Prognostic value of inflammation in prostate cancer progression and response to therapeutic: a critical review

The prostate is an immune-competent organ and is normally populated by a small number of inflammatory cells [1]. In particular regulatory T-cells are mainly located in the fibromuscular stroma and in the peri-glandular area [2]. The origin of prostatic inflammation can be multifactorial, including bacterial infections, viruses, dietary factors, hormones, autoimmune responses and urine reflux [1, 3].

The persistence of some of these factors can constitute to chronic prostatic inflammation. Prostate cancer (PC) is now the most common cancer among men. The role of infection or inflammation is sustained in different cancer sites and also in PC. Epidemiologic, histo-pathologic and molecular pathologic studies provide the emerging evidence of the possible role of prostatic inflammation as a factor involved in PC initiation and/or progression [4‐6]. Inflammation can play a role in PC carcinogenesis through several mechanisms: causing cellular and genomic damage; promoting cellular turnover; creating a tissue microenvironment inducing cell replication, angiogenesis and tissue repair [4, 7].

The high prevalence of chronic inflammation in pathological samples of prostate tissue from surgery or prostate biopsy has sustained a possible link between chronic inflammation and PC [8, 9]. Pro-inflammatory cytokines, inflammatory mediators and growth factors related to inflammation can determine uncontrolled proliferative response, rapidly dividing cells more likely to undergo mutation, as observed in cancer [7, 10]. Several PC susceptibility genes such as RNASEL, MSR1,MIC1 involved in PC carcinogenesis, encode proteins with functions in response to inflammation, infection and oxidative stress: their mutation may reduce the possibility of preventing carcinogenesis through these pathways [4, 11]. Some data show that chronic inflammation can induce proliferative events and post-translational DNA modifications in prostate tissue also through oxidative stress [8, 12]. Repeated tissue damage and oxidative stress related to this event may produce compensatory cellular proliferation with the risk of hyperplastic growth or neoplastic transformation [13]. It is accepted that prostatic inflammation can generate free radicals such as nitric oxide that can be converted by cyclo-oxigenase enzymes to different eicosanoids and prostaglandins that have been recognized as regulator of prostate cell proliferation [14]. Normally glutathione transferase activity defends prostate cells against the genomic damage induced by oxidants isolated at sites of inflammation [8]. The methylation of this gene produce the loss of his protective activity and it could be implicated in the translation from inflammation to pre-neoplastic lesions such as high grade prostatic intraepithelial neoplasia (PIN) and therefore to PC [14].

As clinical data, a 5-year follow-up study based on prostate biopsies established that chronic inflammation accounts for nearly 20% PC development [15, 16]. Moreover, inflammatory atrophy was found in about 40% of PC cores in that study. Observations from many lines of research indicate the existence of role of inflammation in prostate carcinogenesis and progression [17].

The aim of the present review is to critically analyze the role of prostatic inflammation as a prognostic factor for the progression and aggressiveness of PC. In particular, we did not analyze the relationship between inflammation and PC development or incidence but, we verified data on the association between inflammation and PC aggressiveness, or PC response to therapies.

Our aim is to analyze whether, in cases with PC diagnosis, an association with an inflammatory status may determine a poorer prognosis, higher aggressiveness of the tumor and lower response to therapies. We also try to verify a possible role of prostatic inflammation as a prognostic factor in PC patients, for tumor progression despite surgical, radiotherapic or medical therapies.

We searched in the Medline and Cochrane Library databases (primary fields: prostate neoplasm and inflammation; secondary fields: aggressiveness, Gleason score, prognostic, radical prostatectomy, radiation therapy, androgen deprivation therapy, CRPC without language restriction from the literature of the last 10 years. We included original articles, clinical trials conducted in humans and reviews.

Our review was divided in different sections analyzing the following items: relationship between inflammation and PC aggressiveness; influence of inflammation on the response to surgery, radiotherapy, androgen deprivation therapies; role of inflammation in CRPC. At the end of each section we critically analyzed data in terms of a possible prognostic role of prostatic inflammation in these specific settings.

Different clinical and pathological parameters and classifications are used to determine PC aggressiveness and risk of progression. In particular the Gleason grading is used in conjunction with other clinical and molecular variables, including serum prostate specific antigen (PSA) levels and clinical stage, to classify PC in risk classes of progression and to help make treatment decisions for PC.

PC is one of the most common causes of cancer-related deaths among males; however a significant percentage of diagnosed PC have a relatively indolent form that is unlikely to progress with time. A subset of PC, on the contrary, exhibit aggressive properties, such as rapid proliferation, metastatic spread, resistance to therapies. To obtain a correct management of PC, the major challenge remains to distinguish patients who need definitive treatments from cases who have a latent tumor.

Gurel et al. [18], histologically examined prostate biopsy samples for inflammation using the United States National Institute of Health consensus grading system. Cases with inflammation in ≥ 1 biopsy core were 1.78 times (95% CI 1.04–3.06) more likely to develop PC in the core examined in the biopsy than those who had no evidence of inflammation.

This association was more pronounced for high grade tumors (Gleason sum 7–10; OR 2.24, 95% CI 1.06–4.71) than for low grade cases (Gleason sum 6; OR 1.57,95% CI 0.83–300) [18].

From the placebo arm of the Prostate Cancer Prevention Trial, Gurel et al. [18] showed a significant association between the presence and extent of chronic inflammation in prostate biopsy cores and high grade PC (Gleason sum 7–10; OR-2.24;95% CI 1.06–4.71) but not in low grade cases (Gleason sum 6; OR 1.57; 95% CI 0.83–3.0).

As adaptive immune cells, analysis of two major prostate tumor-infiltrating lymphocyte subsets showed that capsular and perineural invasion as well as biochemical progression was related to strong infiltration of T and B lymphocytes [15, 19]. Mc Ardle et al. [20] showed that the presence of increase CD4+ T-lymphocyte infiltrate was associated with poor PC prognosis and it can predict survival.

Ammirate et al. [21] found that PC progression and metastasis is associated with inflammatory infiltration of macrophages and lymphocytes and activation of I kappaB kinase (IKK) by an NF-kappaB independent mechanism.

Davidsson et al. [22] in a case control study on cases diagnosed for PC, showed that men with greater numbers of CD4+ regulatory lymphocytes in their PC environment have an increased risk of dying of PC and this identification may predict clinically relevant tumors.

Although the role of B-lymphocytes in cancer has been overshadowed by the interest in T-cell responses, B-cells can play a complementary role in the immune response against tumor [15]. Gannon et al. [23] compared metastatic and non-metastatic pelvic lymph nodes from PC cases and controls. They found a decreased presence of CD20+ B-lymphocytes and CD38+ lymphocytes in metastatic lymph nodes comparatively to control lymphnodes.

Recently neutrophil-to-lymphocyte ratio (NLR), an inflammatory parameter, has been reported to be of prognostic value in some solid tumors including PC [24]. Yin et al. [24] conducted a meta-analysis to obtain a reliable assessment of the prognostic significance of NLR calculated from the white blood cell differentiated counts in PC patients. A total of 14 studies were included and they showed that elevated NLR was not significantly associated with the poor overall survival (OS) (HR 1.45; 95% CI 0.77–2.71; p = 0.248) or recurrence-free survival (HR = 1.34; 95% CI 0.89–2.02; p = 0.155) of patients with localized PC. On the contrary elevated NLR predicted poorer OS (HR 1.57, 95% CI 1.41–1.74; p < 0.001) and progression-free survival (HR = 1.97;95% CI 1.28–3.04; p = 0.002) in metastatic PC. This finding was also confirmed in a second meta-analysis on 9418 PC patients from Tang et al. [25]. In the majority of studies, NLR was not significantly associated with Gleason grading in PC patients (p = 0.573) [25, 26].

Various innate immune cells have been involved in PC progression. Macrophages are a source of secreted pro-inflammatory cytokines and they are distinguished in two phenotypes: M1 with tumor inhibitor properties and M2 with tumor stimulatory properties [15, 27, 28]. Lanciotti et al. [28] showed that a high density of macrophages in PC is associated with more aggressiveness and poorer prognosis. In particular M2 phenotype was more presented in Gleason score 8–10 and pT3a stage. As showed by Fang et al. [29] in co-cultured prostate epithelial cells, M2 induction of aggressiveness involved the alteration of signaling of macrophage, androgen receptor (AR), inflammatory chemokine CCL4-STATE3 activation, down regulation of p53/PTEN tumor suppressors.

Besides immune cells, chemokine receptors have been shown highly expressed on tumor cells in comparison to normal glands [15] and evidences have been provided that chemokines are involved in the metastatization and progression of different tumors. In vivo based studies on chemokines in PC have been focused on CXCL8, CXCL12 and CCL2 [30‐32]. Chemokines can facilitate tumor progression and dissemination either by shaping the functional profile of infiltrating lymphocytes or by activating stromal and neoplastic cells [15]. Changes in chemokine receptors may be necessary to activate tumor progressing signals. For example, more aggressive PC cells express higher levels of CCR2 compared with less aggressive cells and they increase also in metastatic compared to localized PC [33]. Moreover increased serum levels of CXCL8 and CCL2 have been associated as possible prognostic markers of a high grade and progressive PC [34, 35].

Multiple inflammatory cytokines have been identified as potential mediators between prostatic inflammation and PC risk and aggressiveness [36]. One of these is macrophage inhibitory cytokine 1 (MIC-1), a member of the transforming growth factor beta family. Some evidence suggested an up regulation of MIC-1 related to PC aggressiveness and that circulating levels of MIC-1 can predict poor prognosis for PC [37]. IL-6 is a multifunctional cytokine which is expressed in most PC cell lines and the expression is at higher levels in those which do not express the androgen receptor (AR) and show an enhanced malignant potential [38]. Aberrant IL-6 signal transducer and activator of transcription-3 (STAT3) signalling and loss of p53 occur during PC progression to metastatic disease [39].

Neveu et al. [40], in primary prostate cell culture from patients submitted to radical prostatectomy, showed that the baseline IL-8 secretion level was associated with PC aggressiveness and the median baseline IL-8 levels were significantly lower in cases with non-aggressive PC compared to those having aggressive PC (p = 00005). In particular the Gleason score significantly increased with the basal level of IL-8 (rho =0.63; p = 0.0007).

Some studies reported on the association between one or multiple single nucleotide polymorphism (SNP) in inflammation-related pathways and prostate cancer risk and aggressiveness [36, 41]. Zabaleta et al. [42] evaluated 15 SNP in five cytokine genes (IL-1B,IL-10, TNF alfa, IL-6 and IL-8) in relation to risk of aggressive PC (Gleason score ≥ 8 and PSA > 20 ng/ml) and they found an association between aggressive PC and an IL-8 (IL-8-47CT) genotype (OR = 3.5;95% CI 1.13–10.88), as well as an increased risk with combined genotypes in IL-1B and IL-10 (OR = 3.38;95% CI 1.70–6.71).

SUMO-specific protease 1 (SENP1) is a member of the SUMOylation protease family and has a crucial role in the regulation of androgen receptor (AR) dependent transcription and inflammatory hypoxia signaling [43]. Wang et al. [43] analyzed 150 human PC specimens and they showed that SENP1 expression directly correlates with PC aggressiveness. In fact higher SENP1 scores corresponded with higher Gleason scores of PC patients. Moreover, silencing SENP1 level in metastatic PC cells, perturbs their ability to metastatize to the bone [43].

Fucosylation haptoglobin (Fuc-Hpt) is an important oligosaccharide modification associated with inflammation and cancer. Fujita et al. [44] in 98 PC cases submitted to radical prostatectomy showed that serum levels of Fuc-Hpt were significantly associated with Gleason score P < 0.01) and Fuc-Hpt levels were significantly higher in patients with Gleason score ≥ 7 than in those with Gleason score 6. Authors proposed Fuc-Hpt serum determination as a novel PC biomarker for predicting aggressiveness and prognosis [44].

Another important cytokine involved in PC progression appears to be the macrophage migration inhibitory factor (MIF) [45]. Tawdros et al. [46] showed that an increased extracellular release in MIF induces neuroendrocrine (NED) differentiation in PC stimulating an AR independent progression. The relevance of this study is to link MIF release in chronic prostatic inflammation, to the development of an aggressive phenotype such as NED in PC.

Recently, an emerging class of inflammasomes is considered as master regulators of inflammation. Inflammasomes are a group of multimeric proteins that consist of NLR protein, an apoptosis-associated speck-like protein containing a carboxyterminal CARD (ASC), and procaspase-1 [47]. Assembly of inflammasomes complex activates caspase-1 leading to the secretion and the maturation of proinflammatory cytokines IL-1 and IL-18, which causes a wide variety of biological effects associated with infections. NLRP3 is one of the most studied inflammasomes and senses pathogens and danger signals in response to injury or infection. Therefore, various stimuli such as urine reflux, uric acid crystals, bacteria, or fungi may lead to activation of inflammasome-medicated proinflammatory cytokines in the prostate driving tumor development [47].

MCs (MCs) are recognized as important effectors in Immunoglobulin-E (Ig-E) associated immune responses, and potent effector cells of the innate immune system [48]. In some tumor settings MCs have a protective role, exerted by their proinflammatory mediators, while in other tumors MCs may directly influence the advancement of the cancer cells by stimulating the neovascularity, tissue remodeling and modulation of the host immune response [48]. In PC, MCs have been only recently indicated as potential independent prognostic marker and MCs targeting associated with castration suggested as a potential therapy. Xie et al., in vitro and in vivo models, [49] demonstrated that PCa cells may have better capacity than normal prostate to recruit mast cells and infiltrating mast cells could enhance PCa resistance to chemotherapy and radiotherapy via activation of p38/p53/p21 and ATM protein kinase signals.

Several studies in the literature tried to correlate different inflammatory factors with the aggressiveness and metastatization of PC. Innate and adaptive immune cells, chemokines and cytokines, all showed to significantly correlate to PC aggressiveness (Table 1). These data together confirm the role of inflammation in PC progression but they also produce confusion to identify a reliable clinical prognostic marker. However, innate and adaptive immunity cells may play a dichotomous role, acting in the context of both pro-tumorigenic and anti-tumorigenic depending on the stage of disease, type of inflammation, and/or the tumor microenvironment [6].

Some of these factors can be detected only on prostate tissue samples, others have been suggested as serum markers. NLR can be simply obtained in the clinical practice and it is presented in several studies as a strong indicator of PC prognosis and risk of progression in metastatic cases. Its role seems to be independent to the Gleason score of the tumor.

Among Inflammatory cytokines, MIC-1, IL-6 and IL-8 may be prognostic markers of aggressiveness and metastatic progression but data on their serum determination remain limited. MCs have been recently indicated as novel independent prognostic markers in PC, although previous studies on prostatic biopsies associated high MC densities with favorable tumor characteristics and good prognosis. An explanation for these appearing to be a discrepant finding remains the observation that PC is a multifocal and heterogeneous disease [48, 49].

The potential mechanistic relationships between the molecular events associated with the persistent inflammatory response and prostate carcinogenesis have important implications for optimizing the current therapies against different prostatic disorders and PC [50].

Data on a possible prognostic role of prostatic inflammation in the clinical response to RP in PC are limited in the literature (Table 2). As of now, the studies are mainly retrospective, but some findings may suggest that the grade of inflammation in preoperative biopsy specimens could be used to risk-stratify men with PC for risk of biochemical progression after RP [54, 55]. However, in a multivariate analysis after adjustment with the other clinical parameters (PSA, clinical stage, Gleason score), inflammation did not result an independent predictor. The suggestion that serum markers of systemic inflammation may be independently related to the biochemical progression free survival must be confirmed in prospective studies.

More clinical data on the relationship between inflammation and response to radiotherapy for PC are present in the literature (Table 3). The populations analyzed are larger and more homogeneous and data are examined by multivariate analysis so to define their independent prognostic value. Markers of systemic inflammation more than marker specific for prostatic inflammation have been examined. In particular NLR but also PLR and CRP showed a significant and independent prognostic value for patients submitted to radiotherapy, either in terms of clinical progression or of overall survival, also after adjustment for Gleason score and clinical stage [59, 60]. In the intermediate and high risk classes as described by D’Amico et al., NLR and CRP are able to furthermore stratify cases in terms of clinical progression free survival [62].

Elevated NLR and CRP levels may reflect increased concentrations of these pro-inflammatory cytokines that create a microenvironment favouring PC proliferation and metastatization despite radiotherapy [62]. Furthermore, raised CRP concentrations may produces increased serum levels of vascular endothelial growth factor (VEGF) [63, 64]. However NLR, PLR and CRP are not specific markers of inflammation that might be influenced by several other conditions such as active infection, inflammatory diseases and smoking behaviour independent of PC.

Systemic inflammation represents a truly novel and unexplored prognostic variables in CRPC [74], independent to other clinical parameters such as stage. It does not substitute but rather integrates with established prognostic factors. Pinato et al. [74] suggested that inflammation may predict not only survival but also cancer-related symptomatic burden (association between NLR and performance status) and risk of chemotherapy-related toxicity (inflammation-related alterations in drug pharmacokinetic).

Data on CRPC are strongly significant and are obtained in large populations from randomized trials. Different phase II and III trials [66‐70] used OS and clinical progression free survival as end-points to evaluate the prognostic role of inflammatory variables (Table 4).

Both NLR and CRP resulted independent and significant predictors of OS and progression free survival in CRPC cases during chemotherapy with Docetaxel and for NLR also during abiraterone . An optimal threshold of NLR for the prediction of OS was set at 2–3 whereas of CRP at 8 [71].

Externally validated prognostic algorithms, in patients with metastatic CRPC considered for docetaxel, including NLR or CRP have been proposed [66‐70].

The results of these studies qualify the NLR as an independent predictor of mortality in CRPC patients and support its incorporation in prognostic models to stratify patients with CRPC.