Purpose: The nuclear transcription factor nuclear factor-κB (NFκB) and its inhibitor, IκB, regulate the transcription of various genes involved in cell proliferation, adhesion, and survival. The NFκB transcription factor complex plays a role in cancer development and progression through its influence on apoptosis. More recently, NFκB has been shown to be activated in human and androgen-independent prostate cancer cells. To our knowledge, this is the first study demonstrating the prognostic significance of NFκB immunoreactivity in prostate adenocarcinomas (PACs).
Experimental Design: Using prostatectomy specimens, we performed immunohistochemical staining for NFκB and IκBα (Santa Cruz Biotechnology) on formalin-fixed, paraffin-embedded sections obtained from 136 patients with PAC. Cytoplasmic and nuclear immunoreactivity was scored for intensity and distribution, and results were correlated with preoperative serum prostate-specific antigen, tumor grade, stage, DNA ploidy (Feulgen spectroscopy), and biochemical disease recurrence.
Results: Forty-nine percent of PACs overexpressed cytoplasmic NFκB, and 63% showed decreased IκB expression. Cytoplasmic NFκB overexpression correlated with advanced tumor stage (P = 0.048), aneuploidy (P = 0.022), and biochemical disease recurrence (P = 0.001). When we compared the means for the NFκB-positive and -negative subgroups, NFκB overexpression correlated with preoperative serum prostate-specific antigen (P = 0.04) and DNA index (P = 0.05). Fifteen percent of PACs expressed nuclear NFκB, which correlated with high tumor grade (P = 0.001) and advanced stage (P = 0.05). Decreased IκBα expression correlated with high tumor grade (P = 0.015). On multivariate analysis, tumor stage (P = 0.043) and NFκB overexpression (P = 0.006) were independent predictors of biochemical recurrence.
Conclusion: These results support a role for NFκB pathway proteins in the tumorigenesis of PACs. The findings are also consistent with reported experimental studies suggesting a new strategy of combined chemotherapy and specific NFκB blockade in decreasing the rate of disease relapse.
The nuclear transcription factor nuclear factor-κB (NFκB) and its inhibitor, IκB, regulate the transcription of various genes involved in cell proliferation, adhesion, and survival. NFκB represents a group of structurally related and evolutionarily conserved proteins with five members in mammals, including RelA (p65). NFκB/Rel proteins share a highly conserved N-terminal Rel homology domain responsible for DNA binding, dimerization, and association with the IκB inhibitory proteins (1) . NFκB binds to multiple DNA sequences initiating the transcription of a wide variety of genes including cytokines [interleukin-1 (IL-1), IL-6, IL-8, and tumor necrosis factor], angiogenesis factors (vascular endothelial growth factor), cell adhesion molecules (intercellular adhesion molecule-1 and vascular cell adhesion molecule-1), enzymes (cyclooxygenase-2 and nitric oxide synthase), and antiapoptotic factors, including bcl-2 and survivin (2, 3, 4) .
IκBα is a member of the IκB gene family containing seven known mammalian members characterized by their 6–7 ankyrin repeats, which allow them to interact with members of the Rel family of transcription factors (1 , 5) . In quiescent cells, NFκB is located in the cytoplasm in an inactive form, bound to its inhibitor molecule, IκBα. Stimulation of cells through a variety of mechanisms (e.g., viruses, growth factors, antigens, radiation, or chemotherapeutic drugs) triggers a cascade of signaling events that ultimately result in the degradation of IκBα by the proteasome. This degradation releases active NFκB, which then translocates into the nucleus, where it binds to specific DNA sequences on its target genes.
A series of studies have recently found that NFκB expression may be involved in the development and progression of a variety of malignant processes, including both solid tumors (3 , 4 , 6) and hematopoietic neoplasms (7 , 8) . Studies of NFκB expression in prostate cancer have mostly been limited to analysis of cell lines and preclinical models (9, 10, 11, 12, 13, 14, 15, 16, 17, 18) . The significance of NFκB and IκBα expression as prognostic factors in clinical prostate cancer has not been reported previously.
MATERIALS AND METHODS
Specimen Collection, Tumor Grading, and Pathological Staging.
We randomly selected 136 patients who had undergone radical prostatectomy for biopsy-proven prostate adenocarcinoma (PAC) between 1987 and 1997 at the Albany Medical Center Hospital. H&E-stained slides from each radical prostatectomy case were reviewed, and a Gleason grade (19) and pathological stage (20) were assigned. During review, multiple blocks were identified based on the presence of adequate tumor and the representative nature of the overall grade. Tumors were classified as high grade when the combined Gleason score was ≥7 and as low grade when the combined score was ≤6. Serum prostate-specific antigen (PSA) levels were obtained from the patient’s medical records in every case. Serum PSA was measured by the Hybritech tandem method (Beckman Coulter, Inc., Brea, CA). A postsurgical elevation of the PSA level from a baseline level of 0 ng/ml to >0.4 ng/ml on two consecutive occasions was considered biochemical evidence of disease recurrence. Follow-up information was obtained from review of the patients’ medical records.
Immunohistochemical staining for NFκB and IκBα proteins was performed by an automated method on the Ventana ES (Ventana Medical Systems, Inc., Tucson, AZ) with an indirect biotin-avidin-diaminobenzidine detection system on contiguous 4-μm formalin-fixed, paraffin-embedded sections from a representative block in each case. After deparaffinization to water, the antigenic determinant sites for NFκB were unmasked in citrate buffer with steam for 60 min. The primary antibody used for NFκB, an IgG1 class mouse monoclonal directed against the p65 (F-6) RelA component of the NFκB complex (Santa Cruz Biotechnology Inc., Santa Cruz, CA) was used at a dilution of 1:80 for 32 min at 37°C. The primary antibody for IκBα, an IgG1 class mouse monoclonal (Santa Cruz Biotechnology) was used at a dilution of 1:5 overnight at room temperature. The secondary antibody for both proteins was biotinylated goat antimouse immunoglobulins (DAKO, Carpenteria, CA) at a dilution of 1:250. After the development of the color with diaminobenzidine, the slides were counterstained with hematoxylin. Similarly processed sections from human adenoid and adenoid and breast cancers were used as positive controls for NFκB and IκBα immunostaining, respectively. To confirm the specificity of both primary antibodies, negative control slides were run with every batch, which included an isotype-matched immunoglobulin at the same concentration as that of the primary antibody.
Expression of both proteins was essentially cytoplasmic with scattered positive nuclei. To further verify the predominantly cytoplasmic immunoreactivity obtained with the NFκB p65 (F6) clone used in our study, we ran a set of controls along with 30 of the 136 randomly selected PACs with two additional NFκB p65 antibodies. Additionally, to further verify the nuclear NFκB positivity rate in this study, a phosphorylation-specific antibody, phospho-NFκBp65 (Ser536; Cell Signaling, Beverly, MA) was run on the same subset. To confirm the specificity of this primary antibody, immunoreactivity was blocked by preabsorption of the primary antibody with the antigen peptide (Cell Signaling) and controlled by preabsorption with an unrelated antigen peptide. Protocol details for these three antibodies are summarized in Table 1⇓ .
Immunoreactivity for both NFκB and IκBα was interpreted without previous knowledge of any of the clinicopathological parameters. The intensity and distribution of cytoplasmic staining were considered in the semiquantitative assessment of the immunohistochemical results for both antibodies. Additionally, nuclear immunoreactivity for NFκB was scored independently. The intensity of cytoplasmic and/or nuclear staining was subjectively graded as weak, moderate, or intense. The distribution of staining in the tumor cells was graded as focal (≤10%), regional (11–50%), or diffuse (>50%). Cases in which the cytoplasmic staining patterns were categorized as intense diffuse, intense regional, and moderate diffuse were considered overexpression of the two proteins, whereas cases with any pattern of nuclear staining were considered positive for nuclear NFκB.
Quantitative DNA Analysis.
We stained 5-μm formalin-fixed, paraffin-embedded sections by the Feulgen method and analyzed them for DNA content with the CAS 200 Image Analyzer (Cell Analysis Systems, Lombard, IL). After the instrument was calibrated against similarly stained tetraploid rat hepatocytes, the DNA content of the PACs was measured in a minimum of 100 tumor cells, and the tumor DNA index was determined by comparison with the control diploid cells of the benign prostatic epithelium. All of the tumor cell histograms were reviewed without knowledge of the tumor grade, stage, recurrence status, or immunohistochemical results. A DNA index of 0.77–1.22 was considered diploid. Peaks in the tetraploid region containing <15% of the total cell population were considered to be the G2-M components of diploid cell populations. Tumors with tetraploid peaks >15% and hyperdiploid, nontetraploid peaks were considered to be nondiploid (aneuploid; Ref. 21 ).
Statistical comparisons were carried out with the STATA software (Stata Corporation, College Station, TX). The χ2 test was used to determine the significance of the associations between protein expression and prognostic variables. The t test was used to compare the equality of the means between positive and negative subgroups for each protein. Multivariate analysis including clinicopathological parameters and expression of each protein was performed with the Cox proportional hazards model. The level of significance was set at P ≤ 0.05.
The mean age of the patients was 66 years (range, 49–94 years), and the mean preoperative PSA level was 12.4 ng/ml (range, 1.6–87.8 ng/ml). Of the 136 PACs, there were 76 (56%) low-grade (Gleason score ≤6) and 60 (44%) high-grade (Gleason score ≥7) tumors. At prostatectomy, there were 80 (59%) organ-confined tumors (stages I and II) and 56 (41%) advanced stage (III and IV) tumors. Of the 96 cases previously analyzed for total DNA content, 64 (67%) were diploid and 32 (33%) were nondiploid. Follow-up information was available for all patients, of which 55 (40%) had biochemical postsurgical disease recurrence.
NFκB and IκBα Expression by Immunohistochemistry (Table 2)⇓ .
The immunostaining pattern for both proteins was essentially cytoplasmic, with tumor cells showing moderate to intense positivity as opposed to relatively weaker expression in benign elements (Fig. 1)⇓ . There was scattered nuclear immunoreactivity for the NFκB protein in both tumor and benign epithelial cells. Sixty-six of 136 (49%) PACs overexpressed cytoplasmic NFκB, and 86 (63%) showed decreased expression of IκBα. There was no correlation between the expressions of the two proteins in each case. NFκB overexpression correlated with advanced tumor stage, with 33 of 66 (50%) stage III or IV tumors overexpressing NFκB compared with 23 of 70 (33%) not overexpressing NFκB (P = 0.048). NFκB overexpression was also associated with DNA ploidy status, with 23 of 51 (45%) nondiploid tumors overexpressing NFκB compared with 9 of 45 (20%) not overexpressing NFκB (P = 0.022). The mean DNA index for NFκB-positive tumors was 1.28 compared with 1.09 for the NFκB-negative cases (P = 0.05).
NFκB overexpression correlated with biochemical disease recurrence, with 36 of the 66 tumors (50%) that recurred overexpressing NFκB compared with 19 of 70 (27%) PACs that did not overexpress NFκB (P = 0.001). The mean serum PSA level at the time of diagnosis for the NFκB-positive group (17.14 ng/ml) was significantly greater than that for the NFκB-negative group (10.06 ng/ml; P = 0.04). NFκB overexpression did not correlate with tumor grade. Decreased IκBα expression correlated with tumor grade with 45 of 60 (75%) high-grade tumors showing decreased IκBα compared with 41 of 76 (54%) low-grade tumors (P = 0.015). IκBα immunoreactivity did not correlate with other prognostic variables.
There were no appreciable changes in the staining patterns or results with either the NFκBp65 clone 20 (BD Transduction) or the NFκBp65 RelA (Chemicon) antibodies.
Nuclear NFκB Expression by Immunohistochemistry.
Nuclear immunoreactivity was noted in both tumor and benign epithelial cells. Twenty of 136 (15%) PACs expressed nuclear NFκB. Although nuclear NFκB positivity correlated with high tumor grade (P = 0.01) and advanced stage (P = 0.05) with a trend for association with DNA ploidy status (P = 0.08), in 6 of 20 (30%) of these cases, the tumor cell immunoreactivity was equal to or less than the immunoreactivity of the benign cells. There was no appreciable change in the staining patterns or results with either the NFκBp65 clone 20 (BD Transduction) or the NFκBp65 RelA (Chemicon) antibodies.
When we stained with the phospho-NFκBp65 antibody, both a nuclear and cytoplasmic staining pattern was observed. In addition, nuclear positivity was noted in both tumor and benign epithelial cells. Although the nuclear positivity rate for the subset was 56% (17 of 30), in 71% (12 of 17) of cases, the tumor cell immunoreactivity was less than or equal to the immunoreactivity of the benign cells.
Disease Recurrence Analysis.
On univariate analysis, NFκB overexpression correlated with biochemical disease recurrence (P = 0.001; Fig. 2⇓ ). On multivariate analysis, NFκB immunoreactivity (P = 0.006) and advanced tumor stage (P = 0.043) were independent predictors of biochemical disease recurrence. IκBα immunoreactivity did not correlate with biochemical disease recurrence.
A wide variety of morphology-driven and molecular markers have been studied for their ability to predict disease outcome in prostate cancer (22, 23, 24, 25) . Traditional morphology-driven measures have included tumor grade, volume, and pathological stage. Numerous molecular markers have been proposed for their potential clinical utility, including the determination of p21, p27, cyclin D1, p53, bcl-2, E-cadherin, HER-2/neu, matrix metalloproteases, telomerase, and glutathione S-transferase-π (22, 23, 24, 25, 26) . Expanded use of these markers for the individualization of therapy, however, has been hampered by a lack of universal acceptance of their prognostic significance, problems concerning the specificity and sensitivity of the available testing platforms for each marker, limited available tissue, and the concern that the inherent heterogeneity of prostate cancer could cause false-negative results, particularly for patients for whom a narrow-bore needle biopsy is the only sample available for testing (23) .
In the present study, overexpression of NFκB determined by immunohistochemistry on prostatectomy specimens significantly correlated with higher mean preoperative serum PSA levels, mean tumor grade, nondiploid DNA content, and advanced stage and independently predicted biochemical disease recurrence. This appears to be the first attempt to link NFκB levels with prostate cancer outcome. Studies on NFκB levels in other solid tumors are similarly at an early stage of characterization. To date, NFκB has been studied extensively in breast cancer preclinical and experimental models, but not as a clinical prognostic factor (4 , 27 , 28) . Interestingly, NFκB activation in breast cancer cell lines has been linked to the expression of both cyclin D1 (29) and HER-2/neu (30) , two well-known adverse outcome predictors in breast and, possibly, prostate cancer (25 , 26) . NFκB activation has also been associated in experimental models and noncorrelative tissue assessments of gastric (31) , colorectal (32) , lung (33) , and head and neck squamous cell cancers (34) . The major emphasis of these studies has been the potential role of NFκB in the development of resistance to standard-of-care anticancer drugs designed to treat these tumors (2, 3, 4 , 35) .
Studies of NFκB activation in prostate cancer cell lines and tissues have emphasized the impact on the expression of bcl-2 family members and survivin as well as apoptosis (9, 10, 11, 12, 13) . Bcl-2 expression has been studied extensively in benign prostate tissues (36) and preneoplastic conditions (37) . Bcl-2 expression has generally correlated with adverse outcome in invasive prostate cancer specimens (38 , 39) , although survivin expression has not (40) . The blockade of NFκB activity by transfection in PC-3 cells has been associated with decreased angiogenesis, invasion, and metastasis in experimental models (14) . In a recent study, androgen-independent prostate cancer xenografts had higher constitutive NFκB-binding activity than androgen-dependent xenografts (15) . In another recent study, constitutive activation of NFκB was observed in androgen-receptor-negative, but not androgen-receptor-positive, prostate cancer cell lines (16) .
In a recent study, human prostate cancer tissues were assessed by immunohistochemistry using an anti-p65 (RelA) antibody (Clone G96-337; PharMingen, Inc., San Diego, CA); nuclear staining was identified but not linked to other prognostic variables or disease outcome (16) . In the present study, we used another anti-p65 (RelA) antibody (Santa Cruz Biotechnology) and found that immunoreactivity was essentially distributed in the cytoplasm with only scattered nuclear positivity. This pattern was confirmed with two additional NFκB p65 clones. Cytoplasmic NFκB is considered to be in an inactive form with nuclear translocation required for constitutive activation (2, 3, 4) . Interestingly, various other groups have demonstrated a predominant cytoplasmic and focal nuclear localization with both polyclonal and monoclonal antibodies. Using polyclonal anti-p65 antibodies, studies have shown a predominant cytoplasmic staining in tumor cells and focal positivity in the benign elements in colorectal (41, 42, 43) , gastric (31) , and oral carcinomas (44) . Using monoclonal antibody with high selectivity and specificity, studies on differentiating B cells have demonstrated that only ∼10–20% of the NFκB protein was detectable in the nucleus, whereas the remaining 80–90% remained in the cytoplasm (45 , 46) . It is therefore obvious that immunohistochemical localization may or may not reflect the dynamic functional state of the protein. Possible mechanisms to explain this phenomenon include a mutated IκB masking the nuclear translocation signal in RelA and mutations in RelA prohibiting IκB binding, resulting in cytoplasmic accumulation of both active and inactive forms of the NFκB protein (47) .
In quiescent cells, NFκB is bound to IκB inhibitory proteins and retained in the cytoplasm. Phosphorylation of IκBs by IκB kinase tags them for polyubiquitination. After ubiquitination, the IκB proteins are rapidly degraded by the proteasome, freeing NFκB to enter the nucleus, bind to DNA, and activate transcription (48) . In the present study, decreased expression of IκBα in these primary tumors correlated only with tumor grade. Although based on a general concept that IκB should be inversely related to NFκB, it is not always the case, as demonstrated in this study and others. In fact, Charalambous et al. (43) and Nakayama et al. (44) have reported an increased, rather than decreased, IκB protein expression paralleling that of NFκB positivity in human colorectal and oral cavity squamous cell cancers, respectively. Several others have documented that the interaction between these two proteins is complex and involves numerous other proteins, including cyclooxygenase-2, IL-1β, IL-6, and AKT kinase, among others.
More recently, the PTEN tumor suppressor protein and AKT pathway have emerged as potential prognostic factors in prostate cancer (49, 50, 51, 52, 53, 54) . Studies of PC-3 cells indicate that PTEN can inhibit the phosphatidylinositol 3′-kinase/AKT/NFκB pathway (52) . Phosphatidylinositol 3′-kinase may stimulate the growth of prostate cancer through the androgen receptor pathway (53) . Thus, AKT activation may be associated with prostate cancer cell cycle dysregulation, resistance to apoptosis, and progression to androgen independence (49 , 50) .
These studies of the NFκB pathway in prostate and other cancers have led to the concept of NFκB as a target in anticancer therapy (3 , 55) . Recently, proteasome inhibitors and IκBα kinase inhibitors have been used to target the NFκB pathway in both preclinical models and clinical trials involving patients with prostate cancer and other solid tumors (3 , 55, 56, 57) . With the information presented here concerning the adverse prognostic impact associated with overexpression of NFκB in prostate cancer, it is anticipated that the NFκB pathway will continue to be of significant interest as a target for therapy in this disease.
Grant support: Funded in part by Millennium Pharmaceuticals, Inc.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Note: Presented in part at the 92nd Annual Meeting of the United States and Canadian Academy of Pathology, March 2003, Washington, DC.
Requests for reprints: Jeffrey S. Ross, Department of Pathology and Laboratory Medicine, Albany Medical College, Mail Code 81, 47 New Scotland Avenue, Albany, NY 12208. Phone: (518) 262-5461; Fax: (518) 262-3663; E-mail:
- Received April 7, 2003.
- Revision received December 22, 2003.
- Accepted December 30, 2003.