Prostate cancer (PCa) is the most commonly diagnosed cancer in men and the second most common cancer worldwide (1). Despite the fact that more than 70% of cases of localized PCa are cured by local treatments [brachytherapy, (BT), surgery and/or external beam radiation therapy (EBRT)], or are under active surveillance before receiving treatment without altering the benefit of the latter, the median survival of metastatic patients is still less than 5 years (2). Oligometastatic disease (OMD) is first defined by Hellman and Weichselbaum as an intermediate state between local and systemic disease, but there are no validated biomarkers. The ESTRO-ASTRO consensus notes that there are currently no clinical studies showing a lack of benefit beyond a certain number of metastases to define OMD. It is thus a disease where all metastatic sites are treatable by radiotherapy with curative intent. Moreover, OMD can be split in two subtypes: i) synchronous when OMD is detected at the time of the initial diagnosis, ii) metachronous or oligo-recurrent, when OMD is discovered during the course of the disease. Biologically, all metastases are synchronous but our ability to detect them makes them metachronous. Furthermore, metachronous metastases are known to have a better prognosis than synchronous metastases. Elimination of oligometastatic burden by radiation may prevent additional metastatic spread and improve overall survival (3–5). This approach may change the paradigm from palliative to potentially curable disease for oligometastatic PCa patients (6, 7). Indeed, the phase 2 SABR-COMET (Stereotactic ABlative Radiotherapy for the Comprehensive Treatment of OligoMETastases) trial recently evaluated the value of ablative stereotactic radiotherapy to metastatic sites in patients with 1 to 5 metastases (6). There was an overall survival benefit of 13 months (41 vs 28) (Hazard Ratio (HR) 0.57, 95% CI 0.3 - 1.1; p= 0.09) in favor of radiotherapy to all sites. Similarly, there was a randomized phase 2 study observation vs stereotactic ablative radiation for oligometastatic prostate cancer (ORIOLE) in which 54 patients with recurrent hormone-sensitive oligometastatic PCa were randomly assigned to stereotactic ablative radiotherapy (SABR) or observation in 2:1 ratio (8). At 6 months, disease progression was reported in 19% of patients receiving SABR versus 61% of patients undergoing observation (P=.005). The disease progression rate was 11% vs 50% (P=.005) and median progression-free survival was not reached vs 5.8 months (HR, 0.30; P=.002). Given progress in knowledge and treatments that allow some metastatic patients to be treated with a curative rather than a palliative objective, the concept of oligometastasis is also evolving. The European Society for Radiotherapy and Oncology (ESTRO) and European Organization for Research and Treatment of Cancer (EORTC) consensus seems to relate oligometastases less and less to their number and more to the possibilities of their treatment in terms of technical barriers to the volume and location of metastases (9).

Over the past decade, immunotherapy has revolutionized the treatment of metastatic cancer but has shown only modest efficacy in PCa patients. Nevertheless, recent advances in molecular diagnostics and understanding of immune mechanisms promise to improve the efficacy of immunotherapy in PCa as well. Immunomodulation induced by radiotherapy is a topic of current interest. Indeed, radiotherapy can promote immunogenic cell death and induce the immune response by enhancing antigen cross-presentation and CD8+ cytotoxic T cell response. However, radiation also enhances an immunosuppressive microenvironment by promoting myeloid cells infiltration and macrophage polarization toward an M2-like phenotype, as well as an increase of regulatory T cell subsets involved in the inhibition of naive T cell proliferation and activation (10). Therefore, the combination of radiotherapy and immunotherapy may induce synergistic effects to cure PCa (11). This review aims to highlight the advances in PCa physiopathology and summarize the state-of-the-art knowledge of radiotherapy and immunotherapy in oligo-recurrent PCa.

Induction of immune tolerance is a key process throughout tumor development to metastasis. Basically, tumor antigens, neo or not, must be processed and presented by antigen-presenting cells (APCs) such as dendritic cells (DCs). They then migrate to secondary lymphoid organs to activate specific T lymphocyte (T) cells. Conventional CD8α+ DCs appear to be critical APCs for cross-presentation of neoantigens for tumor rejection by T cells (12). Activation of APCs occurs in coordination with other innate immune cells, including natural killer (NK) cells, natural killer T (NKT) cells and γδ T cells in response to damage-associated molecular patterns (DAMPs).

Immunologically, tumors are classified as hot and cold tumors according to their immune infiltrate. Features of hot tumors include increased T cell and cytotoxic T lymphocyte (cTL) infiltration, primarily due to a high tumor mutational burden (TMB), and increased proteins that activate checkpoint proteins. In contrast, features of cold tumors include exhausted cTL cells in the tumor or their absence at the tumor margins, the presence of tumor-associated macrophages (TAM) polarized to an M2-like phenotype (pro-tumor), a low mutational load and poor antigen presentation. PCa can be considered as an immunologically cold tumor (13).

Cancer progression and response to immunotherapy may be strongly influenced by the tumor microenvironment (TME), including immune cells (14). In PCa, tumor-associated antigens (TAAs) are expressed both in normal and tumor cells, but at higher levels in cancer cells. These TAAs are, for example, prostate-specific antigen (PSA), prostate-specific membrane antigen (PSMA), prostatic acid phosphatase (PAP) or CD155. Nevertheless, no anti-tumor response can be triggered due to the immunosuppressive TME (15). Indeed, a lower density of immune cells has been observed in prostatic adenocarcinomas compared to benign nodular hyperplasia of the prostate (16). Anti-tumor CD8+ T cells are also suppressed by the depletion of arginase and tryptophan from the TME after upregulation of secretion of nitrous oxide synthase and indoleamine 2,3-dioxygenase (IDO) by myeloid-derived suppressor cells (MDSCs) (16), or by the presence of a large amount of regulatory T cells (Tregs) compared to other cancers (17), and other immunosuppressive cells such as M2 TAM or neutrophils, both associated with poor survival (18). This immunosuppressive environment is promoted by specific factors such as TGF-β (19) and CXCR2 (20) secreted under the TME. Then, inhibition of CXCR2 may be interesting to improve immunotherapy as tested in a current clinical trial (NCT03473925).

Immune checkpoint inhibitors (ICIs) are antibodies designed to activate an effective immune response by targeting negative regulators of T cells such as PD-L1, PD-1 or CTLA-4 (48) ( Figure 2 ). The use of a CTLA-4-targeted monotherapy, known as ipilimumab, was tested in PCa in an unselected population, but did not result in significant benefit (49). This could be explained by increased expression of PD-1/PD-L1 as a compensatory mechanism that maintains inhibition of the T cell response (50). Alternatively, an anti-PD-1, pembrolizumab, has been commercialized for mCRPC with mismatch repair deficiency and/or microsatellite instability (51), although the relevance of this ICI is still debated. Indeed, the Keynote 19 trial demonstrated that pembrolizumab monotherapy induced antitumor activity in only a small number of mCRPC patients, with an objective response rate (ORR) up to 5% (52). However, in a phase 2 trial (Checkmate 650), double-blockade immunotherapy with nivolumab and ipilimumab showed an ORR of 26% in asymptomatic or minimally symptomatic patients with mCRPC (26). Therefore, determining the subpopulations that might benefit from ICIs’ immunotherapy appears essential.

A promising new approach uses bispecific antibodies (bsAb), also known as bispecific T cell engager (BiTE^®), to mobilize T cells against tumor cells (53) ( Figure 2 ). BsAbs are designed to recognize a TAA with their target arm and to stimulate the T cell receptor (TCR)/CD3 complex with their effector arm (54). Once BsAbs target tumor cells and activate T cells, they induce T cell proliferation and production of cytokines, perforins and granzymes, thereby killing surrounding tumor cells. Various combinations of bispecific conjugates have been tested in PCa. One of the most studied combinations uses PSMA, a specific-prostate antigen whose expression increases with disease progression, and an anti-CD3. Preclinical data demonstrated its efficacy to induce an anti-tumor CD8 T cell response in vitro, ex vivo and in vivo (55, 56). Several clinical trials with PSMA-targeted T cell engagers (NCT03792841; NCT03577028; NCT03926013) are currently running, particularly for patients with mCRPC (NCT04104607). Furthermore, other TAAs, such as A-disintegrin and metalloprotease 17 (ADAM17) (57) and CD155 (58), are being evaluated as targets of bsAbs in PCa. Interestingly, such therapy may not need to be personalized for each patient.

Chimeric antigen receptor (CAR) T cells are also an interesting tool to fight PCa ( Figure 2 ). These therapies are based on re-engineering patients’ T cells to express a TCR directed against a specific tumor antigen. Cells are then expanded and reinjected into the patient to selectively destroy target cells harboring the epitope of interest. Several studies are currently underway with CAR T cells directed specifically against PSMA (NCT04053062; NCT03089203; NCT03873805). Initial results for NCT03089203 demonstrated that adoptive cell transfer of CAR-PSMA-TGFβRdn is safe and feasible in mCRPC patients (59).

Vaccine-based therapies are also being developed to treat PCa ( Figure 2 ). In 2010, the Food and Drug Administration approved Sipuleucel-T for the treatment of castration-resistant PCa. Sipuleucel-T is a vaccine based on the transfer of autologous DC to cross-present prostatic acid phosphatase (PAP), a specific prostate antigen, to T cells and active adaptive immune cells (60). Another vaccine therapy evaluated in PCa is PROSTVAC, which uses a genetically engineered poxvirus encoding prostate specific antigen (PSA) to generate a T cell response. It also contained three co-stimulatory molecules: CD80, intercellular adhesion molecule 1 (ICAM-1) and lymphocyte function-associated antigen 3 (LFA-3). However, a phase III study showed no effect on overall survival in mCRPC (61). Combination therapy with a monoclonal antibody directed against PD-L1 and a recombinant vaccine of Avipoxvirus is currently in use (NCT03315871).

A new class of immunotherapy is oncolytic viruses (62, 63) ( Figure 2 ). Oncolytic viruses selectively replicate in tumor cells, and induce an immunogenic cell death. The subsequent release of TAAs is thought to trigger an anti-tumor immune response by recruiting DCs and activating T cells. A recent study by Zafar et al. indicated that oncolytic adenoviruses expressing CD40L (Ad3-hTERT-CMV-hCD40L) can effectively stimulate DCs in the immunosuppressive microenvironment of PCa (64). Another genetically engineered oncolytic virus, Ad5-yCD/mutTKSR39rep-hIL12 shows promising effects in preclinical model of PCa through the enhancement of anti-tumor response by cytotoxic immune cells (NK and cTL) (65). In a phase I clinical trial (NCT02555397), improvement in local and metastatic tumor control resulted in significant prolongation of survival.

Extracellular vesicles and microRNA-based therapies represent future perspectives in immunotherapy (66, 67) ( Figure 2 ), as they are involved in immunomodulation and tumor progression (68, 69). Exosomes are small extracellular vesicles (50-150nm) formed inside cells (70) and secreted by almost all cell types, including tumor cells. They appear as an interesting tool in cancer immunotherapy due to their low immunogenicity and toxicity (71). However, caution should be taken when targeting exosomes as they are involved in many physiological pathways. They play a role in intercellular communication through a specific interaction between transmembrane proteins of exosomes and receptors on the plasma membrane of recipient cells, and influence physiological and pathological functions in the recipient cell. In PCa, exosomes from cell lines expressing various regulatory proteins such as FAS ligand (FASL) or PD-L1, lead to suppression of T or NK cell responses (19, 72, 73).

Of note, exosomes are naturally enriched with non-coding RNAs such as microRNAs (miRNAs) (74), which are readily transferred to recipient cells (75). miRNAs are a subset of small non-coding RNAs with a length of 19 to 22 nucleotides that regulate gene expression at the post-transcriptional level by translational repression or degradation of the target mRNA. miRNAs included into tumor exosomes (TEX) can participate in tumor immune escape by reducing the CD8+ T cell response (69). Thus, the use of engineered EVs containing miRNA mimic or miRNA antagonists may be a promising therapy to enhance the anti-tumoral response (76–78).

Irradiation induces immunogenic cell death leading to the release of tumor antigens, including miRNA patterns (79) or DNA breaks. Radiation therapy also affects the TME, inducing immune cell recruitment and vascular changes ( Figure 3 ). Of interest, irradiation of the vasculature may promote infiltration of immune-inflammatory cells [reviewed in (80)]. Conventional 2 Gy dose fractions, a single large dose fraction, or high dose hypofractionated radiotherapy are effective in tumor control. Recent preclinical studies have shown that tumor-resident T cells may be relatively radioresistant and can be amplified to control irradiated tumors (81). The main question remains to determine the dose or fractionation regimen that can transform an immunocompromised tumor into a highly immunogenic one (11). In that regard, stereotactic body radiation therapies (SBRT) can be divided into three categories based on their effects on the immune system or the TME: immunogenic ablative (15-35 Gy fractions), immunomodulatory sub-ablative (8-12Gy), and modulatory low-dose fractions (≈ 2 Gy). Ablative doses lead to profound cell death with concomitant depletion of radioresistant immune suppressor cells in the TME. They may also increase levels of fibrosis and chronic inflammatory/immunosuppressive pathways. However, high ablative fractionation is not considered due to normal tissue tolerance (82). Hypofractionated radiotherapy is considered the most suitable with the goal of immunomodulation, whereas ablative or sub-ablative doses remain more controversial. Preclinical data on fractionation showed that immunomodulatory fractionation of 3 x 8 Gy was more effective than a single ablative dose of 20-30 Gy (83). While modulatory doses (e.g. three x 8 Gy) can produce similar effects to standard fractionation, they resulted in a strong type I interferon (IFN) response (84, 85). On the other hand, low doses of irradiation also have profound effects, leading to remodeling of vessels, reprogramming of macrophages or increased lymphocyte infiltration (86).

Recently, a prospective observational study compared the effects of internal irradiation BT followed by EBRT (15 Gy high-dose rate BT, followed 2 weeks later by 46 Gy in 23 fractions the entire pelvis) with EBRT alone (46 Gy in 23 fractions to the entire pelvis followed by 32 Gy in 16 fractions to the prostate) on immunological cells in PCa patients with Gleason score 9 (87). An enhancement of cTL response was observed in patients receiving BT + EBRT compared to patients receiving EBRT alone, which was associated with IL-2 and granzyme B secretion. In addition, a reduction in CD4+ T cells was observed 3 months after treatment. The authors observed an increase in PD-1 expression by CD4+ and CD8+ cells following radiotherapy. Thus, it might be useful to combine anti-PD-1 checkpoint inhibitors with BT/EBRT to obtain a reliable immunological response.

In the 1940s, prostate cancers (PCa) were found to have a dependence on androgens. This discovery led to the approach to treat PCa using androgen deprivation therapy (ADT) (98). Moreover, the addition of RT to ADT appears to improve outcomes by enhancing both local and distant disease control (99). Mechanisms of synergy are partially understood, but are likely mediated by the androgen receptor (AR) (100). The AR is a nuclear hormone receptor activated by engagement of its ligands, testosterone and dihydrotestosterone (DHT). Ligand binding exposes the AR in the nucleus, the receptor dimerizes and binds to androgen response elements in the promoter regions of target genes like the PSA (101). Additional co-regulatory proteins are recruited to allow transcription, leading to downstream cellular responses such as growth and survival (102). Thus, androgen ablation therapies repress transcription of AR target genes, which causes activation of tumor cell apoptosis and the eradication of most of the androgen-dependent cancer cells (103) Thus, inhibiting the tumor cell’s ability to repair double-stranded DNA damage by ADT can act as a “radiosensitizer” (104). Combined treatment also induces permanent cell cycle arrest or apoptosis (105). Also, ADT reduces intraprostatic hypoxia which is an important risk factor for poor locoregional disease control and biochemical failure after RT (106, 107)

Finally, enhanced immune responses have also been reported after the association between ADT and RT (108). In addition, change has been also observed in ADT-treated mice CRPC following RT with more TILs associated and an attenuated MDSC recruitment (109). In fact, RT promotes T cell priming by the release of tumor antigens and pro-inflammatory soluble mediators. On the other hand, ADT promotes lymphopoiesis, immune cell trafficking and tumor infiltration (110). Associating immunotherapy to this combination may enhance these processes. Also, there is the question about the precisely timing of immune modulation and depends on many factors, such as the type of ADT, the RT strategy used as a drug (11). In a clinical report, ADT promoted strong adaptive anti-tumor T- and B-cell responses; however, peripheral TH1 and TH17 effector memory subsets decreased after 2 years of treatment (111).

Both preclinical and clinical data showed that androgen-depriving therapy (ADT) synergizes with prostate cancer radiotherapy (100, 112). The impact of testosterone on tumor immune response is ambiguous (113). On the one hand, hormone-naive prostate cancer may respond better to immunotherapy than castration-resistant prostate cancer. In mice, orchiectomy synergizes with immunotherapy, whereas androgen receptor (AR) antagonists suppress the effects of immunotherapy by impairing the adaptive immune responses through interference with initial T cell priming (114). On the other hand, ADT induces T cell infiltration of the prostate (115). Neoadjuvant ADT promotes immune infiltration with proinflammatory effects, but the anti-tumor cells (CD8+ T) are counterbalanced by local pro-tumor cells (TAMs and T reg) (116). ADT also does not increase or diminish PDL1 expression (117). Following ADT and vaccination, prostate cancer-specific T cells expand and develop effector functions (118), suggesting that neoadjuvant ADT may increase the efficacy of immunotherapy. Indeed, the androgen receptor antagonist, enzalutamide, has been tested in phase 2 in combination with immunotherapy and has shown interesting results with a 20% objective response rate (ORR) in patients with mCRPC treated with abiraterone naive chemotherapy (119).

There is growing evidence that combination of immunotherapy and radiotherapy is a promising strategy to achieve overall survival benefits for patients. Radiation therapy of the primary tumor and/or metastases in combination with immunotherapy increases overall survival in preclinical models of prostate cancer. However, despite clinical evidence of increased immune response, clinical studies have failed to show improved survival following combined immunotherapy and radiotherapy. It appears essential to better understand the mechanisms of metastases and notably the communication between tumor cells and immune cells. These may open up the development of new therapeutic approaches.

SS initiated the study. LO and ML performed the scientific literature search and wrote the manuscript. LO designed the tables. ML designed the figures. DF, VP and SS supervised, helped to revise and edit the manuscript. All authors contributed to the article and approved the submitted version.

ML was supported by a regional grant (EpiSAVMEN), and by Ms. Suzanne Saulnier. ML and DF were recipients of grant “Ligue contre le Cancer, comités 16, 22, 29, 35, 44, 49 and 56”. ML and DF were also recipients of “Cancéropôle Grand Ouest AO Structurant ExomiR”. SS and VP were recipients of grant “Ligue contre le cancer”, transports Loizeau, De Graet consulting, Astra- Zeneca and Ms. Suzanne Saulnier. This study received funding from Astra Zeneca. The funder was not involved in the study design, collection, analysis, interpretation of data, the writing of this article or the decision to submit it for publication. LO was supported by an institutional grant from CHRU Brest. The POSTCARD F-GETUG P13 study received funding from Astra Zeneca.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.