Comparison of microvessel density and PSA level in prostatic adenocarcinoma of Gleason score 6 and 7

Purpose: The aim of our investigation was to determine differences in microvessel density (MVD) and serum levels of prostate-specific antigen (PSA) between groups of patients with Gleason score (GS) 6 adenocarcinoma and patients with GS 7 adenocarcinoma. Methods: The study was done in a series of 26 patients with prostatic adenocarcinoma. Specimens were divided into two groups: GS 6 (13 cases) and GS 7 (13 cases). Intratumoral microvasculature was highlighted by immunohistochemical means using an antibody against endoglin. Endoglin stained microvessels in and around the tumor but showed weak or no staining for blood Izvleček


INTRODUCTION
Angiogenesis is the formation of new blood vessels.Capillaries "sprout" from pre-existing vessels and have an important role in the progression and metastasis of tumors (1,2).Studies have suggested that the microvessel density (MVD) of prostatic adenocarcinoma may be of prognostic value (3)(4)(5)(6)(7).Endoglin (CD 105) is a receptor for transforming growth factor β1.It is expressed on endothelial cells during tumor angiogenesis and inflammation with weak or negative expression in the vascular endothelium of normal tissue (5,(8)(9)(10)(11)(12).MVD evaluation as determined using antiendoglin monoclonal antibodies has been shown to be an independent prognostic factor for certain types of malignant neoplasia, such as breast carcinoma and non-small-cell lung carcinoma (13,14).Adenocarcinoma of the prostate gland is the most commonly diagnosed male malignancy in the Euro-pean Union and USA (15,16).Parameters that can stratify patients for type of therapy based on likelihood of tumor progression are clinical stage, serum levels of prostate-specific antigen (PSA) and histological differentiation, which is conventionally reported as the Gleason score (17,18).Approximately 80% of men diagnosed with prostate cancer have moderately increased serum levels of PSA (3-10 ng/mL) and a non-palpable localized tumor with a Gleason score of 6 or 7 (GS 6 or 7) (19,20).For prostatic adenocarcinoma with a GS 6, the clinical course is often unpredictable.Hence, MVD as a possible independent prognostic factor could be of great value.

MATERIALS AND METHODS
Thirty tissue specimens of radical prostatectomy were re-examined.Twenty-six were considered suitable for the study (paraffin blocks intact, enough material for re-cutting, basal clinical and follow-up data complete).The median age of patients at diagnosis was 64 (range 53-71) years.Specimens were divided in two groups: GS 6 (13 cases) and GS 7 (13 cases).All cases were stage pT2.Paraffin-embedded biopsy tissue blocks were cut into 4-µm sections, deparaffinized and rehydrated.Antigenic recovery was achieved by heating the slides in an autoclave with sodium citrate buffer (30 min).Endogenous peroxidase was inhibited with a Peroxidase Block Kit (Novocastra Laboratories, Newcastle upon Tyne, UK).Immunohistochemical staining was undertaken using primary antibodies against endoglin (1:50 dilution; Novocastra Laboratories).A Novolink Polymer Detection System (Novocastra Laboratories) was used for visualization.Primary antibodies were omitted in negative controls.As positive controls, sections of tonsil tissue were used.Tissue sections were counterstained using Mayer's hematoxylin and mounted.Immunoreactivity was evaluated without knowledge of patient data.After scanning the immunostained section at low magnification (×40), three areas of maximal angiogenesis ("hotspots") within the tumor were identified.Then, microvessels were counted at ×400 magnification (0.19 mm 2 field).Any single cell or spot stained by the immunohistochemical marker was counted as a vessel.As in previous reports (5,7,13), a visible vascular lumen was not required to count as a microvessel.The highest number of vessels counted in any hotspot was recorded (MVD per field).Then the mean vascular count per mm 2 was calculated (MVD per mm 2 ).Both values were used in the statistical analysis.Groups were compared with Student's t-test for independent samples.Correlations of MVD and preoperative PSA level were calculated using Pearson's correlation test.P<0.05 was considered significant.Statistical analyses were carried out using SPSS ver19 (SPSS, Chicago, IL, USA).

RESULTS
Endoglin expression in specimens with GS 6 and GS 7 is shown in Figures 1 and 2, respectively.The group of specimens with GS 6 had lower MVD per field than the group with GS 7 (19.9 vs. 25.6;P=0.05; Table 1), but this difference was not significant.The same was true when MVD per mm 2 was compared between the two groups (89.7 vs. 117.7;P=0.05; Table 1).The preoperative serum level of PSA was 1.8-8.8ng/mL in the GS 6 group (median, 5.6 ng/mL), and 2.8-34.4ng/mL in the GS 7 group (median, 5.9 ng/ mL; Figure 3).The mean PSA level in serum was not significantly different in GS 6 group compared with the GS 7 group (5.7 vs. 10.2 ng/mL; P=0.10; Table 1

DISCUSSION
Angiogenesis enables tumor growth and metastasis by providing nutrients and oxygen for metabolism as well as removal of the resultant waste products.Angiogenesis initially develops by incorporating existing blood vessels, but solid tumors cannot probably grow >1 mm 3 unless they synthesize their own network of morphologically and functionally primitive and abnormal microvessels (i.e., neovascularization) (14,(21)(22)(23).The association of increasing tumor vascularity with various measures of tumor aggressiveness (such as a greater incidence of metastases and/or reduced patient survival) has been shown in studies of patients with various types of carcinoma (4,13,14,24,25).
In the present study, we investigated angiogenesis in specimens with GS 6 and GS 7, currently the most commonly assigned Gleason scores in prostatic adenocarcinomas (18).Preliminary results showed lower MVD in GS 6 specimens than in GS 7 specimens even  Laboratorijska študija / Laboratory study though the difference was not significant.Some authors have shown a correlation between Gleason score and MVD (5-7, 15, 26, 27), whereas other authors did not (28)(29)(30).Such discrepancies may be because details of the methodology used to assay MVD, such as the choice of the antibody (e.g., CD31, CD34, von Willebrand factor (vWF), endoglin) have been reported to influence study outcome (31).It was shown that CD31, CD34 and vWF do not stain all microvessels, and particularly not newly formed microvessels (5,7,26).Nevertheless several authors used these antibodies (3,6,15,27,28,30).We used endoglin, which was consistently present in all cases and which stained microvessels in and around the tumor but showed weak or no staining for blood vessels in non-neoplastic tissue.Studies also differ with regard to the quantification of angiogenesis.Most authors examined areas of maximal angiogenesis (hotspots) at ×200 magnification (5,6,15,20,26,27).Only a few reports determined MVD at ×400 magnification (3,7,30).In the present study, we evaluated angiogenesis at ×400 magnification, which allowed more precise quantification of the number of vessels than if we had evaluated MVD at ×200 magnification.Furthermore, in statistical calculations we used two series of data for each specimen: MVD per field and MVD per mm 2 .The difference in MVD between GS 6 and GS 7 specimens in our preliminary report was of borderline statistical significance (P=0.05).This was attributable (at least in part) to the small sample size.Therefore, we expect that the final results of our future large-scale study will provide more information about the stage of vascularity in both groups of prostatic adenocarcinoma.
A correlation between MVD and serum levels of PSA was not observed in the present study.This finding is in agreement with those in other reports (26,27,29,30).Conversely, several studies lack information about the association between these two parameters (3,5,6,15,20,28).Furthermore, a significant difference was not shown when serum levels of PSA between the two groups of patients in the present study were compared.
One reason for this is the degree of dispersion of the data, especially in the GS 7 group.PSA is a key variable in prognostic models for clinically localized prostate cancer.It is used to assess pathologic tumor stage and the risk of disease recurrence after local therapy.However, elevation of serum levels of PSA do not solely reflect the presence of cancer, but may also be driven by certain non-malignant causes such as nodular hyperplastic changes in the prostate gland, and prostatic inflammatory processes (32).This apparent lack of specificity limits the application of PSA for early detection.However, there is a strong evidence that a cutoff point of 4.0 ng/mL may lead to missing a significant number of cancers (32,33).Consequently, other PSAbased strategies are being tested for clinical use.These include PSA density (ratio of an individual serum PSA and its corresponding prostate volume as assessed by transrectal ultrasonography), percent free PSA (%fPSA; calculated from analyzed free PSA and total PSA) and complexed PSA (bound to plasma proteins).Of these, only %fPSA is already used in the clinic (32).Furthermore, several alternative biomarkers (cell cycle, invasion, cell adhesion, signal transduction, apoptosis, angiogenesis, genetic) have been suggested to supplement (or even replace) PSA to improve strategies for early detection and predict the natural behavior of the tumor (32,34,35).Angiogenesis markers indicate that MVD has already been found to be a prognostic factor in several malignancies.However, there remain controversies in prostatic adenocarcinoma due to the small number of such studies (5,7,20,27,30).Therefore, future studies with large numbers of specimens and a precise methodology of evaluating angiogenesis are needed.If MVD is proved to be an independent prognostic factor in prostatic adenocarcinomas, we believe that it will help to determine which patients may require aggressive adjuvant therapy because of being at high risk for carcinoma recurrence and death.
Presented as a poster at the International Symposium of Clinical and Applied Anatomy, Maribor, Slovenia, July 2011

Figure 4 .Figure 2 .
Figure 4. Correlation between PSA level and MVD per mm 2 (r = 0,20; P = 0,30) Our preliminary results suggest a possible difference in the neovascularization between the two groups of specimens, but a difference in PSA level was not proved.Nevertheless, a future large-scale study is needed to confirm these findings.