MIB-1, Her2/neu veya CD34 testisin evre II ve III non-seminomatöz germ hücreli tümörlerinde kullanışlı prognostik faktörler midir?

PDF Dosyası: 

arastirmax_2092_pp_196-200.pdf

Introduction
Currently, testicular cancer is considered as a curable cancer and durable remission may be achieved
with conventional chemotherapies, even in advanced
stages. The survival rate is reported between 55% and
75% in those patients (1). Although conventional or
salvage chemotherapies are applied, some patients do
not achieve a remission and die due to the progression of their disease (2).
Up to date many prognostic factors have been investigated to define a correlation between the prognosis and survival. Among those factors, MIB-1 which is a nuclear antigen expressed in all phases of the
cell cycle except G0 (3), is a monoclonal antibody
and has been used to assess the proliferative activity
in human tumors (4). It has been found in conformance with the prognosis in some human solid tumors, such as breast cancer (5), colorectal cancer (6)
or prostate cancer (7). It has also been shown that
MIB-1 is a significant marker to detect the low risk
group of patients prone to development of metastasis (8). On the other hand HER2/neu, which is a
membrane glycoprotein and belongs to the epidermal growth factor receptor (EGFR) family has been
shown to have a crucial role in angiogenesis (9). The
CD34 antigen is a transmembrane glycoprotein and
is expressed on lymphohematopoietic progenitor cells, vascular endothelial cells, fibroblastic spindle cells
and some cells in the neural tissue (10). CD34 reflects
the vascular density of tumor and has been shown as
a blood vessel marker (11). Although many studies
have been performed concerning the expression of
the CD34 antigen in hematopoietic, soft tissue and
skin tumors, there are limited numbers of reports in
regard to its expression in neoplasms of other organ
systems (12,13). Two studies have reported that yolk
sac tumor is the only germ cell tumor component
positive for the expression of CD34 antigen (14,15).
Therefore, the results of current data are somewhat
* Department of Medical Oncology, Gulhane Military Medical Faculty
** Department of Epidemiology, Gulhane Military Medical Faculty
*** Department of Pathology, Gulhane Military Medical Faculty
Reprint request: Dr. Selmin Ataergin, Department of Medical Oncology,
Gulhane Military Medical Faculty, Etlik-06018, Ankara, Turkey
E-mail: sataergin@superonline.com
Date submitted: April 28, 2008 • Date accepted: October 10, 2008Volume 50 • Issue 3 MIB-1, Her2/neu and CD34 in testicular cancer • 197
still conflicting and no enough data are present yet, to
show if MIB-1, HER2/neu or CD34 are good prognostic factors in testicular germ-cell tumors (8,15,16).
In the present study, we investigated the immunostaining pattern of MIB-1, HER2/neu and CD34 in
clinical stage II and III testicular non-seminomatous
germ-cell tumors and determined whether these factors may predict the outcome of the disease in advanced stage tumor.
Material and Methods
Patient characteristics: Patients’ characteristics are
detailed in Table I. A total of 64 patients (19-40 years
old) with initial stage II and III testicular germ cell
tumors were retrospectively analyzed. Patients with
seminomas, extragonadal germ cell tumors or other
malignancies were excluded from the study. Clinical
staging was done according to AJCC Cancer Staging
(17) with computed tomography of the chest, abdomen and pelvis at initial presentation of the patients.
Serum AFP and β-HCG levels, initial localization sites of tumor, metastasis sites (lung versus other than
lung) were also noted for each patient. By this way, it
was noted that 38 patients had stage II disease and 26
patients had stage III disease.
Histopathological evaluation: All archival tissue
blocks from each tumor were initially checked by
hematoxylin and eosin-stained sections to select the
representative block with available tissue for immunohistochemical staining. A 4-μm thick section from
each formalin-fixed paraffin-embedded tumor was
stained using the primary antibody against CD34
(1:100; Neomarkers, Fremont, CA, USA), c-erbB-2 /
HER2/neu (1:200; Neomarkers, Fremont, CA, USA)
and Ki-67 (1:100; Clone SP6, Neomarkers, Fremont,
CA, USA). Immunohistochemical staining was performed by the labeled streptavidin-biotin method using
UltraVision Large Volume Detection System (Cat #
TP-060-HL, LabVision, Fremont, CA, USA) kit using
autostainer (Labvision, Fremont, CA, USA) according
to procedure briefly described below.
Immunohistochemical staining procedure was
performed as follows: sections were deparaffinized
and rehydrated in graded ethanols. After rinsing in
distilled water, sections were microwaved for 5 minutes at 750 watts in 0.01 mol/L sodium citrate buffer (pH 6.0); this step was repeated three times. The
slides were immersed in 3% H2
O2
in distilled water
for 5 minutes and then in blocking solution for 30
minutes to block endogenous peroxidase activity and
unspecific binding sites, respectively. Sections were
then rinsed in phosphate-buffered saline (PBS) and
incubated at room temperature with the primary
antibody for 60 minutes, followed by a rinse in PBS.
Omitting the primary antibody performed negative
controls. The sections were thereafter treated with biotinylated secondary anti-rabbit antibodies in a dilution of 1:200, and antibody-binding sites were finally
visualized by avidin-biotin peroxidase complex solution, using AEC as a chromogen.
Immunohistochemically stained slides were reviewed by two pathologists. The percentage of stained
cells and staining intensity were evaluated for HER2/
neu expression. MIB-1 labeling index was measured
morphometrically using a computer system composed of a personal computer, a light microscope with
motorized stage (Zeiss Axioscope, Zeiss, Göttingen,
Germany), a frame grabber card (Matrox Meteor), a
digital camera attached to the microscope (Sony AVT
Horn 3 CCD). Using KS 400 software, a semiautomatic, interactive “macro” (a batch file containing all of
the instructions to accomplish a predetermined set of
measurements) was prepared. During measurement,
a section of tumor immunostained with MIB-1 was
examined under ×2.5 magnification. At this stage, a
suitable quadrangular area was selected interactively
with the aid of the tools of the software. Then, the
entire area selected under ×2.5 magnification was
examined with an ×20 objective as consecutive nonoverlapping fields. Measurement cycles were completed in the following order: acquiring the image,
adjusting focus, superimposing randomly oriented
grids and dots on top of the image, counting dots
over stained nuclei and non-staiend tumor nuclei,
respectively, and acquiring image of the next field.
After completing the last of the preselected fields, the
Table I. Patient characteristics
Number Percentage (%) Total number
Age
< 20
20-30
>30
5
53
6
8
83
9
64
Stage
II
III
38
26
60
40
Occurrence of relapse
Stage II
Stage III
9
14
15
24
23
Location of relapse
Lung
Other than lung
9
14
15
24
Number of death
Stage II
Stage III
5
12
9
21198 • September 2008 • Gulhane Med J Ataergin et al.
dot counts summed up by the software and the results were shown in a message window as percent of
MIB-1 labeled nuclei in the tumor. These results were
recorded for each sample.
CD-34 immunostained slides were assesed for microvessel density. Briefly, the tumor section was scanned at low power, and subjectively the three areas
with the highest number of discrete microvessel profiles were selected (hot-spots). One microscopic field
(Leica DMLB; Leica, Wetzler, Germany) was identified within each hot-spot at x200 magnification providing a 0.754 mm
2
field size. All individual microvessel profiles within the applied circular microscopic field were counted. MVD was defined as the number of
manually counted vessel profiles per mm
2
(vp/mm
2
)
taken as the average from the three hot-spot counts.
Then, they are categorized into three group: (+), <50;
(++), 50-70; (+++), ≥70.
Statistics: All values in the text were given as median
(min-max) and percents. The differences between the
groups were tested using Kruskall Wallis and MannWhitney U tests. The relationships between the variables were detected by Spearman’s correlation test.
The survivals of patients were estimated by Kaplan
Meier method. Log rank test was used to compare the
survivals of groups. Cox proportional hazard analysis
was used to determine the prognostic factors in univariate and multivariate analysis. All statistical calculations were performed with SPSS 10.0 for Windows
statistical software package (Chicago, IL, USA). P value < 0.05 was considered statistically significant in
all analysis.
Results
Patients and HER2/neu, CD34 and MIB-1 immunostaining findings: Sixty percent of patients had stage II
disease and 40% of patients had stage III disease. All
pathologic specimens were stained for MIB-1 by immunohistochemistry. Median level of expression of
MIB-1 was 32 (range: 0-55). MIB-1 level was under
32 in 18 patients (47%) with stage II disease and 8
patients (30%) with stage III disease; the level was ≥
32 in 20 patients (53%) with stage II disease and in
18 patients (70%) with stage III disease (Table II). The
median level of expression of MIB-1 was 30 in stage II
and 34 in stage III tumors. The difference was found
significant statistically (p: 0.018). CD34 immunostaining was positive in all specimens. The level of CD34
immunostaining was (+) in 12 patients (31%), (++) in
15 patients (4%), and (+++) in 11 patients (29%) in
stage II tumors. This level was (++) in 7 patients (26%)
and (+++) in 19 patients (73%) (Table II). None of these specimens were stained with HER2/neu.
Uni and multivariate cox proportional hazard modeling
results: The expression level of MIB-1 was not found
as a significant prognostic factor statistically at a median level of 32 (OR 1.20, 95% CI (0.47-3.04), p=0.70)
(Table III). Similarly, CD34 expression was not shown
as an important factor for prognosis statistically (OR
31.8, 95%CI (0.32-3127.8), p=0.14) in univariate
analysis. Multivariate analysis was not performed due
to the lack of significance in univariate analysis.
Discussion
High proliferation indices are reported as a risk factor for relapse in testicular germ-cell tumors (18). DNA
flow-cytometry (19), DNA image cytometry (20) and
proliferating cell nuclear antigen (PCNA) were used
to determine tumor proliferation rate (21). However,
the analysis of MIB-1 or HER2/neu by immunohistochemistry is considered as a more useful and less
expensive method in some studies to give information about the cell proliferation activity in testicular
germ-cell tumors (8,22-25).
Albers et al. investigated 76 patients with stage
I germ cell tumors and emphasized that MIB-1 is a
good prognostic parameter and that it could be used
even to classify patients with low metastatic risk due
Table II. Distribution of the patients according to the evaluated prognostic factors
Stage II (n) Stage II (%) Stage III (n) Stage III (%) Total (n) Total (%)
MIB-1
<32§
≥32§
18
20
47
53
8
18
30
70
26
38
40
60
HER2/neu (-) 38 100 26 100 64 100
CD34
(+)
(++)
(+++)
12
15
11
31
40
29
0
7
19
0
26
73
12
22
30
19
34
47
§: Median valueVolume 50 • Issue 3 MIB-1, Her2/neu and CD34 in testicular cancer • 199
Table III. Univariate cox hazard proportional analysis result
Prognostic
factor
Hazard ratio
95% Confidence
interval
p value
MIB-1 1.20 0.47-3.04 0.70
CD34 31.8 0.32-3127.8 0.14
HER2/neu NA NA NA
to high expression level in metastatic cases (26). On
the other hand, Heidenreich et al. and Data et al. failed to show that MIB-1 expression did correlate with
pathologic stage in NSGCTs (16,23,27). In our study,
we detected MIB-1 expression in all specimens. The
median level of MIB-1 expression was found significant statistically according to stages. However, at a
median level of expression, we found that MIB-1 was
not a significant prognostic factor in univariate analysis. In various studies, investigators have selected different cut-off value for MIB-1, which made difficult
the interpretation on risk factor analysis (13). The
cut-off value of Albers et al. was 70% (8). We selected
the mean value 32% as a cut-off value. The difference
in results between the Albers’ and our study may be
related to the cut-off value selected (28).
CD34 has been shown as a blood vessel marker in
various tumors and is used to identify the vascular
density of the tumor (11). It has been shown that
quantitative evaluation of angiogenesis by CD34
immunohistochemistry may be helpful to differentiate hepatocellular carcinoma (HCC) and non-HCC.
However, the authors emphasized that no prognostic
information may be provided by the staining of CD34
(29,30). Similarly to these reports, in our study, all
specimens were stained positively with CD34 and the
intensity of staining was ranging from (+) to (+++).
However, the expression level was not found significant statistically between stage II and III tumors
(p>0.05), and the intensity of immunostaining with
CD34 was found as a significant prognostic factor to
predict the survival in stage II and III tumors.
Finally, although HER2/neu is among EGFR family,
in our study none of our germ cell tumor specimens
were stained positively with HER2/neu by immunohistochemistry. The role of other subtypes of EGFR
family in predicting the survival of germ cell tumors
needs to be investigated.
In conclusion, our study revealed that MIB-1, HER2/
neu or CD34 are not useful as a prognosticator to predict the survival in patients with stage II and III testicular germ cell tumors. The intensity of CD34 and
MIB-1 immunostaining increased with advancing
stage. The role of these markers in clinical stage and
prognostic evaluation needs further investigation.