Surgery, radiotherapy and endocrine therapy for oligometastatic prostate cancer efficacy: a systematic review and network meta-analysis

Introduction

Oligometastatic prostate cancer (omPCa) has garnered increasing attention in recent years, with a growing body of prospective studies exploring its management. Androgen deprivation therapy (ADT) and other endocrine therapies remain the cornerstone of treatment for metastatic prostate cancer, representing a pivotal advancement in the care of advanced disease by substantially improving patient survival (Cornford et al., 2017). In contrast, surgical interventions are seldom recommended in this setting.

A recent review has challenged conventional treatment paradigms, reporting a 5-year overall survival rate of 55% for surgery, compared to 25.3% for radiotherapy and 41.3% for endocrine ADT in oligometastatic prostate cancer (Tosoian et al., 2017). These findings suggest a potential survival benefit with surgical intervention, yet the evidence remains insufficiently explored. Consequently, the optimal therapeutic strategy—whether surgery, radiotherapy, or endocrine therapy—for improving prognosis in oligometastatic prostate cancer remains unresolved.

To address this uncertainty, we conducted a systematic review and network meta-analysis (NMA) to evaluate the comparative efficacy of surgical, radiotherapeutic, and pharmacological endocrine therapies in oligometastatic prostate cancer. Our aim was to synthesize existing evidence and identify the most effective treatment approach for this patient population.

Methods

The methodology for this systematic review and network meta-analysis (NMA) was adapted from established protocols described in prior research (Page et al., 2021), with adherence to the Preferred Reporting Items for Systematic Review and Meta-Analysis extension for Network Meta-Analyses (PRISMA-NMA) guidelines. The study protocol followed standard registration procedures through PROSPERO (CRD42022370203).

This framework structured the Medline, EMBASE, and Cochrane database searches into three conceptual tiers: (1) intervention terms (“Radical prostatectomy,” “Radiotherapy,” “Androgen receptor axis-targeted therapy”), (2) condition terms (“Prostate Neoplasms” AND “Oligometastatic”), and (3) study design filters (“Randomized Controlled Trial”). The full search syntax with Medical Subject Headings (MeSH) terms and keyword permutations has been archived on INPLASY Protocol to ensure reproducibility (Ying et al., 2022). Inclusion criteria were restricted to randomized two-arm trials evaluating active interventions for oligometastatic prostate cancer.

We included patients with oligometastatic prostate cancer in randomized controlled trials and combined androgen-deprivation therapy (ADT) with one or more of the previously listed treatments. In the 13 articles eventually included, the diagnosis of oligometastatic prostate cancer was based on conventional imaging techniques (such as scintigraphic bone scans, thoracic computed tomography (CT), abdominal CT, and pelvic magnetic resonance imaging (MRI) or CT), as well as histological and/or cytological confirmation.

Initially, two independent authors (Wenwei Ying and Zhenshan Ding) performed the search strategy and screened the search results for relevance based on the topics and abstracts. Any disagreements between them were resolved by consulting a third author (Yanqing Gong). The full texts of the 58 articles identified as potentially relevant were then reviewed by the same two independent authors (Wenwei Ying and Zhenshan Ding) to determine if they met the predefined inclusion and exclusion criteria. If multiple reports of the same trial were found, only the most recent publication was included in the analysis. Data extraction was also conducted by the two independent authors (Wenwei Ying and Zhenshan Ding).

The primary outcome was overall survival (OS), measured as the time from randomization to death from any cause. We also evaluated secondary endpoints, including progression-free survival (PFS), defined as the time from enrollment to prostate-specific antigen (PSA) progression, quality of life (QoL), and treatment-related adverse events (TRAEs).

Results

After a rigorous screening process involving abstract review and full-text assessment, 13 eligible studies comprising 2,524 patients were ultimately included in this systematic review and network meta-analysis (Fig. 1). Table 1 presents the detailed clinical characteristics of the included studies. The investigated interventions encompassed a range of treatment modalities: observation, androgen-deprivation therapy (ADT), external beam radiation therapy (EBRT), metastasis-directed therapy (MDT), radiation therapy (RT), stereotactic ablative radiotherapy (SABR), as well as combination therapies such as ADT+RT, ADT+SABR, ADT+abiraterone, ADT+enzalutamide (ENZA), ADT+EBRT, and ADT+abiraterone+RT.

The risk of bias for each trial is reported in Fig. S3. For network meta-analysis purposes, the ADT arm from each trial served as the common comparator across all treatment comparisons (Fig. 2).

Progression-free survival

Eight randomized trials evaluating the impact of various treatment modalities on progression free survival (PFS) were included in this NMA, with the treatment network illustrated in Fig. 2A. The funnel plot demonstrated a roughly symmetrical distribution of included studies, indicating no substantial publication bias (Fig. S1A). Compared to ADT alone, significant reductions in disease progression risk were observed with combination therapies: ADT+ RT (HR = 0.39, 95% CI [0.27–0.56]), ADT + SBRT (HR = 0.35, 95% CI [0.21–0.58]), and ENZA + ADT (HR = 0.27, 95% CI [0.16–0.46]). In contrast, RT alone failed to demonstrate significant clinical benefit (HR = 0.60, 95% CI [0.31–1.13]) (Fig. 3A). Interestingly, while SUCRA analysis initially suggested ADT monotherapy as the most likely optimal treatment, comprehensive league table evaluation revealed that ADT combined with RT may provide superior PFS outcomes. Furthermore, ADT + SBRT also showed considerable potential for improving PFS (Table 2, Fig. S2A).

Table 2:

League table with pooled standardized mean difference for PFS, OS, TREA, QoL.

HR with 95% confidence interval on PFS
ADT	2.59 (1.80, 3.74)	2.86 (1.73, 4.71)	3.70 (2.18, 6.28)	1.00 (0.10, 10.00)	.	.
2.56 (1.78, 3.69)	ADT + RT	.	.	1.00 (0.10, 10.00)	.	.
2.86 (1.73, 4.71)	1.12 (0.60, 2.07)	ADT + SBRT	.	.	.	.
3.70 (2.18, 6.28)	1.45 (0.76, 2.75)	1.30 (0.63, 2.68)	ENZA + ADT	.	.	.
1.60 (0.31, 8.24)	0.62 (0.12, 3.22)	0.56 (0.10, 3.11)	0.43 (0.08, 2.42)	Observation	1.67 (0.87, 3.18)	3.33 (1.23, 9.05)
2.67 (0.46, 15.52)	1.04 (0.18, 6.06)	0.93 (0.15, 5.82)	0.72 (0.11, 4.53)	1.67 (0.87, 3.18)	RT	.
HR with 95% confidence interval on OS
ADT	1.11 (0.59, 2.08)	1.54 (0.88, 2.71)	1.69 (0.84, 3.40)	2.00 (0.94, 4.27)	.	.	1.00 (0.09, 10.74)
1.11 (0.59, 2.08)	ADT+EBRT	.	.	.	.	.	.
1.52 (0.87, 2.66)	1.37 (0.59, 3.17)	ADT+RT	.	.	1.00 (0.01, 103.70)	.	1.00 (0.09, 10.74)
1.69 (0.84, 3.40)	1.53 (0.60, 3.90)	1.11 (0.46, 2.71)	ENZA+ADT	.	.	.	.
2.00 (0.94, 4.27)	1.80 (0.67, 4.82)	1.31 (0.51, 3.36)	1.18 (0.42, 3.31)	ERBT	.	.	.
1.52 (0.01, 163.36)	1.37 (0.01, 153.34)	1.00 (0.01, 103.70)	0.90 (0.01, 101.50)	0.76 (0.01, 86.83)	MDT	0.60 (0.30, 1.22)	.
0.91 (0.01, 103.37)	0.82 (0.01, 96.98)	0.60 (0.01, 65.64)	0.54 (0.00, 64.19)	0.46 (0.00, 54.91)	0.60 (0.30, 1.22)	observation	.
1.23 (0.23, 6.77)	1.11 (0.18, 6.82)	0.81 (0.15, 4.44)	0.73 (0.12, 4.58)	0.62 (0.10, 3.98)	0.81 (0.01, 113.63)	1.35 (0.01, 199.15)	Observation
OR with 95% confidence interval on TRAEs
ADT	0.65 (0.48, 0.88)	.	0.90 (0.58, 1.40)	0.63 (0.10, 4.00)
0.65 (0.48, 0.88)	ADT+abiraterone	1.08 (0.56, 2.09)	.	.
0.70 (0.34, 1.45)	1.08 (0.56, 2.09)	ADT+abiraterone+RT	.	.
0.90 (0.58, 1.40)	1.39 (0.81, 2.37)	1.28 (0.55, 2.98)	ADT+ENZA	.
0.63 (0.10, 4.00)	0.98 (0.15, 6.32)	0.90 (0.12, 6.51)	0.71 (0.11, 4.69)	ADT+RT
SMD with 95% confidence interval on QoL
ADT	−0.22 [−0.78; 0.34]	0.00 [−0.51; 0.51]	.
−0.22 [−0.78; 0.34]	ADT+RT	.	.
−0.00 [−0.51; 0.51]	0.22 [−0.53; 0.98]	MDT	0.06 [−0.43; 0.56]
0.06 [−0.65; 0.77]	0.28 [−0.62; 1.19]	0.06 [−0.43; 0.56]	observation

DOI: 10.7717/peerj.19819/table-2

Discussion

Oligometastatic prostate cancer (omPCa) represents an intermediate stage of cancer dissemination between localized and widely metastatic disease. With dramatic improvements in staging—particularly through advanced functional imaging such as PSMA-PET and robotic-assisted (RA) targeted biopsies (RA-TB)—omPCa is now more frequently detected, both at primary diagnosis and in oligorecurrent disease (Patel et al., 2020). Singh et al. (2004) investigated the correlation between survival and the number of metastatic lesions observed in each prostate cancer patient. They found that men with ≤5 lesions had survival rates comparable to those without metastases, and these rates were significantly higher than those of patients with >5 lesions. These findings underscore the growing rationale for aggressive, metastasis-directed therapy (MDT) in omPCa. Retrospective and prospective studies suggest that surgical interventions (cytoreductive prostatectomy (CRP), salvage lymph node dissection (SLND), or metastasectomy) can achieve favorable short-term oncologic outcomes and local disease control (von Deimling et al., 2022). However, high-quality evidence on long-term oncological benefits remains limited. For de novo oligometastatic hormone-sensitive prostate cancer (omHSPC), ongoing trials are evaluating the role of primary tumor resection, while in metachronous disease, SLND should be reserved for carefully selected patients within a multimodal approach to eradicate detectable disease and prolong progression-free survival (PFS) (von Deimling et al., 2022). Notably, clinicians should remain vigilant for the potential presence of neuroendocrine prostate cancer (NEPC) in metastatic prostate cancer (mPCa) cases. This rare but highly aggressive variant necessitates immediate therapeutic strategy modifications (Wishahi, 2024).

Recent advances in cancer genomics, sequencing technologies, and functional genomics have led to the development of novel therapeutic approaches, including immunotherapies that have demonstrated promising efficacy (Gillessen et al., 2023; Ashrafizadeh et al., 2022). However, further research is required to fully elucidate their clinical benefits (Miyahira et al., 2022). For oligometastatic prostate cancer (omPCa), current management strategies typically involve local consolidation therapy (e.g., stereotactic body radiotherapy (SBRT)) targeting the primary tumor, followed by metastatic lesion treatment and subsequent systemic hormonal therapy. Clinical evidence from completed trials (NCT02192788, NCT02680587, NCT02264379) available on ClinicalTrials.gov has demonstrated sustained clinical benefits of SBRT in oligometastatic castration-sensitive prostate cancer (omCSPC). Ongoing investigations through currently recruiting studies (NCT04610372 [PROMPT], NCT04115007, NCT04599686) are expected to provide additional insights. While enhanced recovery after surgery (ERAS) protocols have been successfully implemented across various surgical specialties (Wishahi, Kamal & Hedaya, 2024), surgical intervention is generally contraindicated in metastatic prostate cancer. Emerging evidence suggests that primary tumor treatment may reduce the need for palliative interventions in locally advanced cases (Heidenreich, Pfister & Porres, 2015; Won et al., 2013) and potentially delay the initiation of systemic therapies such as androgen deprivation therapy (ADT)—an approach that could significantly improve quality of life (Taylor, Canfield & Du, 2009). This consideration is particularly relevant given that while ADT remains a cornerstone of metastatic prostate cancer treatment, it is associated with substantial adverse effects and may, in some patient populations, negatively impact overall survival (Taylor, Canfield & Du, 2009; Sammon et al., 2015). Our study findings align with these observations, demonstrating that ADT—whether administered alone or in combination with SBRT—showed the least significant improvement in PFS among treated patients.

Retrospective studies indicate that interventions like radical prostatectomy and targeted radiotherapy for either local or metastatic lesions can be performed in metastatic settings with minimal risk of severe toxic effects. Heidenreich, Pfister & Porres (2015) reported no Clavien grade IV or V complications in men undergoing radical prostatectomy, and the incidence of grade I–III complications was similar to or better than controls. Additionally, palliative intervention was needed in 28.9% (11 out of 38) of control patients compared to none who underwent radical prostatectomy. Moreover, 21 (91.3%) patients reported postoperative continence, requiring zero or one pad per day. A retrospective multi-institutional analysis of radical prostatectomy in distant metastasis settings showed similarly promising results (Sooriakumaran et al., 2016). The rates of complications, readmission, and reoperation were 20.8%, 3.8%, and 1.9%, respectively, compared to 19.4%, 3.0%, and 2.3%, respectively, in open cases for standard indications (Tewari et al., 2012). These treatments appear to reduce the requirement for subsequent palliative care, though there is insufficient data to reliably conclude their impact on survival. Therefore, a standardized clinical protocol for managing patients with metastatic prostate cancer would be a valuable clinical tool.

Our network meta-analysis (NMA) of 13 trials (n = 2, 524) underscores the superiority of ADT combined with RT (ADT + RT) in both progression-free survival (PFS: SUCRA = 51.4%, HR = 0.39) and QoL (OS: SUCRA = 74.3%, HR = 0.22), aligning with landmark trials like ENZAMET and ARCHES (Sweeney et al., 2023; Armstrong et al., 2019). However, combination therapies, including ADT + abiraterone, increased treatment-related adverse events (TRAEs) by 54% (OR = 1.54), while ranked lowest in safety (SUCRA = 25.8%). These findings highlight a critical dilemma—aggressive regimens maximize survival but exacerbate toxicity, whereas ADT monotherapy, though safer (SUCRA = 79.8%), offers limited efficacy.

Our study has limitations reflective of the broader field. First, the heterogeneous definition of “oligometastatic” disease complicates cross-trial comparisons; standardized criteria (such as ≤5 lesions on PSMA-PET) are urgently needed. Second, approximated hazard ratios for studies reporting median survival times (such as Yildirim et al. (2019)) introduce bias, emphasizing the need for raw Kaplan–Meier curves in publications. Third, QoL outcomes were inconclusive due to sparse data (ADT + RT: SMD = 0.22, 95% CI [−0.34–0.78]), underscoring the imperative to integrate patient-reported outcomes (PROs) in future trials. Finally, the combination of ADT with abiraterone demonstrated significantly elevated adverse event risk, necessitating rigorous clinical monitoring. Interestingly, our analysis revealed the counterintuitive finding that ADT+abiraterone+RT showed better safety outcomes than ADT+RT alone. We hypothesize this paradoxical result may stem from abiraterone’s potential to enhance RT feasibility through two mechanisms: (1) significant tumor downstaging prior to radiation, and (2) reduced tumor volume facilitating post-treatment recovery. The apparent TRAE reduction likely reflects decreased tumor progression rather than true treatment-related toxicity mitigation. Despite these limitations, the results of our analysis provide strong evidence for the treatment of oligometastatic prostate cancer and can serve as an important reference.

Conclusion

This network meta-analysis of 13 studies (2,524 patients) found that ADT + RT and ADT + SBRT significantly improved PFS (HR = 0.39 and 0.35 vs. ADT alone) but showed no OS benefit. ADT monotherapy had the lowest toxicity, while ADT + abiraterone increased grade ≥3 TRAEs (OR = 1.54). Limited data suggested ADT + RT may improve QoL.

Supplemental Information