The Lack of Epithelial Interleukin-7 and BAFF/BLyS Gene Expression in Prostate Cancer as a Possible Mechanism of Tumor Escape from Immunosurveillance

Prostate tissue is composed of a mixture of glands and intervening fibromuscular stroma (1). We have recently characterized a third element, namely, the prostate-associated lymphoid cell population (2). This is embedded in both components in the form of intraepithelial lymphocytes, mainly CD8⁺ T cells, and stromal lymphocytes, sometimes arranged (as observed in a significant number of subjects) in lymphoid follicles (LF), which are CD20⁺ B-cell follicles that may develop a germinal center (GC) surrounded by parafollicular CD3⁺ T-cell areas and are located adjacent to the double-layered glandular epithelium (2). This anatomic setting strongly suggests the possibility of an intimate cross-talk between glandular epithelia and associated lymphoid cells and leads to speculation on the local production of cytokines endowed with a lymphoid cell growth, differentiation, or survival function.

Interleukin (IL)-7 is a pleiotropic cytokine with central roles in modulating T-cell and B-cell development, and peripheral naive and memory T-cell homeostasis (3–5). Binding of IL-7 to its receptor (IL-7R), which consists of two components, the α chain (IL-7Rα) and the γ chain, shared by receptors for IL-2, IL-4, IL-9, IL-15, and IL-21, activates multiple pathways that regulate lymphocyte survival, proliferation, and differentiation. IL-7 is primarily important for the survival of memory CD8⁺ T cells through its antiapoptotic effect via up-regulation of bcl-2 (6–9). The physiologic role of IL-7 as a trophic factor for developing B cells through the modulation of bcl-2 family members has not yet been definitively determined (4), at least in humans.

The key role of the B-cell–activating factor of the tumor necrosis factor family (BAFF; also called BLyS, TALL-1, zTNF-4, or THANK; ref. 10) in B-cell biology includes immunoglobulin isotype switching, GC maintenance, and expression of surface molecules (11), whereas its prime function is the control of B-cell maturation and homeostasis (12) by stimulating the expression of prosurvival proteins bcl-2 and bcl-xL and decreasing that of proapoptotic factors such as Bak and, potentially, Blk (13–15). Signals through BAFF-R/BR3, the BAFF-specific receptor (16), increase bcl-2 expression in both T cells (17) and B cells (13), which may result in enhanced basal T-cell survival, particularly by BAFF-R⁺ T cells, such as effector and central memory subsets.

In this study, we set out to determine whether IL-7 and BAFF are expressed in the human prostate and affect intraprostatic T and B lymphocyte homeostasis. Our approach was based on laser capture microdissection (LCM), a technology that allows isolation of individual cell populations, followed by reverse transcription-PCR (RT-PCR) and real-time RT-PCR. In addition, immunofluorescence and laser scanning confocal microscopy were used to investigate protein production, combined with immunohistochemistry to characterize intraprostatic lymphocyte populations.

Patients and samples. Normal prostates were obtained from 12 untreated patients ages 57 to 63 following prostatectomy for bladder cancer (control patients). They were histologically negative for prostate cancer (PCa) or benign prostatic hyperplasia. PCas were from 46 patients (PCa patients) ages 64 to 74 following radical prostatectomy for histologically verified adenocarcinomas at biopsy. Both normal and cancer specimens, from each PCa patient, were collected.

Preoperative androgen deprivation had not been used. PCa samples were graded as Gleason score 5 (n = 9), 6 (n = 7), 7 (n = 13), 8 (n = 10), and 9 (n = 7) and staged as pT2 [organ-confined cancer; n = 27 (7 T2a, 12 T2b, and 8 T2c)] and pT3 [capsular penetration; n = 19 (17 T3a and 2 T3b)].

One half of each sample was fixed in 4% formalin and embedded in paraffin, and the other was embedded in Killik frozen section medium (Bio-Optica), snap frozen in liquid nitrogen, and preserved at −80°C.

For histology, paraffin-embedded samples were sectioned at 4 μm and stained with H&E. Single and double immunohistochemistry was done on paraffin-embedded or frozen sections depending on the antibody used.

Written informed consent was obtained from patients and the study was approved by the Ethical Committee of the “SS. Annunziata” Hospital. This investigation conformed with the principles outlined in the Declaration of Helsinki.

Antibodies and immunohistochemistry. For immunohistochemistry on the formalin-fixed, paraffin-embedded samples, sections were treated as previously reported (18) with the primary antibodies listed in Table 1.

For immunohistochemistry on the frozen samples, cryostat sections were fixed in acetone for 10 min. After washing in PBS/Tween 20, sections were stained as reported (18) with the primary antibodies listed in Table 1.

Automated cell count was done by light microscopy (19) using a Leica Imaging Workstation (Leica Microsystems) by applying a dedicated algorithm in Qwin image analysis software (version 2.7).

After exclusion of lymphocyte-forming LFs, CD8⁺ and CD20⁺ cells were counted by adding together the intraepithelial and stromal scattered lymphocytes in randomly chosen fields for the normal prostate samples (of both control and PCa patients) and in fields randomly chosen within neoplastic areas for the PCa samples. Values are represented as the mean ± SD of positive cells/field evaluated by light microscopy on single-immunostained (CD8, CD20), formalin-fixed, paraffin-embedded sections at ×400 magnification in an 85,431.59 μm² field. Eight to 12 high-power fields were examined for each section and two sections per sample were evaluated.

Both bcl-2 and terminal deoxynucleotidyl transferase–mediated dUTP nick end labeling (TUNEL) positivity were assessed by the evaluation of positive lymphocytes (double stained: CD8/bcl-2, CD20/bcl-2, and CD8/TUNEL) in the normal or in PCa samples, and the results were expressed as percentage positive staining.

Double immunohistochemistry on paraffin-embedded tissue sections was done using the EnVision G/2 Doublestain System, Rabbit/Mouse (Dako), according to the manufacturer's protocol, and analyzed under a Leica DMLB light microscope.

Double immunofluorescent stainings on frozen sections were done as described (18) and examined under an LSM 510 Meta confocal microscope (Zeiss).

TUNEL assay. DNA fragmentation associated with apoptosis was detected in 4-μm prostate sections by TUNEL staining with the ApopTag Plus Peroxidase In Situ Apoptosis kit (Millipore) according to the manufacturer's protocol. For double staining, after the TUNEL assay, sections were incubated with anti-CD8 antibody (Table 1), washed, and then incubated with biotinylated horse anti-mouse antibody (Vector Laboratories). Unbound immunoglobulin was removed by washing and slides were incubated with StrepABComplex/AP followed by development with the fuchsin substrate-chromogen system (both from Dako).

RNA isolation and RT-PCR. IL-7 and BAFF expression was investigated on frozen sections from normal prostate specimens of control patients. Two 10-μm sections per specimen were cut and put in a sterile vial. RNA was immediately isolated with RNeasy Mini kit (Qiagen). RT and PCR were done using the iScript cDNA Synthesis kit and the iTaq DNA Polymerase kit, respectively (both from Bio-Rad), according to the manufacturer's protocol. The PCR amplification primers were as follows: IL-7, 5′-AAAAACTGAATGACTTGTGT-3′ (forward) and 5′-GGTTTAGCATCCTTAGACT-3′ (reverse); BAFF, 5′-ATGGATGACTCCACAGAAAGG-3′ (forward) and 5′-TGAGGCAGCAAGACAGCAGT-3′ (reverse). The housekeeping gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a control to assess RT and PCR efficiency using the following primers: forward, 5′-AGCCGCATCTTCTTTTGCGTC-3′; reverse, 5′-TCATATTTGGCAGGTTTTTCT-3′. All primers were synthesized by Sigma-Aldrich Corp. The sizes of amplified cDNA fragments were 764 bp for IL-7, 184 bp for BAFF, and 763 bp for GAPDH. The PCR procedures were carried out on an MJ Mini Gradient Thermal Cycler (Bio-Rad) under the following conditions: IL-7, initial denaturation at 95°C for 3 min followed by 35 amplification cycles (denaturation at 94°C for 30 s, annealing at 58°C for 30 s, and extension at 72°C for 30 s); BAFF, initial denaturation at 95°C for 12 min followed by 35 amplification cycles (denaturation at 94°C for 30 s, annealing at 60°C for 30 s, and extension at 72°C for 30 s); GAPDH, initial denaturation at 95°C for 3 min followed by 35 amplification cycles (denaturation at 94°C for 30 s, annealing at 57°C for 30 s, and extension at 72°C for 30 s). For each sample, a mock reaction without reverse transcriptase (RT−) was done. The PCR products were separated on a 2% agarose gel stained with ethidium bromide. DNA bands were visualized with a Transilluminator 2000 (Bio-Rad). All experiments were done in triplicate.

LCM and real-time RT-PCR. For LCM, we used the P.A.L.M. Micro Beam System (P.A.L.M. Microlaser Technologies). Frozen sections (10 μm), from normal prostate samples (of both control and PCa patients) and PCa samples, were mounted on polyethylene naphthalate membrane-covered slides (P.A.L.M.), thawed at room temperature, and immersed in cold acetone (5 min). Two sections per sample were analyzed. All reagents were prepared using Ultrapure DNase/RNase-free distilled water (Invitrogen). Immediately after staining, sections were used for LCM. Selected cells were cut and catapulted intact into the cap of an LPC-Microfuge Tube (P.A.L.M.), and RNA was immediately isolated with the RNeasy Micro kit (Qiagen). RT-PCR and visualization of amplified cDNA fragments were done as described above.

The real-time RT-PCR was carried out using the iScript cDNA Synthesis kit for RT and the iQ SYBR Green Supermix for real-time PCR (both from Bio-Rad) with the following primers: IL-7, 5′-AGAGTGTTCTAATGGTCAGCATCG-3′ (forward) and 5′-AGTGGAGATCAAAATCACCAGTGC-3′ (reverse); BAFF, 5′-CGTGCCGTTCAGGGTCCAG-3′ (forward) and 5′-TCGAAACAAAGTCACCAGACTCAAT-3′ (reverse). The housekeeping gene hypoxanthine phosphoribosyltransferase 1 (HPRT) was used as endogenous control using the following primers: forward, 5′-AGACTTTGCTTTCCTTGGTCAGG-3′; reverse, 5′-GTCTGGCTTATATCCAACACTTCG-3′. Primers were designed with Beacon Designer software (Premier Biosoft International) in our laboratory and synthesized by Sigma-Aldrich. The sizes of the amplified cDNA fragments were 198 bp for IL-7, 297 bp for BAFF, and 101 bp for HPRT. The real-time RT-PCR was done using the MiniOpticon System (Bio-Rad) with SYBR Green fluorophore under the following conditions: denaturation at 95°C for 3 min followed by 40 amplification cycles (denaturation at 95°C for 10 s and annealing/extension at 60°C for 30 s). Melting curve analysis was done to assess the specificity of PCR products (samples were heated from 65°C to 90°C at a ramp rate of 0.3°C every 5 s). The efficiency of reaction for each target was evaluated by amplifying serial dilutions of cDNA. Relative quantification of mRNA was done according to the comparative threshold cycle method with HPRT as calibrator using the Gene Expression Analysis for iCycler iQ real-time PCR detection systems (Bio-Rad). The samples were processed in triplicate and wells without added cDNA served as negative controls.

Statistical analysis. Data are reported as mean and SD. Differences between lymphocyte populations, the percentage of bcl-2⁺ or TUNEL⁺ lymphocytes, and the relative expression of IL-7 and BAFF in groups of samples from PCa patients or control patients were evaluated with the Mann-Whitney U test. Differences between groups of patients with different Gleason scores for the relative expression of IL-7 and BAFF by real-time RT-PCR were assessed by one-way ANOVA test. Statistical tests were evaluated at an α level of 0.05. The Statistical Package for the Social Sciences software version 11.0 (SPSS, Inc.) was used with P < 0.05 as the significance cutoff.

Human prostatic epithelium expresses IL-7 mRNA and produces IL-7 protein. Because the human prostate is regularly embedded by lymphocytes, mainly T cells (2), we used RT-PCR to assess the expression of IL-7, a cytokine essential for T-cell development and homeostasis (4), in normal prostate sections from untreated patients following prostatectomy for bladder cancer.

IL-7 mRNA was expressed as shown in Fig. 1A, where the specificity of the amplified band was validated by its predicted size (764 bp). Double immunofluorescent staining and confocal microscopy revealed IL-7 protein production in most of the glandular epithelium in both the apical (CK18⁺) and the basal (p63⁺; ref. 20) epithelial cell layer (Fig. 1B, a-g) and frequently in blood vessel endothelial cells (marked by anti-vWF; Fig. 1B, h). LCM of the whole epithelia, marked by CK5/6/18 immunostaining on frozen sections, followed by RT-PCR analyses of the isolated cells, showed their expression of IL-7 mRNA (Fig. 1C) in accordance with the immunohistochemical findings.

Lymphocytes and endothelial cells express IL-7Rα. This demonstration of IL-7 production by prostatic epithelial and endothelial cells prompted us to look for its target by investigating the expression of IL-7Rα (CD127), the specific component of the IL-7R complex (5).

Immunohistochemistry on paraffin sections revealed IL-7Rα expression on periepithelial and intraepithelial small mononuclear cells displaying a lymphocyte morphology (Fig. 1D, a), mostly identified as CD3⁺, and on stromal lymphocytes, particularly those located at the parafollicular T-cell zone of periglandular lymphoid aggregates (Fig. 1D, b). Expression of IL-7Rα could also be found on stromal microvessels (Fig. 1D, c) and sometimes on vessels with HEV-like features, such as those found in lymphoid aggregates (Fig. 1D, d).

Human prostatic epithelium expresses BAFF mRNA and produces BAFF protein. In view of the B-cell content of the normal human prostate (2), we investigated the local expression of BAFF, which plays a fundamental role in regulating peripheral B-cell survival and homeostasis (12).

RT-PCR showed BAFF mRNA in the prostate tissue (Fig. 2A). The specificity of the related amplified band was validated by its predicted size (184 bp). Immunohistochemistry on frozen sections disclosed BAFF protein expression in the basal epithelial layers of many glands of the prostate (Fig. 2B and C). RT-PCR examination of microdissected p63⁺ basal cells supported this result by showing their expression of BAFF mRNA (Fig. 2D).

BAFF protein was also expressed by CD3⁺ T lymphocytes, mostly those with the CD4⁺CD45RO⁺ memory phenotype, arranged in loosely or well-organized follicle-like structures, as assessed by confocal microscopy (Fig. 3A, a-i).

Expression of BAFF-R, mostly by CD20⁺ cells. BAFF binds to three receptors, namely, BCMA (B-cell maturation antigen; ref. 12), TACI (transmembrane activator and calcium modulator and cyclophilin ligand interactor; ref. 12), both of which also bind APRIL (a proliferation-inducing ligand), and BAFF-R/BR3 (16), which is specific for BAFF. We therefore looked for the expression of BAFF-R/BR3 by immunohistochemical analysis of prostate paraffin sections. BAFF-R was expressed by many lymphocytes scattered in the periglandular stroma (Fig. 3B, a), whereas in the LFs it was mainly found on CD20⁺ B cells and rarely on T cells, as assessed by double immunohistochemistry (Fig. 3B, b-d). BAFF-R was weakly expressed by GC B lymphocytes and clearly expressed by B cells colonizing the mantle zone, as illustrated in Fig. 3C (a and b).

Prostate-infiltrating lymphocytes display low proliferation but wide expression of the antiapoptotic protein bcl-2. We have recently shown that most prostate-infiltrating T and B cells are located inside (in the form of intraepithelial lymphocytes) or close to (sometimes in the form of LFs) the glandular epithelium (2). In the present study, we have observed that these lymphocytes mostly express IL-7Rα and/or BAFF-R. It can thus be hypothesized that production of IL-7 and BAFF by the prostatic epithelia is involved in modulating intraprostatic lymphocyte proliferation and/or survival.

Intraprostatic lymphocytes were scarcely proliferating, as assessed by Ki-67 immunostaining. As previously described (2), in addition to a few basal cells of the glandular epithelium, Ki-67 positivity was only detected in GC lymphocytes of LFs (Fig. 4A and B).

The vast majority of prostatic lymphocytes, both stromal (organized in LFs, that is, CD20⁺ B-cell follicle encircled by parafollicular T-cell area, Fig. 4C-H, or not organized, Fig. 4I-K) and intraepithelial lymphocytes (Fig. 4L), strongly expressed the antiapoptotic protein bcl-2. Accordingly, they showed very low apoptotic rate as revealed by their rare positivity in the TUNEL assay (Fig. 4M). Prostate-homing lymphocytes thus have low proliferative and high antiapoptotic activity.

Loss of IL-7 and BAFF production in neoplastic prostatic glands is accompanied by reduction of CD8⁺ and CD20⁺ cell infiltrates. To find out whether epithelial production of IL-7 and BAFF and thus their trophic action on intraprostatic lymphocytes are subverted by tumor onset, we next did immunohistochemistry, LCM, and real-time RT-PCR to determine their expression level on prostate adenocarcinoma sections from 46 untreated patients.

By LCM, we isolated malignant glands from cancer samples and the whole normal glandular epithelium from normal prostate samples (of both control and PCa patients). Real-time RT-PCR revealed that, in contrast to normal prostatic glands, malignant glands showed low to absent, ∼60 times lower than in the normal glands (Fig. 5A), IL-7 mRNA expression. In addition, immunohistochemical examination of subsequent serial sections showed that IL-7 protein production was distinct in normal glands and scarce to absent in neoplastic glands (Fig. 5B, a and b). To determine whether this deficiency of IL-7 production was associated with alterations in T lymphocyte content, we next compared the lymphocyte infiltrates of normal (from both control and PCa patients) and neoplastic tissues. In comparison with normal prostate, PCa was largely lacking in the LFs but still detectable in the remaining nonneoplastic tissue. Expression of the chemokines B lymphocyte chemoattractant, also called CXCL13, and secondary lymphoid tissue chemokine, also known as CCL21, which we previously found (2) associated with LFs, was lost too (Fig. 5B, c-e). Accordingly, CD4⁺ T lymphocytes, mostly lying in the parafollicular areas of LFs, were scarce in PCa. The remaining T cells were mainly CD8⁺ (Fig. 5B, f and g), and the vast majority displayed the CD45RA⁻CD45RO⁺CCR7⁻ effector-memory phenotype and were CD27⁺CD28⁻ preterminally differentiated or CD27⁺CD28⁺ at an early differentiation stage (21) as assessed by immunohistochemistry (Fig. 5C, a-f). CD8⁺ T lymphocytes were significantly (P < 0.05) more numerous in the normal prostate, 26.2 ± 5.5 (mean ± SD of positive cells/×400 field), located inside and within glandular epithelium, than in PCa, 17.8 ± 4.2, located within, rather than inside, neoplastic glands (Fig. 5B, f and g). At the same time, the percentage of CD8⁺ cells with positive cytoplasmic staining for the antiapoptotic protein bcl-2 was decreased in PCa (85.1 ± 5.3%; Fig. 5D, a and b), in comparison with the normal prostate (97.0 ± 2.4%), as assessed by double CD8/bcl-2 staining on paraffin sections. Accordingly, the percentage of TUNEL⁺ CD8⁺ cells was increased in PCa, 16.5 ± 3.7%, when compared with normal prostates, 4.4 ± 2.0% (Fig. 5D, c and d).

LCM followed by real-time RT-PCR examination of neoplastic glands versus normal glands revealed a low to absent, about five times lower than in the normal glands, BAFF mRNA expression (Fig. 5E). Accordingly, a loss of BAFF protein production was assessed by immunohistochemistry (Fig. 5F, a and b). Concomitant decline in the number of periglandular CD20⁺ B lymphocytes was revealed in PCa, within neoplastic glands, 3.0 ± 2.4, compared with the normal prostate, 13.1 ± 3.9 (Fig. 5F, c and d). The percentage of bcl-2–expressing CD20⁺ B cells was also reduced in PCa (88.5 ± 4.2%) compared with the normal prostate (95.8 ± 3.0%).

CD8⁺ and CD20⁺ lymphocyte counts in PCa were significantly lower than in the normal prostate starting from the lower Gleason score and no significant association was disclosed by Mann-Whitney U test or the χ² test between these counts and the different Gleason scores, tumor-node-metastasis stages, and patient age.

Notably, data about cell counts and IL-7 or BAFF production obtained in normal samples from PCa patients were analogous to those obtained in normal samples from control patients. Furthermore, no significant differences of either IL-7 or BAFF expression emerged, with real-time RT-PCR analyses, by ANOVA test, between groups of patients with different Gleason scores.

We here provide the first evidence that the prostatic epithelium is a source of both IL-7 and, to a lesser extent, BAFF, two cytokines that are needed for lymphocyte development and homeostasis, and may also act on intraprostatic T and B cells because most of them express IL-7Rα and BAFF-R, respectively. This observation strongly indicates the existence of an epithelium lymphocyte cross-talk that may be subverted during tumor onset because replacement of the normal epithelium by neoplastic cells implies the dramatic decline of IL-7 and BAFF/BLyS production associated with a reduced lymphocyte bcl-2 expression, increased apoptosis, and consequent impoverishment of intraepithelial and stromal lymphocytes with loss of LFs.

IL-7 is a nonredundant trophic factor for immune cell development that primarily acts by promoting cell survival (22, 23). In humans, it has been identified from thymic stromal cells (primarily MHC class II⁺ epithelial cells) and subsequently intestinal epithelial cells (24), keratinocytes (25), and hepatocytes (26). It is also produced by follicular (27) and peripheral blood dendritic cells (28), endothelial cells (27), smooth muscle cells (27), and fibroblasts (26). Even so, it is not really clear which cells supply the IL-7 for peripheral T cells nor whether its synthesis by these cells is regulated or constitutive (29). Our study shows that the normal prostate, which is endowed with a discrete number of scattered intraepithelial and stromal lymphoid cells (2, 30, 31), constitutively produces IL-7 in the whole glandular epithelium as the major source with various degrees of cooperation on the part of the vessels. In the light of the myriad effects of IL-7 on mature T cells and antigen-presenting cells, along with its proven ability to modulate the immune response in infections and malignancies (32, 33) by inducing antitumor immunity, its modulation of this response in prostate-associated lymphocytes during prostatitis and PCa should not be surprising. Whether alterations of the prostate microenvironment by inflammatory stimuli regulate epithelial and/or endothelial IL-7 production, as observed for keratinocytes (34) or intestinal epithelial cells (35), remains to be investigated. This production is certainly impaired in PCa because neoplastic glands express ∼60 times less IL-7 mRNA than their normal counterparts and hence fail to produce detectable levels of IL-7 protein. This, in fact, may only be found in the remaining nonneoplastic glandular epithelia where the prostate-associated lymphoid cell population is substantially unaffected, whereas it is clearly altered in cancer as shown by a significant decrease in T and B cells and loss of well-organized LFs that was dramatically evident starting from the lower Gleason score. The PCa T-cell infiltrate is mainly composed of effector-memory CD8⁺ cells devoid of terminal differentiation (21) possibly as a result of IL-7 deficiency (4, 8), an issue we are currently investigating. They are mainly confined to the stroma, undergo apoptosis more frequently than in the normal prostate, and, like B cells, display reduced bcl-2 expression.

Periglandular B lymphocytes and B-cell follicles, along with production of CXCL13 by follicular-like dendritic cells and CCL21 at the site of HEV-like vessels in LFs, usually disappear in the neoplastic areas together with the epithelial production of BAFF because of the loss of p63⁺ basal cell layer in neoplastic glands. PCa may lack the upstream modulators of CXCL13 and CCL21 production that are needed for B-cell recruitment into LFs (36, 37), and naive T-cell recruitment, and enter the parafollicular T-cell zones of LFs (38). BAFF also affects the formation of LFs supporting B-cell differentiation and GC development (11, 12), but it essentially functions as a regulator of B-cell survival via bcl-2 antiapoptotic protein (12). Initially believed to be produced solely by cells of the myeloid lineage (39), BAFF has now been shown to be expressed by cells of nonhematopoietic origin, such as fibroblast-like synoviocytes, osteoclasts, and ductal epithelial cells of salivary glands in Sjögren's syndrome (11, 40). By contrast, in the normal prostate, BAFF is produced by the basal layer of the gland in the absence of evident signs of inflammation and by memory CD4⁺ T lymphocytes arranged in follicle-like structures. These are lost too in PCa, resulting in a further decrease of BAFF available for local B-cell survival.

Taken together, our data suggest that in the human prostate, IL-7, essentially, and BAFF, to a lesser extent, regulate the local mucosal immune system by sustaining T-cell and B-cell survival via bcl-2 up-regulation and thus cooperate to preserve the prostate-associated lymphoid cell arrangement and functions. In addition to the already known mechanisms influencing lymphocyte survival in prostate as in other types of cancer (41, 42), the dramatic decline in the production of these two key lymphotropic molecules in PCa is clear evidence of a new pathway (i.e., deficiency in the prostate-associated lymphocyte population), enabling a tumor to escape from surveillance on the part of the immune system.

No potential conflicts of interest were disclosed.

We thank Prof. John Iliffe for editing and Mario P. Colombo for critical reading of the manuscript.