized trial of treatment deintensification, in which a de-intensified regimen (radiation and 12 months of ADT) is compared with the standard of care (radiation and 24 months of ADT).

In addition to improving risk prediction in treatment-naïve patients who are at high risk, biomarkers also have a role in improving the assessment of risk of disease progression after radical prostatectomy, particularly for patients at high risk. Current NCCN guidelines state that the “Decipher molecular assay is recommended to inform adjuvant treatment if adverse features are found post-radical prostatectomy”; these guidelines are based on a number of studies validating the prognostic value of Decipher in the postprostatectomy space, including a large meta-analyses of 975 patients demonstrating that Decipher independently predicts metastatic progression within nearly all clinicopathologic, demographic, and treatment subgroups.⁷⁶ More recently, a study validated the prognostic value of Decipher in patient samples from the RTOG 9601 trial, in which patients with PSA recurrences after surgery were randomly assigned to radiation alone or in combination with 2 years of high-dose bicalutamide.⁷⁷ Intriguingly, this study also suggested that only patients with intermediate to high Decipher scores derived benefit from the antiandrogen therapy, and patients with low Decipher scores did not.⁷⁸ Based on these findings, a number of trials have used or are incorporating Decipher to select patients at high risk for adjuvant therapies after prostatectomy. These include the NRG-GU002 RADD randomized trial (NCT03070886), which is investigating the addition of adjuvant docetaxel to radiation and ADT for patients with persistently elevated PSAs after prostatectomy, as well as the ECOG ERADICATE trial, in which patients who had a prostatectomy with high Decipher scores were randomly assigned to 12 months of ADT with or without 12 months of darolutamide.

It should be noted that the major assays (Decipher, Prolaris, Oncotype DX, or Promark) were designed as prognostic, but not specifically predictive, biomarker panels. A prognostic biomarker provides information on outcomes independent of the treatment received. Thus, aggressive disease, as identified by a prognostic test, may be aggressive regardless of the treatment received. In contrast, a predictive biomarker specifically identifies response or resistance to a particular therapy, but not to all treatments. Although prognostic biomarkers are useful (e.g., in identifying patients with aggressive disease who should be enrolled in studies of treatment intensification), predictive biomarkers are ultimately needed to select the right treatment of each patient. A predictive biomarker panel has been developed and validated to predict the benefit of postoperative radiation; this panel is called the postoperative radiation therapy outcome score and is available through the Decipher platform⁷⁹; however, it should be noted that postoperative radiation therapy outcome score has not been validated prospectively. In addition, another predictive biomarker panel, the PAM50 classifier, has been optimized and validated to predict the benefit of postoperative ADT; this panel, which was originally derived in breast cancer samples and predicts benefit from endocrine therapy in that disease, has also been validated, in retrospective cohorts, to predicting benefit from androgen-directed therapies in prostate cancer.⁸⁰ PAM50 is also being prospectively tested in the NRG GU006 BALANCE trial (NCT03371719), in which patients with PSA recurrences after prostatectomy are randomly assigned to salvage radiation alone or in combination with 6 months of apalutamide.

The biomarker assays discussed thus far have involved testing of samples of prostate cancer. However, it should be noted that NCCN guidelines now includes a recommendation of germline genetic testing (i.e., testing of blood or saliva samples for alterations present in a patient) for all patients with high-risk prostate cancer. Specific genes that should be assessed via germline testing include BRCA1, BRCA2, ATM, PALB2, MLH1, MSH2, MSH6, and PMS2.⁸¹ Several studies have investigated the impact of germline mutations in BRCA2, a key DNA repair gene involved in homologous recombination, on treatment outcomes in patients with prostate cancer. Germline mutations in DNA repair genes occur in approximately 4.6% of patients with localized prostate cancer and 11.8% to 16.2% of patients with metastatic disease; among these, BRCA2 alterations are by far the most common.⁸² For men with localized prostate cancer treated with local therapy (i.e., either radical prostatectomy or radiation), the presence of a pathogenic germline BRCA2 mutation is associated with worse metastases-free survival and prostate cancer cause-specific survival compared with patients who do not harbor such a mutation.⁸¹ A more recent study suggested that, in a cohort of 67 carriers of BRCA2 or BRCA1 germline mutations (compared with 1,235 noncarriers), prostate cancer–specific survival was decreased in patients with the BRCA mutation treated with radiotherapy compared with patients who did not have germline BRCA alterations; no difference was seen in prostate cancer–specific survival between BRCA carriers and noncarriers treated with radical prostatectomy.⁸³ Although this study has led some researchers to conclude that patients with BRCA2-mutant prostate cancer might be better treated with surgery, it is difficult to draw practice-changing conclusions from a small retrospective study in which patients treated with radiation had far more aggressive disease than patients treated with surgery (radiation therapy vs. radical prostatectomy cohort: median PSA, 14 vs. 7.5 ng/mL; Gleason score of at least 8 in 20% vs. 10%; patients at high risk, 24% vs. 56%) and in which there was no significant statistical interaction between BRCA status and treatment modality for either metastasis-free survival or prostate cancer–specific survival. Thus, the issue of whether BRCA status should influence treatment selection remains an area of active research.

CONCLUSIONS

Multimodal treatment strategies of surgery, radiation therapy, and systemic therapy offer the greatest potential for improved long-term outcomes for patients with high-risk prostate cancer who may harbor occult metastatic disease. Integrated multidisciplinary teams of urologists, medical oncologists, radiation oncologists, radiologists, and pathologists will be instrumental in shifting the treatment tide for patients. Novel systemic therapies, including hormonal, cytotoxic, targeted, and immunologic agents, tested in the context of rationally designed clinical trials will help better refine therapies for this heterogeneous group of patients. Such biomarker-driven trials will be of increasing importance in the era of precision medicine.

PRACTICAL APPLICATIONS.

• Patients with high-risk prostate cancer have an increased risk of disease recurrence and death from prostate cancer.

• Local treatment strategies include definitive radiotherapy or radical prostatectomy with or without adjuvant or salvage radiation therapy.

• The backbone of systemic therapy for patients with high-risk disease includes androgen deprivation therapy, although many questions remain regarding the use with surgery, intensity, and duration of androgen deprivation therapy.

• Blood- and tissue-based biomarkers to guide therapy selection continue to be an area of active research, and contemporary clinical trials are integrating such predictive biomarkers to better guide therapy selection for patients at high risk.