Prostate cancer is a leading cause of morbidity and mortality in men worldwide. Radical prostatectomy (RP) is one of the most effective treatments of localized prostate cancer. However, recurrence after RP should not be overlooked. It has been reported that 20-50% of men still experience sustained or delayed increase of prostate specific antigen (PSA) within 5-10 years after prostatectomy, which is called biochemical recurrence (1).The level of PSA has proven to be a sensitive indicator for monitoring the recurrence of prostate cancer after operation (2). The recurrence of local prostate tumor in situ, metastasis to occult lymph nodes, or distant metastasis may account for biochemical recurrence (3–5). While imaging shows no signs of recurrence (6). Thus, patients with biochemical recurrence are at greater risk of metastasis (3). Consequently, biochemical recurrence after prostatectomy is a key clinical management issue for patients with prostate cancer. As a common salvage treatment for biochemical recurrence of PSA after RP, salvage radiotherapy (SRT) should be recommended to most patients when PSA elevates after prostatectomy (7).Whereas, the efficacy of SRT is unsatisfactory. Endocrine therapy can reduce local recurrence and distant metastasis of prostate cancer after RP (8). It is reported that (9), radiotherapy combined with endocrine therapy showed good efficacy as salvage therapy for biochemical recurrence of prostate cancer. Since most of the previous studies were retrospective, the addition of endocrine therapy to salvage radiotherapy was inconclusive. At present, some results in randomized controlled studies have come out, so we performed a systematic review and meta-analysis to evaluate the efficacy and safety of SRT combined with endocrine therapy in patients with biochemical recurrence after RP.

This study was registered in the PROSPERO database (CRD42022365432) and was conducted according to the preferred reporting project for systematic review and meta-analysis (PRISMA) statement (10).And this study aims to compare the efficacy and safety of SRT combined with endocrine therapy versus SRT alone in patients with biochemical recurrence after RP.

Serum PSA is considered as one of the most sensitive tumor markers. Therefore, for early detection of prostate cancer, PSA screening is reckoned as an effective method to reduce prostate cancer mortality (20), for biochemical recurrence seems to be the only evidence of disease progression in the absence of clinical symptoms. It is worth noting that bout a third of patients experience biochemical recurrence after RP, depending on their initial prognosis (1) which entails a comprehensive assessment of local and systemic recurrence so as to carry out active intervention. SRT has proven to enhance the outcomes of patients with biologically recurrent prostate cancer (7). But whether to add endocrine therapy to radiotherapy still remains controversial. This meta-analysis summarized randomized controlled studies evaluating the effect of SRT combined with endocrine therapy on postoperative biochemical recurrence of prostate cancer. The results indicated that SRT combined with endocrine therapy significantly improved bPFS and MFS, with a trend of benefit in the term of OS, even though the difference was not statistically significant.

A previous meta-analysis (21) involving 12 studies also showed a significant benefit of SRT combined with endocrine therapy for bPFS. However, most of the studies included in this meta-analysis were retrospective, with only two RCTs, so the credibility of the conclusion was very low. Our meta-analysis, which included only randomized controlled studies, not only updated the latest results of two previous RCTs, but also included two new studies. Preclinical studies have found superadditive effects of the combination of radiotherapy and endocrine therapy (22), suggesting a potential benefit of the combination. This was confirmed by clinical results, with all four included studies indicative of benefit for bPFS, which was in accordance with our findings.

As the first study to evaluate the effect of adding endocrine therapy to SRT on the prognosis of patients with biochemical recurrence after RP, the RTOG 9601 trial (18) revealed that SRT plus 24 months of bicalutamide anti-androgen therapy significantly improved long-term OS after preforming an analysis of 760 patients. SPPORT (17) and GETUG-AFU 16 (16) did not show significant benefits for OS. It is noteworthy that, endocrine therapy in SPPORT (17) and GETUG-AFU 16 (16) lasted no more than 6 months, whereas anti-androgen therapy in RTOG 9601 (18) lasted 24 months, which may account for the difference. The RADICALS-HD trial, whose results have not been released, enrolled 3000 men undergoing postoperative prostate bed radiotherapy and compared SRT alone versus SRT plus 6 months of ADT versus SRT plus 24 months of ADT (NCT00541047). The results of that trial are expected to provide more valuable guidance for clinical practice. In addition, longer follow-up is needed to determine the impact of interventions on OS in patients with biochemical recurrence after RP. The incidence of death events was low in all three studies, so follow-up time was also a factor affecting OS. As can be seen, SPPORT (17) and GETUG-AFU 16 (16) also showed a trend of prolonged OS, though it was not statistically significant. Our review indicates a similar result. In this case, MFS is seen as a good surrogate end point for OS. Xie et al. (23) collected individual patient data from 28905 patients from 28 trials. At the patient level, Kendall’s tau coefficient correlation with OS was 0.91 for MFS, demonstrating that MFS was a strong surrogate for OS in localized prostate cancer and was significantly associated with the risk of death from prostate cancer. Given the strong correlation between MFS and OS, it makes perfect sense to use MFS as the primary end point instead of OS. In our review, SRT plus endocrine therapy resulted in longer MFS, suggesting that this combination therapy is related to better prognosis of patients with biochemical recurrence after RP.

The four included trials did not fully agree on the definition of biochemical recurrence. RTOG 9601 (18) and GETUG-AFU 16 (16) triggered SRT at a PSA level of 0.2ng/ml, while a PSA level of 0.1ng/ml was required for SPPORT (17) and SALV-ENZA (15). Therefore, subgroup analyses were performed to explore this difference. The results showed that the definition of biochemical recurrence before treatment had no effect on bPFS. In terms of MFS and OS, the results of the subgroup triggering SRT at 0.2ng/ml PSA level were consistent with the overall effect. The definition of biochemical recurrence after treatment was also slightly different. However, a Japanese study reported that this difference contributed to the overall effect difference, but the effect was small (24). This study reviewed the data of 118 patients who underwent SRT for biochemical recurrence after RP. The 2-year bPFS rates defined using Nara, RTOG 9601 and GETUG-AFU 16 were 59%, 70% and 69%, respectively. In addition, all four trials had a clear benefit with respect for bPFS which seemed unaffected by differences in definition.

The baseline characteristics of prostate cancer patients have potential impact to the effect of endocrine therapy. In a recently published meta-analysis, multiple analyses involving androgen-receptor-axis-targeted (ARAT) based treatment were conducted by pooling the PFS- and OS-HRs of different between subgroups defined according to different baseline characteristics, with the intent to identify predictors of efficacy in ARAT therapies for metastatic castration-sensitive prostate cancer (25). The authors concluded that prior docetaxel use and tumor volume/presence of visceral metastasis were the only factors that negatively affected the efficacy of ARAT, which seems to indicate that metastatic prostate cancer with a low tumor burden may benefit more from ARAT treatment. However, patients with pT3 localized prostate cancer in the SALV-ENZA trial benefited more significantly from enzulamide than patients with pT2 (interaction p =0.031) (15). Similarly, the subgroup analysis of the RTOG 9601 trial found the greatest overall survival benefit in the subgroup of patients with more aggressive prostate cancer (18). In another meta-analysis, an overall survival benefit from hormone therapy was found in RTOG 9601 trial limited to follow-up extended to ≥10 years, pre-SRT PSA ≥0.7ng/ml, or the presence of a higher gleason grade or positive margin (26). In the secondary analysis of RTOG 9601, higher pre-SRT PSA levels (PSA>0.6ng/mL) and higher Decipher scores were associated with greater benefit of hormone therapy (27, 28). Previous studies have found that providing SRT in lower PSA levels is associated with improved outcomes (29, 30). As a result, men with low PSA levels get less absolute benefit from hormone therapy. Patients with higher PSA levels are more likely to benefit from ADT, possibly because additional ADT can eliminate radiation-offsite subclinical metastases. In contrast, the insufficient effect of positive margins on the risk of metastasis may be related to higher dose administration and more timely use of SRT, which may lead to better local disease control (26, 31, 32). A cohort study found that ADT was beneficial only in patients with more aggressive traits, namely, pT3b/4 and grade group>4 or pT3b/4 and PSA at eSRT>0.4 ng/ml (8), which is recommended by the current csco guidelines. In our analysis of these limited clinical trials, we found that patients with more aggressive prostate cancer (including high baseline PSA, gleason score ≥ 8, and pT3b/4) were more likely to benefit from endocrine therapy. Unfortunately, we failed to further explore the effect of baseline characteristics on endocrine therapy in patients with prostate cancer due to insufficient data for subgroup analyses. So these conclusions are still premature and need to be confirmed by more clinical trials.

Exploring prognostic factors helps establish accurate risk stratification. GETUG-AFU 16 found that baseline PSA levels, surgical margin status, PSA doubling time at recurrence, and seminal vesicle status had a negative impact on disease progression (16, 33). Moreover SPPORT found that baseline PSA levels, seminal vesicle involvement, and Gleason score were prognostic factors for disease progression (17). Heterogeneity of prostate cancer is one of the common reasons of biochemical recurrence after RP, so better stratification of prostate cancer patients from diagnosis to early biochemical recurrence is essential. However, current risk stratification models are limited. An emerging field of research is helping us solve this problem how to more accurately describe the prognosis for each patient and each stage of the disease. As a rapidly growing field of research, radiomics features (RFs) analysis transforms visual image information into in-depth features for quantitative study that can inform detection, risk stratification, and treatment (34). For example, due to its very high sensitivity and specificity, prostate-specific membrane antigen positron emission tomography (PSMA-PET) can improve the detection of biochemical recurrence and metastasis, especially at low PSA levels (35), which has been difficult to achieve with previous detection methods. PSMA-PET can identify patients who need additional lymph node radiotherapy or ADT by locating the lymph node recurrence in patients with biochemical recurrence with aims to provide more personalized treatment (36–38). Studies have found that SRT guided by PSMA-PET has a better prognosis than non-PSMA-PET guided SRT (39). In addition, the artificial intelligence (AI) and machine-learning (ML) techniques are attractive research fields in prostate cancer. With high predictive accuracy, AI and ML can integrate medical imaging, genetic and clinical big data to obtain more information about the molecular and biological aspects of prostate cancer (34). By combining RFs with AI and ML methods, these applications could hold promise for personalized and precision therapy in the future.

For adverse effects, this meta-analysis found that endocrine therapy did not induce more severe grade 3+ acute toxicities, but would increase the risk of overall grade 3+ chronic toxicities. Among the common chronic toxicities, compared with placebo, endocrine therapy induced higher severity of sexual dysfunction, without increasing the incidence of renal or genitourinary, hematologic and gastrointestinal toxicities. Endocrine treatment-related cardiotoxicity is a common concern because severe cardiotoxicity can be fatal. In the RTOG 9601 trial, cardiac events were 17 in the bicalutamide group and 6 in the placebo group (18). The secondary analysis of RTOG 9601 found that an increase in other-cause mortality (OCM) from the primary cause of early PSA was seen in low PSA (0.2 to 0.3 ng/mL) patients, who had a 13.3% reduction in 12-year OS (27). LHRH agonists and GNRH antagonists also lead to higher rates of cardiac events, but the rates of cardiovascular events are somewhat different (40). A meta-analysis found that degarelix was associated with a lower incidence of cardiovascular events compared with LHRH agonists (41). But degarelix had a higher incidence of cardiovascular events in patients with preexisting cardiovascular disease (42). Overall, long-term antiandrogen therapy is associated with increased cardiovascular events and mortality, which may offset the increased potential survival in patients with low pre-SRT PSA.

This meta-analysis has several strengths including the prospective registration of the systematic review protocol and no heterogeneity in any efficacy end point. Additionally, to eliminate the limitation of follow-up time, we pooled MFS data with the aim of attempting to replace OS to evaluate the efficacy of endocrine therapy and showed that additional endocrine therapy could improve MFS. Moreover, our review represents the full range of randomized evidence for the effect of adding endocrine therapy to SRT in patients with biochemical recurrence after RP. There are also limitations in our study. Firstly, only 4 trials were included in our review. Secondly, there were some differences in the definitions of biochemical recurrence before and after treatment among the included trials. Thirdly, it is difficult to explore the effect of baseline characteristics on endocrine therapy for lacking enough data to perform subgroup analyses of baseline characteristics in prostate cancer patients. Fourthly, the follow-up time of most studies was insufficient, making it difficult to evaluate the effect of endocrine therapy on OS more comprehensively, leading to certain deviations.

MFS is a strong surrogate of overall survival in localized prostate cancer. This meta-analysis found superior bPFS and MFS associated with additional endocrine therapy in patients with biochemical recurrence after RP. Although the follow-up time was limited, there was also a trend of benefit in the term of OS. Therefore, adding endocrine to SRT is a reasonable and necessary. But more high-quality studies are needed to determine the duration and specific drugs of endocrine therapy. Patient, tumor, and treatment factors should be considered when using hormone therapy, as individualized therapy is the current trend.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/ Supplementary Material .

ZL, SL, HL, and CF conceived and guide the study. ZL and SL performed statistical analysis, and drafted the manuscript. HZ, LL, and JW collected the literature, edited figures and revised the manuscript. All authors contributed to the article and approved the submitted version.