How to tailor the radiotherapy after radical prostatectomy in the modern era of molecular imaging, genomic testing, and technology development?

Abstract

In recent years, we have observed a significant increase in the number of radical prostatectomies (RP) performed in patients at very high risk of prostate cancer. This group of patients is very heterogeneous due to the presence of multiple risk factors for recurrence, which makes decisions regarding the use of salvage radiotherapy (RT) very challenging. On the one hand, there is a subgroup of patients who can only be observed. For a more advanced subgroup of patients postoperative RT alone is effective treatment. For patients with very aggressive biology characteristics RT should be combined with systemic therapy. Unfortunately, we still do not have ideal tools to precisely assign individual patients to these subgroups. Clinicopathological factors are very helpful in this regard. So, introduction of modern molecular diagnostics, i.e., positron emission tomography (PET) using a radiotracer that binds to prostate-specific membrane antigen (PET-PSMA), allowed for a change in the general strategy for salvage RT. Information from PET scans included in the RT plan allow for increasing the effectiveness of RT. Another method for personalizing RT planning is the use of genomic testing. To date, the most clinically validated method allows for the identification of a subgroup of patients in whom combined treatment [androgen deprivation therapy (ADT) + RT] yields the greatest clinical benefits. Improvement in RT outcomes is associated with the introduction of technologically advanced RT methods, such as adaptive RT (ART), which provides the opportunity for a precise delivery of dose to a tumour bed. Another very promising development involves new predictive tests for RT which are based on genomic analysis of cancer tissue.

Introduction

In recent years, we have observed a significant increase in the number of radical prostatectomies (RP) performed in patients at high and even very high risk of prostate cancer [1]. In these risk groups, even in high-volume centers the biochemical recurrence, observed within 5 years from RP ranged from 40% to 70%, depending on case mix [2–4]. Generally, for loco-regional recurrences after RP the only effective local therapy is postoperative RT, which is considered to be a second- line treatment [5]. Depending on the timing of RT implementation, we distinguish early adjuvant RT [prostate-specific antigen (PSA) levels below the biochemical recurrence threshold < 0.1 ng/mL and not rising] and salvage RT (early salvage; PSA levels > 0.1 ng/mL or rising in consecutive measurements and late salvage; PSA > 0.5 ng/mL) [6, 7]. Nowadays, we significantly improved our ability to administer postoperative RT to patients after RP. Better understanding the relationship between PSA level/kinetics, PET-PSMA results, and genetic testing allows us to use the RT more precisely [8–11]. Currently we are moving toward personalization of RT, especially when RT is combined with hormonal therapy. What’s more, the technological advances such as adaptive RT combined with knowledge from positron emission tomography (PET) using a radiotracer that binds to prostate-specific membrane antigen (PET-PSMA) scans enable safe dose escalation [12–14].

Early adjuvant versus early salvage radiotherapy

Historically, early adjuvant RT was preferred, based on results from four randomized clinical trials (RCTs) in patients with risk factors [pT3a, pT3b, R1, International Society of Urological Pathology (ISUP) group 4–5] but without biochemical recurrence and/or distant metastases [15–18]. Results from RCTs clearly showed that early adjuvant RT significantly prolongs biochemical progression-free survival, which raises the question of how improvements in biochemical control translate into overall survival (OS) and prostate cancer-specific survival (PCSS). Unfortunately, the results of these trials to date do not clearly confirm improvement in OS or PCSS. The main issues of early adjuvant RT are the extra side effects of RT, which are added to the side effects of RP. The rationale for adjuvant RT was further questioned by the fact that, despite the presence of pathological recurrence risk factors, a significant number of patients do not experience biochemical failure and, consequently, cancer progression and death from prostate cancer. Therefore, it is logical to assume that adjuvant RT should only be used when biochemical recurrence has already occurred. Unfortunately, this approach has its drawback. The arbitrarily adopted cut-off PSA level may be the result remaining in the postoperative bed non-cancerous tissue. On the other hand, delaying RT may result in missing the therapeutic window for salvage RT due to the development of distant metastases. Some help could come from the ultrasensitive PSA tests but this is hampered by the fact that PSA is not a specific prostate cancer marker, and its level represents the sum of PSA production by normal tissue and prostate cancer cells. However, earlier identification of PSA trends may, to some extent, aid in making decisions about early initiation or discontinuation of salvage RT.

With the aim to determine which treatment option (adjuvant versus salvage) should be preferred 3 RCTs (RADICALS, GETUG-AFU 17, and RAVES trial) were conducted to compare the outcomes of early adjuvant RT with early salvage RT [19–21]. Although the goal of these studies was the same, the primary endpoints were different: metastasis-free survival in the RADICALS trial; event-free survival in the GETUG-AFU 17 trial; and biochemical failure-free survival in the RAVES trial. The timing of salvage RT between studies also varied depending on the PSA level and it was triggered at the PSA level of 0.2 ng/mL for RAVES trial, PSA level of 0.2 ng/mL and rising in GETUG-AFU 17 trial, and PSA > 0.1ng/ml or 3 consecutive rises below 0.1 ng/mL for RADICALS trial. These trials recruited men with similar pathological features: RADICALS included pT3 and pT4 disease, GETUG-AFU 17 was limited to pT3 or pT4a (with bladder neck invasion) and positive margins, and the RAVES trial included positive margins and pT2 or pT3.

Another difference between the studies concerned the use of ADT. In the GETUG AFU-17 trial, concomitant RT with ADT was used for 6 months, the RAVES trial used RT alone, while the RADICALS trial introduced the second randomization: long (24 months) vs short (6 months) ADT vs. no ADT.

A crucial conclusion from these studies was that early adjuvant RT compared to early salvage RT did not show survival benefit, but the toxicity of early adjuvant RT was significantly more pronounced in all studies. For example, acute GU toxicity in GETUG AFU 17 was higher for early RT (27% vs. 7%), and significantly worse erectile dysfunction (36% vs. 13%).

To summarize all the studies, the ARTISTIC meta-analysis was conducted, which confirmed no survival benefit with early adjuvant RT compared to early salvage RT [22]. Additional key finding from the meta-analysis was that patients in the salvage RT arm reported a significantly lower risk of post-RT complications compared to early adjuvant RT. This is because only a minority of patients are eligible for salvage RT (approximately 60% of patients do not receive RT), whereas 100% of patients received early adjuvant RT. Therefore, the use of early adjuvant RT in all patients with risk factors for cancer progression clearly poses the risk of overtreatment.

However, it is important to emphasize that in the above studies, there was poor representation of patients with multiple recurrence risk factors. Therefore, if pT3a or pT3b and Gleason score of ISUP 4–5 ± positive margin are present, which are related to the higher risk of rapid disease progression and risk of metastasis, it is advisable to consider early adjuvant RT [23].

The results of a multicenter study analyzing 2,379 patients after RP could be helpful in making decisions regarding the early or late salvage RT [24]. The study showed that in the low-risk European Association of Urology (EAU) progression group [PSA doubling time (PSA-DT) > 1 year and ISUP < 4], 12-year OS was 87% vs. 78% and PCSS was 100% vs. 96% for early salvage RT vs. late salvage RT, respectively. In the high-risk EAU progression subgroup (PSA-DT < 1 year and ISUP 4–5), 12-year OS was 81% vs. 66%, and PCSS was 98% vs. 82%.

In summary, it can be concluded that for low-risk patients, a wait-and-see approach is acceptable, while for high-risk patients, salvage RT should be implemented.

PSA level and PET-PSMA results help to decide when and what to irradiate

PSA is considered to be one of the best biomarkers in prostate cancer diagnosis but, unfortunately, we cannot precisely define PSA cutoff level above which we define biochemical recurrence and site of the recurrence [25–27]. The introduction of ultrasensitive PSA tests, which detect very low PSA levels, has not changed this situation, as PSA may be produced by the healthy portion of the prostate left during surgery. Therefore, a single PSA measurement is insufficient in such situations because PSA increase is not equivalent to clinical failure and, thus, we do not treat all patients with that entity [26, 28]. A reliable method is to assess the dynamics in PSA level changes, for example, where the level of undetectable PSA after RP with a subsequent detectable PSA that increases on two or more determinations to PSA > 0.1 ng/mL, we can assume with high probability the biochemical relapse. Another factor that may confound the interpretation of PSA is its production by metastatic microfoci, which may coexist with a recurrence in the postoperative bed. It is important to note that the rate of PSA increase (PSA-DT), which represents PSA kinetics, is also helpful. Regarding PSA-DT in clinical practice, we use a wide range of values as low as PSA-DT < 4 months and a high value of PSA-DT > 12 months [29, 30]. PSA level/PSA-DT with the Gleason score, tumor stage (pT), postoperative margin status, and preoperative PSA level gives more information regarding starting the salvage RT [31, 32].

Many guidelines on prostate cancer assume that biochemical recurrence after RP can be considered when the PSA level exceeds 0.2 ng/mL or 0.4 ng/mL and these values are confirmed by a subsequent test. However, such an arbitrary definition of a biochemical failure after RP in light of modern diagnostic imaging does not seem to meet contemporary challenges. The introduction of new diagnostic imaging methods into clinical practice, so-called “next generation imaging”, such as PET-PSMA, forces us to reconsider the current definition of biochemical recurrence after RP [33–36]. Among other guidelines, those developed by the National Comprehensive Cancer Network (NCCN) recommend defining biochemical recurrence as an undetectable PSA level after RP with a subsequent detectable PSA that increases by 2 or more determinations (PSA recurrence) or increases to PSA > 0.1 ng/mL, or a PSA level > 0.2 ng/mL. The EUA has also adopted the principle that conversion from an undetectable to a detectable PSA level should be considered as a biochemical recurrence [23].

The second crucial step in the decision-making process for patients after RP who experience biochemical recurrence is determining the site of PSA production. Results of several clinical studies showed that the best method for defining the site of abnormal PSA production is PET-PSMA [33–39]. The diagnosis of recurrence by PET examination in the prostate bed is more difficult because it overlaps with excreted radioactive contrast. Generally, according to the latest EUA recommendation, PET/PSMA should not be performed when the PSA level is < 0.2 ng/mL, and in the case of PET-Choline, even when the PSA level is < 1.0 ng/mL, due to the low diagnostic usefulness of both methods at very low PSA levels. Nevertheless, even at a PSA level of around 0.2 ng/mL, in approximately 30% of patients we can detect the site of abnormal PSA production, including regional and/or distant metastases identified by PET-PSMA [40].

One of the challenges we face is to find a PSA level of approximately 0.1 ng/mL, which persists for 4–8 weeks after RP. Such a PSA level is then commonly referred to as “persistent”. When that PSA level is accompanied by pathological risk factors, it is a very poor prognostic factor that indicates rapid disease progression, including distant metastases. However, in the absence of pathological risk factors, such a PSA level may indicate a non-cancerous cause, such as a normal fragment of the prostate gland after surgery. If a “persistent” PSA level is related to cancer, early salvage RT may improve the patient’s prognosis.

The importance of molecular imaging in planning salvage RT

Currently, in a high-risk group of patients, PET-PSMA is replacing the conventional imaging for identifying recurrence lesion as well distant metastases [41, 43]. Standard target in salvage RT is tumor bed whose volume is defined by consensus as a result of analysis of recurrence sites after RP. However, PERYTON-study with PET-PSMA used for defining the CTV boundaries showed better compatibility of targets defined in this way [43, 44]. However, it should be taken into account that PET could show concentration of cancer cells when it contains more than 10⁶ cells. So, it is very likely that PET-PSMA imaging only shows a part of the whole volume of cancer.

What it is important, that outcome of salvage RT is better for patients with negative PET-PSMA scan than for those in whom the examination revealed local or regional lesions [45]. So, according to guidelines for negative PET-PSMA, salvage RT should not be delayed.

We have received more results from clinical studies showing how PET-PSMA imaging can be translated into improvement outcomes of salvage RT. For example, the EMPIRE-1 study is one of RCT, which included patients with biochemical recurrence after RP who, after conventional diagnostic procedures (computed tomography, bone scintigraphy), had no distant metastases or local recurrence [12]. In the next phase of the study, half of the patients were randomly assigned to a PET-fluciclovine scan. In patients whose PET scan revealed active spots (distant, regional, or local metastasis), this information was used to plan further treatment strategies, including modifications of the target definition for salvage RT. The remaining group of patients who did not undergo a PET examination underwent salvage RT based on conventional imaging definition of a target. The so called PET-guided RT resulted in a change of general decision in 35.4% cases and it resulted in difference in 4-year failure-free survival in favor of PET based planning: 75.5% vs. 51.2%. It is important to underline that acute and late GI and GU toxicity rates were similar in both groups. Therefore, we can conclude that conventional imaging have low diagnostic yield, especially at lower PSA levels. Therefore, for salvage RT, molecular imaging has the potential to increase accuracy in the decision-making process, which translates into improved survival. Currently, the next phase of the EMPIRE-1 study, the EMPIRE-2 study (ClinicalTrials.gov registration: NCT03762759) is ongoing where patients are randomly assigned to PET-fluciclovine or PET-PSMA. If PET reveals suspected lesions (metastasis, local recurrence), the dose is escalated for these lesions.

So, in summary for patients whose PSA is above 0.2 ng/mL, a PET-PSMA should be performed before salvage RT for two important reasons. Firstly, if the examination confirms distant metastases, the patient is redirected to systemic treatment ± metastases directed RT. Secondly, when in patients with visible metastases in the regional lymphatic system and/or recurrence in the tumor bed, we can use this information for RT dose escalation for these lesions [46, 47].

Role of genomic testing for postoperative RT

In the era of precision oncology, genomic risk classifiers (GC) is a tool which could help us tailor more precisely the patient’s therapy. One of this GC is Decipher GC, which consists of a 22-gene panel representing multiple biological pathways. Based on patients with localized prostate cancer who participated in RTOG 0126 trial (dose escalation trial) Decipher GC scores stratified patients into three risk categories: low (< 0.45), intermediate (≥ 0.45−< 0.6) and high (> 0.6). High-risk GC category was associated with a 10-fold increased risk of biochemical recurrence as compared to low-risk [48].

In the postoperative RT setting the Decipher GC may indicate whether ADT with salvage RT should be offered. The utility of the Decipher GC was analyzed in the group of post-RP patients participating in the RTOG 9601 study which assessed the impact of combining hormonal therapy with salvage RT [49]. The overall conclusion from the RTOG 9601 study is that combined therapy in patients with a PSA > 0.7 ng/mL translates into survival improvement, whereas this benefit was not observed in patients with a PSA < 0.7 ng/mL. In sub-analysis of the trial conducted by Feng et al., the Decipher GC test identified the subgroup of patients with a PSA < 0.7 ng/mL in whom combined therapy would be more effective than salvage RT alone [50]. In patients with Decipher GC > 0.45 and PSA < 0.7 ng/mL, hormonal added to RT improved OS, PCSS, and distant metastasis-free survival. So, for patients planned for salvage RT with a high-risk Decipher GC, the use of RT combined with ADT is recommended. Currently, it is unclear how useful the Decipher GC is for patients with late salvage RT (PSA > 0.5 ng/mL). In addition, the estimation of optimal ADT duration (6 months vs. 24 months) based on Decipher GC score is currently unknown [51].

In summary, in daily practice we should take into consideration many factors, such as age of patient (life expectancy), comorbidities, clinical and pathologic data, PSA levels, PSADT, and 22-gene GC molecular assay to individualize treatment decision regarding salvage RT ± ADT [52]. Patients with high Decipher GC scores (GC > 0.6) should be strongly considered for the addition of ADT to RT, particularly when the opportunity for early salvage RT has been missed.

Another GC that has been shown promise for prostate cancer patients is the PAM-50 classier, which was adapted from breast cancer. The PAM-50 classier is being tested prospectively in the NRG-GU-006 trial where patients were randomly assigned into the arm with salvage RT plus apalutamide or salvage RT + placebo for patients with rising postoperative PSA (ClinicalTrials.gov identifier: NCT03371719). During the ASTRO 2025 conference, results of NRG GU006/BALANCE trial were presented demonstrating PAM 50 molecular cassifier as the first prospectively validated predictive biomarker guiding ADT for salvage RT [53].

The Post-Operative Radiation Therapy Outcomes Score (PORTOS) is another very promising test that identifies patients in whom dose escalation in salvage RT would translate into improved outcomes. Primary validation of the test was based on patients participating in the RTOG 0126 study and then in the SAKK 09/10 study [52, 54]. There were no significant differences in clinicopathological variables between analysed groups according the dose in RT (standard dose RT vs. escalated dose RT). Study showed that patients with a higher PORTOS and dose escalation achieved a biochemical benefit. It is very important to underline that the identification of patients in whom a therapeutic gain can be obtained as a result of dose escalation in RT is not possible based on a simple assessment of clinicopathological variables. Since dose escalation is related to higher risk of toxicity, PORTOS could identify patients who do not require dose escalation and could be particularly helpful where dose constraints are difficult to fulfill.

In July 2025, the Food and Drug Administration (FDA) approved the new ARTERA AI test, which is both prognostic and predictive in intermediate-risk prostate cancer patients who received definitive RT with or without ADT [55]. This test is based on the analysis of the structure of pathological images from histological specimens of prostate cancer using artificial intelligence (AI). This test identified a subgroup of patients for whom combined treatment (ADT + RT) is beneficial compared to RT alone. Therefore, the information obtained through the ARTERA test allows for the personalization of treatment involving RT and ADT. Currently, we do not have validated data for postoperative RT. However, it is expected that such studies will soon be undertaken for such a large group of patients, providing clinicians with a new tool for personalizing RT.

In the future, it may be possible to combine genomic tests with the results of molecular imaging studies to obtain precise prognostic and predictive tests regarding RT in terms of dose (deescalation vs. escalation), irradiated volume (tumor bed ± whole pelvic), or combination with systemic treatment (no vs. short-term vs. long-term).

Radiotherapy technological improvements in salvage treatment after RP

RT dose escalation in many RCTs showed that it is one of the most significant factors, which for localized cancer translates into better local control. Unfortunately, for patients after RP two RCTs did not show clear benefit of higher dose RT. The first trial SAKK 09/10 included patients with biochemical progression (PSA > 0.1 to 2 ng/mL at randomization) randomized to 64 Gy or 70 Gy to the prostate bed [56]. No patients in this study received ADT or pelvic nodal RT. At a median follow-up of 6 years, study showed no difference in freedom from biochemical progression. Another study tested dose escalation from Chine when patients were randomized to 64 vs. 72 Gy, and also showed no benefit of RT dose escalation [57]. With a higher dose of 70–72 Gy, tolerance doses to organs at risk should be carefully considered because higher toxicity could be expected [58]. There is likely a subgroup of patients who would benefit from a higher dose, but currently we lack a reliable tool to identify this group, other than the mentioned PORTOS test, which should be fully validated. Nevertheless, dose escalation should be considered when imaging studies [e.g., magnetic resonance imaging (MRI) or positron emission tomography (PET)] confirm a recurrence in the postoperative bed or metastasis in the pelvic lymphatic system. The priority is to avoid doses exceeding tolerance levels in critical organs.

Dose fractionation is another very important parameter of prostate cancer RT [59–61]. However, only one RCT (level 1) exists for the use of moderate hypofractionated, and no such evidence for ultrahypofractionated salvage RT [62]. Regarding moderate hypofractionation dose, we are aware of mature results from clinical trials that evaluated this issue, e.g., RADICALS-RT (52.5 Gy/20 fractions vs. 64 Gy/32 fractions) [63–66]. Only one study, GU 003, evaluated patients reported outcome (PRO) to compare the conventional fractionation to hypofractionation with dose fraction of 2.5 Gy. In conclusion of this study, RPO is the same for both arms. So, currently according to ASTRO guidelines, hypofractionation could be used but with marginal recommendation.

Stereotactic body RT (SBRT) using ultra-hypofractionated dose fractions is a standard of care for localized cancer. However, data on ultra-hypofractionated adjuvant or salvage RT using SBRT are scarce including small phase I or retrospective studies [67, 68]. Potential severe acute and late/very late toxicities are of major concern, especially toxicity from the vesicourethral anastomosis and bladder. The latest results of the phase 2 study (SCIMITAR) conducted by Casillas and Nikitas et al. showed that SBRT (30–34 Gy in 5 fractions) used as salvage RT after RP was well tolerated, with no measurable decline in urinary or bowel patients reported outcomes (PRO) with median follow-up of 53 months compared with conventional fractionated RT (historical phase III cohort, n = 742 patients) [69, 70].

Technological advancement in RT which has potential to improve treatment outcomes is adaptive RT (ART). ART utilizes the patient’s daily acquired images, such as computed tomography (CT), Con- BeamCT, and MRI, for recontouring and replanning of RT dose distribution. Generally, ART can be applied between individual RT fractions (offline), immediately before the fraction is delivered (online), and during the fraction in real time (real-time, online during the session).

Thanks to technological improvement in the RT device constructions and the imaging methods, such as CT and MRI, and their integration with the RT planning process, it is becoming possible to deliver the dose to the target with increasing accuracy. This is particularly important for the pelvis region, which contains structures such as the postoperative bed volume, urinary bladder, and large bowel, which can move and change the volume over time. Unfortunately, their geometric relationships and volume change very rapidly, significantly altering the dose distribution in an unpredictable manner. This could lead to dose increases in critical organs and, on the one hand, the dose reduction in the postoperative bed (target). However, introducing ART into daily clinical practice presents a significant logistic challenge. One of these challenges is to ensure a reasonable time window, ideally within minutes, for preparing a new, revised RT plan.

Studies on various tumor locations have demonstrated that ART can enhance the accuracy of prescription dose coverage for the target volume while reducing radiation exposure to normal tissues [71, 72].

However, ART is still in its infancy compared to many other RT technological advances, such as intensity-modulated RT, image-guided RT, and stereotactic body radiation therapy. Although clinical studies are emerging, most of the data suggesting a possible benefit to ART come from preclinical or early-phase studies. However, to date, we do not have the results of prospective clinical trials on how ART for salvage RT in prostate cancer can translate into improved treatment outcomes (survival, quality of life).

How to combine ADT with salvage RT

In general, for high-risk and locally advanced prostate cancer numerous RCTs showed that ADT combined with RT translated into benefit in PCSS and OS. The clinical situation is different for postoperative patients because we irradiate microscopic disease in tumor bed and/or pelvic. However, salvage RT has demonstrated high loco-regional efficacy but, unfortunately, relapse is observed in half of the patients. That’s why, there has been considerable interest in combining ADT with salvage RT to improve outcome, especially for those patients with more pathologically adverse features.

So far, we have results from 4 RCTs which compared salvage RT alone to salvage RT combined with hormonal therapy. One of the oldest studies is RTOG 9601 trial where investigators compared the efficacy of salvage RT combined with the long-term (2 years) treatment with bicalutamide (150 mg daily) to salvage RT alone [73]. Patients in this study had pathological risk factors (pT3 or pT2R1) and PSA ranging from 0.2 ng/mL to 4.0 ng/mL (median 0.6 ng/mL). RT was given to tumor bed (without pelvic lymph nodes) to total dose of 64.8 Gy. In conclusion, after 18 years of follow-up biochemical failure was 70.7% in the RT alone arm vs. 48.9% in the RT+ hormonal therapy arm, overall survival was 52.9% for the combined therapy (RT + HT) and 43.1% for RT. Moreover, the 18-year prostate cancer-related death rate was 17.8% for the combined group and 27.6% for the RT. The centrally reviewed non-prostate cancer deaths showed no difference between the two arms. Important message from the study is that for patients with a Gleason score of 8–10, the 18-year rate of distant metastases was 24.3% in the combined arm vs. 53.6% in the RT. In conclusion, authors underlined that the biggest benefit of combined therapy is noted in subgroups of patients with more aggressive cancer (higher PSA level prior RT or Gleason score of 8–10). Currently, bicalutamide is replaced by luteinizing hormone-releasing hormone (LHRH) analogues or LHRH antagonists. Furthermore, the current doses of RT are higher, ranging from 66 to 70 Gy.

Another very important trial which evaluated the efficacy of ADT with or without whole pelvic radiotherapy was RTOG 0534-SPPORT. The study involved the following 3 arms: Arm 1: tumor prostate bed RT (PBRT), Arm 2: 4–6 months of short-term ADT (STADT) + PBRT, and Arm 3: whole pelvic lymph node RT (PLNRT) + STADT + PBRT. Primary analysis showed benefit of adding PLNRT and STADT to PBRT. In the RTOG 0534 trial investigators evaluated the role of short-term ADT with RT of tumor bed or pelvic RT. The RTOG 0534 study showed the benefits of short-term ADT combined with PBRT. The endpoints involving OS and distant metastatic survival have so far not been statistically different among arms with short ADT, but when Whole pelvic RT was combined with ADT the benefit in distant metastases-free survival and PCSS was observed [74].

The third trial, GETUG 16, randomized patients with a median PSA of 0.3 ng/mL to salvage RT to the prostate bed with or without 6 months of ADT. At a median follow up of 9.3 years, the progression free survival was superior in patients receiving ADT. Additionally, there was a statistically significant benefit of combined therapy for metastasis free survival but there was no benefit in OS [75].

In the next trial, the RADICALS-HD 1553 patients were randomized to no ADT vs. 6 months of ADT vs. 24 months of ADT. So far, early results at a median follow-up of 8.9 years for the 6 vs. 24 months comparison showed an improvement in metastasis free survival for the long-term ADT arm, but no OS benefit [76]. Results for the short-term ADT arm showed no benefit, so according to the investigators of this study there is no sufficient evidence for routine recommendation of ADT.

The planned meta-analysis of the above mentioned studies (RTOG 0534 trial, GETUG-AFU 16), the RADICALS-HD) will clarify this problem. In addition, we are waiting for results from ongoing RCTs that evaluated new systemic therapies combined with salvage RT, such as the NRG Oncology GU 002 (docetaxel), GU 006 (apalutamide), GU 008 (abiraterone + apalutimide), and RTOG 3506 (enzalutamide) trials [77–80].

Another very important issue discussed in salvage RT is the optimal volume which should be included in RT. For very high-risk (pT3b, Gleason 8–10, pN1) features the pelvic nodes should be encompassed for salvage RT. Another issue related with the pelvic RT is the upper level of lymph nodes which should be included. More recently, the NRG Oncology International Consensus Atlas recommended the superior border at the bifurcation of the aorta into the common iliac arteries or the proximal inferior vena cava to the common iliac veins, whichever occurs more superiorly (typically at the level of L4–L5). The frequency of lymph nodes metastases increases after template nodal dissection with these interventions disrupting the normal lymphatic network and allowing the onset of different routes of spread.

Summary and key points

Early adjuvant RT should be considered for patients with pT3 and ISUP group 4–5 ± positive margins.

Generally, early Salvage RT (PSA > 0.1 ng/mL and 0.2 ng/mL) is preferred over Late Salvage RT (> 0.5 ng/mL) or Early Adjuvant RT. Salvage RT combined with ADT should be offered for following situations: pN1, persistent PSA (PSA > 0.1 ng/mL), ISUP group 4–5, pT3b, PSA > 0.6 ng/mL, PSA-DT < 6 months.

Patients with negative PET-PSMA scan do better with salvage RT than observation and it reflects a strong recommendation of guidelines that a negative PSMA PET/CT should not delay salvage RT.

The optimal duration of ADT (6–24 months) combined with salvage RT is unknown but longer therapy is advised for higher risk groups. When ADT is combined with salvage RT, also pelvic elective RT should be considered.