close

Вход

Забыли?

вход по аккаунту

?

771

код для вставкиСкачать
520
Tumor Angiogenesis, Hepatocyte Growth Factor, and
c-Met Expression in Endometrial Carcinoma
Satoshige Wagatsuma,
Ryo Konno, M.D.
Shinji Sato, M.D.
Akira Yajima, M.D.
BACKGROUND. This study was designed to evaluate the significance of tumor angio-
M.D.
genesis and angiogenic factors such as hepatocyte growth factor (HGF) and c-Met
in determining the prognoses of 93 patients with endometrial carcinoma.
METHODS. By immunohistochemical staining, this retrospective study investigated
tumor angiogenesis, HGF expression, and c-Met expression, using one tissue slide
Department of Obstetrics and Gynecology, Tohoku University School of Medicine, Sendai, Miyagi, Japan.
that was representative of the invasive edge of the tumor. To evaluate tumor
angiogenesis, the microvessels within the primary endometrial carcinoma were
highlighted by staining their endothelial cells immunohistochemically for von Willebrand factor (VWF). The microvessels were then counted in the most intense
areas of neovascularization. HGF and c-Met were identified with specific antibodies. Tumor angiogenesis, HGF expression, and c-Met expression were correlated
with both the prognostic variables for and the survival of endometrial carcinoma.
RESULTS. A high microvessel count (�0 in a 0.90 mm2 area) was significantly
correlated with surgical Stage III and IV, histologic Grade 3, positive lymph node
involvement, and shorter patient survival. Expression of c-Met was significantly
correlated with surgical Stage III and IV, histologic Grade 3, and shorter survival.
HGF expression was significantly correlated with surgical Stage III and IV by semiquantitative analysis. Multivariate analysis showed that surgical Stage III and IV,
histologic Grade 3, the score for myometrial invasion �2, and a high microvessel
count were independent indicators of the prognoses of patients with endometrial
carcinoma.
CONCLUSIONS. Both tumor angiogenesis, measured by the microvessel count, and
c-Met expression were significant prognostic indicators for patients with endometrial carcinoma. Cancer 1998;82:520?30. q 1998 American Cancer Society.
KEYWORDS: tumor angiogenesis, hepatocyte growth factor, c-Met, endometrial carcinoma.
F
Presented at the 49th Annual Congress of the
Japan Society of Obstetrics and Gynecology,
Tokyo, Japan, April 5?8, 1997.
Address for reprints: Satoshige Wagatsuma,
M.D., Department of Obstetrics and Gynecology, Tohoku University School of Medicine, 11 Seiryo-machi, Aoba-ku, Sendai 980, Japan.
Received February 6, 1997; revisions received
May 27, 1997, and August 11, 1997; accepted
August 11, 1997.
olkman et al.1,2 have suggested that the growth and metastases of
many solid tumors require a switch from a prevascular to a vascular
phase. The prevascular phase is characterized by limited tumor
growth and little or no metastasis, while the vascular phase is characterized by rapid tumor growth and an increased incidence of metastasis. Angiogenesis thus contributes significantly to the progression of
carcinoma. Angiogenesis involves several sequential steps, including
the degradation of the basement membrane of the parent vessel, the
movement of endothelial cells toward a tumor implant, mitosis, lumen formation, the development of both sprout loops and a new
basement membrane, and, finally, the recruitment of pericytes into
the tissue.3
It is thought that an intratumoral microvessel count serves as an
index of the overall steps of angiogenesis. The microvessel count in
malignant tumors has recently been quantified by the immunohisto-
q 1998 American Cancer Society
/ 7ba8$$0201
01-13-98 09:28:49
cana
W: Cancer
Angiogenesis in Endometrial Carcinoma/Wagatsuma et al.
chemical staining of the endothelial cells in such cancers as breast carcinoma,4 ? 8 lung carcinoma,9,10 head
and neck carcinoma,11,12 endometrial carcinoma,13 and
ovarian carcinoma.14 von Willebrand factor (VWF ), or
factor VIII-related antigen, is most frequently used for
the immunohistochemical staining of endothelial
cells. This antigen is a multimeric protein that is synthesized by the endothelial cells lining blood vessels
of all sizes.
The induction of angiogenesis may be mediated
by specific angiogenic molecules released by the tumor and by macrophages that are attracted to it.15,16
Acidic and basic fibroblast growth factor17,18 and vascular endothelial growth factor19 are considered to be
the primary factors that stimulate angiogenesis. The
numerous other factors described include transforming growth factor-a, platelet-derived endothelial
cell growth factor, vascular permeability factor, folliculistellate-derived growth factor, angiotropin, angiogenin, transforming growth factor-b, tumor necrosis
factor-a, and urokinase-type plasminogen activator.2,15,16,20,21
Recent studies indicate that hepatocyte growth
factor (HGF ) is also a powerful inducer of angiogenesis.22,23 HGF was first discovered as a mitogen for adult
hepatocytes.24,25 Later studies revealed that it was
identical to the scatter factor.26 HGF is a heparin-binding glycoprotein that consists of a 60-kD a-chain and
30-kD b-chain linked by disulfide bonds.27 It belongs
to the family of kringle proteins,25 which are characterized by triple disulfide loop structures that mediate
protein-protein and protein-cell interactions. It is
known that HGF has mitogenic, motogenic, and morphogenic functions in vitro on various epithelial cells
derived from rodents and humans.28,29 The HGF receptor is the c-met proto-oncogene product (c-Met)30 and
is a transmembrane tyrosine kinase, a 190-kD glycoprotein consisting of a 145-kD membrane-spanning
b-chain and a 50-kD a-chain. It is thought that both
HGF and c-Met activation and/or expression contribute to embryogenesis, organ regeneration, angiogenesis, and neoplasia.31
The objective of the current study was to determine whether a correlation exists between the prognosis for endometrial carcinoma and the microvessel
count and/or expression of HGF and/or c-Met in tumor tissue. These relationships were investigated in
93 patients with endometrial carcinoma.
PATIENTS AND METHODS
Patients
The retrospective study included 93 Japanese patients
with surgically operable endometrial carcinoma who
underwent surgery between 1991 and 1995 at the To-
/ 7ba8$$0201
01-13-98 09:28:49
cana
521
hoku University Hospital, Sendai, Japan, and affiliated
hospitals. All patients included in this study had endometroid adenocarcinoma. None of the carcinomas
were of the adenosquamous, serous papillary, or clear
cell type. The median age of the patients was 56 years
(range, 29 ? 79 years). The surgical stage of disease was
assigned, following surgery, according to the criteria
outlined in the staging system of the International
Federation of Gynecology and Obstetrics (FIGO).32 The
patients included 65 in Stages I / II and 28 in Stages III
/ IV. The histologic grades of the tumors, determined
according to the FIGO criteria, was as follows: G1, welldifferentiated (n � 44); G2, moderately differentiated
(n � 29); or G3, poorly differentiated (n � 20). Lymph
node involvement was observed in 22 of the 93 cases.
The depth of the myometrial invasion was classified
as tumor invasion within the inner half (n � 64) or
the outer half (n � 29) of the uterine wall. Vascular
involvement was observed in 46 cases. An additional
14 Japanese patients diagnosed with uterine myomas
and who had normal endometrium provided tissue
that was examined for both HGF and c-Met expression.
All 93 patients in this study underwent laparotomy
combined with total abdominal hysterectomy, bilateral salpingo-oophorectomy, pelvic lymphadenectomy, and peritoneal washing for cytology. If a metastatic tumor was present, it was resected as fully as
possible. Adjuvant combination chemotherapy consisted of cyclophosphamide, doxorubicin, and cisplatin, and/or radiation therapy followed the surgery,
if the surgical stage was greater than Ic, if the histologic
grade was poorly differentiated, or if vascular involvement was observed.
Microvessel Staining and Counting
Specimens were fixed immediately in 4 % paraformaldehyde (pH 7.4) for 18 hours at 4 7C and subsequently
embedded in paraffin. The specimens were stained
with hematoxylin and eosin, then used to select areas
that were representative of the invasive tumor. One
representative section per tumor was stained in order
to count the number of microvessels. All microvessels
were highlighted by staining their endothelial cells for
VWF with a standard immunoperoxidase staining
technique using the avidin-biotin method (Histofine;
Nichirei Co. Ltd., Tokyo, Japan) as follows:
Sections 3 mm thick were cut from paraffin-embedded blocks and mounted on silanized slides. The
sections were then deparaffinized and incubated in
methanol containing 0.3 % hydrogen peroxide at room
temperature for 15 minutes to block endogenous peroxidase activity. The slides were then incubated in 0.1
% trypsin (Sigma Co. Ltd., St. Louis, MO) for 15 minutes for antigen retrieval. After treatment with 1% nor-
W: Cancer
522
CANCER February 1, 1998 / Volume 82 / Number 3
mal rabbit serum at room temperature for 30 minutes,
the slides were next incubated at 4 7C overnight with
antihuman VWF (mouse monoclonal; Dako A/S,
Glostrup, Denmark) diluted to l:30. This was followed
by treatment with biotinylated antimouse immunoglobulin at room temperature for 30 minutes, after
which the sections were incubated with peroxidaseconjugated streptavidin at room temperature for 30
minutes. After a final wash, the sections were immersed for 15 minutes in a solution containing 0.06
mM 3.3-diamino-benzidine (DAB) plus 2 mM hydrogen peroxide in 0.05 % Tris-HCl buffered at pH 7.6. The
sections were finally counterstained with hematoxylin.
Microvessel staining was satisfactory and counting
was possible in all cases.
Microvessels were counted by an investigator who
had no knowledge of the patients? disease stage.
Counts were determined by light microscopy of those
areas of the invasive tumor that contained the highest
numbers of capillaries and small venules. These vascular ??hot-spots?? were first detected by scanning the
tumor sections at low power (140 and 1100) and then
identifying those areas of the invasive carcinoma having the greatest number of distinct VWF-stained microvessels per area. Three discrete areas of these hotspots were selected from one section per tumor. Microvessel counts were performed with an image
analyzer (Luzex F; Nireco, Tokyo, Japan) on each area
at a 1100 field (110 objective and 110 ocular). The
frame area for each captured image was 0.90 mm2.
Any endothelial cells or endothelial clusters that were
immunopositive and were clearly separate from the
adjacent microvessels were considered to be a single
microvessel. Red cells did not have to be present in a
vessel lumen for the structure to be identified as a
microvessel. Results are expressed as the highest number of microvessels identified within any three discrete
areas of hot-spots.
Immunohistochemical Staining of HGF and c-Met
Immunohistochemical staining for HGF and c-Met expression was performed with either anti-HGF a (rabbit
polyclonal; IBL, Gumma, Japan) or anti-c-Met (rabbit
polyclonal; IBL, Gumma, Japan) on 3 mm sections of
paraffin-embedded blocks as follows. In brief, after
antigen retrieval by 0.1 % trypsin, the sections were
incubated with 1% normal goat serum at room temperature for 30 minutes. The slides were next incubated overnight at 4 7C with either anti-HGF a or antic-Met at a dilution of l:20 (5 mg/mL). Treatment with
biotinylated antirabbit immunoglobulin at room temperature for 30 minutes followed. The other steps were
the same as those for microvessel staining, except that
the time for immersion in DAB was 5 minutes. For
/ 7ba8$$0201
01-13-98 09:28:49
cana
absorption test of immunoreactivity for HGF and cMet, the mixture of HGF recombinant peptide (Becton
Dickinson Labware, Bedford, MA) and anti-HGF a, or
c-Met recombinant peptide (IBL, Gumma, Japan) and
anti-c-Met was incubated overnight at 4 7C. After centrifugation, the resultant supernatants were used as
preabsorbed antibodies. Placental tissue, which is
known to exhibit high levels of HGF and c-Met protein,
was used as a positive control. Slides stained with the
primary antibodies omitted were used as negative controls. The evaluation of immunohistochemical staining of HGF and c-Met was performed without knowledge of the clinical outcome. The tumor samples were
considered to show diffuse staining when more than
one-third of the gland or cancer cells were positively
immunostained for HGF or c-Met, and to show focal
staining when less than one-third of the gland or cancer cells positively immunostained for HGF or c-Met.
This categorization was based on a previously published study.33 Furthermore, HGF expression was correlated with surgical stage of the tumor by semiquantitative analysis as follows: Score 0, no positively stained
tumor cells detected; Score 1, less than one-third positively stained tumor cells; Score 2, one-third to twothirds positive tumor cells; and Score 3, more than
two-thirds positively stained tumor cells.
Statistical Analysis
Results were analyzed using the StatView 4.5 software
(Abacus Concepts, Inc., Berkeley, CA) or the SPSS-X
statistical computer package (SPSS Inc., Chicago, IL).
For contingency tables we used the chi-square test.
Continuous variables were analyzed with the MannWhitney U test. Survival curves were constructed by
the Kaplan-Meier method and statistically analyzed
with the log rank test. The prognostic role of the variables on survival was evaluated with the Cox regression model, by both univariate and multivariate analyses. The variables found to be statistically significant in
the univariate analysis were subjected to multivariate
analysis. A level of P � 0.05 was considered to be
statistically significant.
RESULTS
As of October 31, 1996, the median duration of followup was 27 months (range, 2 ? 61 months). A total of 21
of the 93 patients had died from the recurrence of
disease. The median time to death was 18 months
(range, 2 ? 61 months). The remaining 64 patients were
disease free at this time, whereas 8 patients had been
lost to follow-up.
The anti-VWF antibody showed intense staining
of the endothelial cells of the microvessels, and background staining was very low. Although hot-spots
W: Cancer
Angiogenesis in Endometrial Carcinoma/Wagatsuma et al.
523
TABLE 1
Distribution of Microvessel Count in Patients with Endometrial Carcinoma
MVC
Characteristic
Age (yrs)
�
�
Surgical stage
I / II
III / IV
Histologic grade
G1 / 2
G3
Lymph node involvement
Negative
Positive
Myometrial invasion
�2
�2
Vascular involvement
Negative
Positive
No. of patients
Median
Range
P valuea
21
72
84
116
35?194
31?250
0.0866
65
28
97
127
31?194
62?250
0.0039
73
20
102
128
31?204
51?250
0.0222
71
22
97
128
31?220
80?250
0.0020
64
29
111
116
31?204
52?250
0.6453
47
46
100
118
31?194
50?250
0.0852
MVC: microvessel count (per 0.90 mm2).
a
Mann-Whitney U test.
FIGURE 1.
Immunohistochemical staining with VWF highlighting microvessels in myometrium of a patient with endometrial carcinoma. The
vascular hot-spot was observed at the margin of the adenocarcinoma (left)
(1250).
could occur anywhere within the carcinoma tissue,
most were observed at the margins of the carcinoma
(Fig. 1). The microvessel count showed marked intratumor and intertumor heterogeneity: for the total population, the median count was 111 (range, 31 ? 250),
and the mean was 114.7 { 44.0 ({ standard deviation).
/ 7ba8$$0201
01-13-98 09:28:49
cana
The distribution of the microvessel count across the
variables is shown in Table 1. The patients were stratified into high (�0) and low (�0) microvessel
counts on the basis of the median value, neither of
which were significantly correlated with age, degree
of myometrial invasion, or vascular involvement.
However, the counts were significantly correlated with
the surgical stage and histologic grade of the tumor,
and with lymph node involvement (Table 2). A significant correlation was also observed between a high microvessel count group (�0) and a decrease in the
length of survival (P � 0.0003) (Fig. 2).
The immunoreactivities for HGF and c-Met were
evaluated in the normal endometrium and in that of
the endometrial carcinoma. A positive immunoreactivity for HGF was observed in both the cytoplasm and
the nuclei of the adenocarcinoma cells (Fig. 3), the
uterine smooth muscle cells, and the vascular smooth
muscle cells. Normal endometrial gland cells were
also stained positively for HGF. The immunoreaction
was abolished when the antibody was preincubated
with the HGF recombinant peptide. Diffuse staining
for HGF was noted in the majority of cases with a
normal endometrium and endometrial carcinoma (Table 3). There were no differences in staining intensity
of HGF in cancer cells as compared with normal endometrial gland cells. Conversely, a positive immunoreactivity for c-Met was observed in both the cell mem-
W: Cancer
524
CANCER February 1, 1998 / Volume 82 / Number 3
TABLE 2
Clinicopathologic Correlation of Microvessel Count in Patients with Endometrial Carcinoma
No. of patients (%)
Characteristic
Age (yrs)
�
�
Surgical stage
I / II
III / IV
Histologic grade
G1 / 2
G3
Lymph node involvement
Negative
Positive
Myometrial invasion
�2
�2
Vascular involvement
Negative
Positive
MVC � 110
MVC � 110
P valuea
13 (61.9)
31 (43.1)
8 (38.1)
41 (56.9)
0.203
37 (56.9)
7 (25.0)
28 (43.1)
21 (75.0)
0.009
40 (54.8)
4 (20.0)
33 (45.2)
16 (80.0)
0.012
40 (56.3)
4 (18.2)
31 (43.7)
18 (81.8)
0.004
31 (48.4)
13 (44.8)
33 (51.6)
16 (55.2)
0.921
26 (55.3)
18 (39.1)
21 (44.7)
28 (60.9)
0.175
MVC: microvessel count (per 0.90 mm2).
a
Chi-square test.
FIGURE 2. Actuarial survival curves for patients with endometrial carcinoma stratified by a microvessel count of fewer (single line) and greater
than (double line)110 microvessels (per 0.90 mm2).
brane and the cytoplasm of the adenocarcinoma cells
(Fig. 4), as well as in that of the normal endometrial
gland cells. The immunoreaction was abolished when
the antibody was preincubated with the c-Met recombinant peptide. Diffuse staining of c-Met was seen in
fewer cases showing a normal endometrium than
those with endometrial carcinoma (Table 3). Among
the 59 patients who showed diffuse staining of c-Met in
endometrial carcinoma tissue, 57 also showed diffuse
staining of HGF.
/ 7ba8$$0201
01-13-98 09:28:49
cana
The percentage of the cases that showed diffuse
staining for HGF in endometrial carcinoma tissue tended
to be higher in those patients who were in surgical Stage
III / IV, had poorly differentiated tumors, had positive
lymph node involvement, and in whom the tumor had
invaded the outer half of the uterine wall. Although the
patients with such diffuse staining of HGF appeared to
survive for a shorter time, the presence of HGF immunoreactivity was not significantly correlated with any of the
clinicopathologic findings, including survival (Table 4, Fig.
5). When the relationship between HGF expression and
the stage of the tumor was examined by semiquantitative
analysis, HGF expression was significantly correlated with
surgical Stage III / IV of the tumor (Fig. 6). The presence
of c-Met expression was significantly correlated with the
surgical stage and the histologic grade of the tumor, but
was not significantly correlated with age, lymph node
involvement, myometrial invasion, or vascular involvement (Table 4). A significant correlation was also observed
between the diffuse staining for c-Met and a decrease in
the length of survival (P � 0.0027) (Fig. 7).
The microvessel count was significantly higher in the
patients who showed diffuse staining of HGF (focal vs.
diffuse; median value [range]: 67 [31?133] vs. 116 [51?
250]; P � 0.0011) and c-Met (focal vs. diffuse; median
value [range]: 92 [31?177] vs. 122 [51?250]; P � 0.0031)
in endometrial carcinoma tissue.
Univariate analysis showed a significant association
W: Cancer
Angiogenesis in Endometrial Carcinoma/Wagatsuma et al.
525
FIGURE 3. Immunohistochemical staining of HGF in endometrial carci-
FIGURE 4. Immunohistochemical staining of c-Met in endometrial carci-
noma. (A) The HGF positive cells were observed in both the cytoplasm
noma. (A) The c-Met positive cells were observed in both the cell membrane and the cytoplasm of most adenocarcinoma cells. This showed
diffuse staining. (B) Less than one-third of the carcinoma cells positively
immunostained for c-Met (arrow). This showed focal staining (1250).
and the nuclei of most adenocarcinoma cells. This showed diffuse staining.
(B) Less than one-third of the carcinoma cells positively immunostained
for HGF (arrow). This showed focal staining (1250).
TABLE 3
Immunoreactivity of HGF and c-Met in Human Endometrium
No. of specimens showing
diffuse staining (%)
Normal endometrium
Proliferative
Secretory
Atrophic
Endometrial carcinoma
Total no. of
specimens
HGF
c-Met
14
5
4
5
93
11 (78.6)
5 (100)
1 (25.0)
5 (100)
84 (90.3)
2 (14.3)
2 (40.0)
0 (0.0)
0 (0.0)
59 (63.4)
Diffuse staining: more than one-third of the gland or cancer cells showed positive staining.
/ 7ba8$$0201
01-13-98 09:28:49
cana
between the prognosis of the patient and the surgical
stage of the tumor, the histologic grade of the tumor,
lymph node involvement, myometrial invasion, vascular
involvement, microvessel count, and the presence of cMet expression. Multivariate analysis of these prognostic
variables revealed the following to be independent prognostic indicators: the surgical stage of the tumor, the histologic grade of the tumor, the presence of myometrial
invasion, and the microvessel count (Table 5).
DISCUSSION
In addition to the variables that are known to be prognostic for endometrial carcinoma, this study showed
that both tumor angiogenesis and c-Met expression
were significant prognostic indicators for this disease.
This study also shows that tumor angiogenesis served
W: Cancer
526
CANCER February 1, 1998 / Volume 82 / Number 3
TABLE 4
Clinicopathologic Correlation of HGF and c-Met Expression in Patients with Endometrial Carcinoma
No. of patients (%)
Characteristic
Age (yrs)
�
�
Surgical stage
I / II
III / IV
Histologic grade
G1 / 2
G3
Lymph node involvement
Negative
Positive
Myometrial invasion
�2
�2
Vascular involvement
Negative
Positive
No. of patients (%)
HGF focal
HGF diffuse
P valuea
c-Met focal
c-Met diffuse
P valuea
2 (9.5)
7 (9.7)
19 (90.5)
65 (90.3)
0.695
10 (47.6)
24 (33.3)
11 (52.4)
48 (66.7)
0.347
8 (12.3)
1 (3.6)
57 (87.7)
27 (96.4)
0.355
29 (44.6)
5 (17.9)
36 (55.4)
23 (82.1)
0.026
9 (12.3)
0 (0.0)
64 (87.7)
20 (100)
0.220
33 (45.2)
1 (5.0)
40 (54.8)
19 (95.0)
0.002
9 (12.7)
0 (0.0)
62 (87.3)
22 (100)
0.179
30 (42.3)
4 (18.2)
41 (57.7)
18 (81.8)
0.073
7 (10.9)
2 (6.9)
57 (89.1)
27 (93.1)
0.817
27 (42.2)
7 (24.1)
37 (57.8)
22 (75.9)
0.149
4 (8.5)
5 (10.9)
43 (91.5)
41 (89.1)
0.973
21 (44.7)
13 (28.3)
26 (55.3)
33 (71.7)
0.153
Focal: less than one-third of the cancer cells showing positive staining; diffuse: more than one-third of the cancer cells showing positive staining.
a
Chi-square test.
FIGURE 5. Actuarial survival curves for patients with endometrial carcinoma stratified by focal (single line) and diffuse (double line) HGF expression.
FIGURE 6. Relationship between HGF expression and the stage of the
tumor by semiquantitative analysis. The score of HGF expression was
significantly higher in the patients who showed surgical Stage III / IV
(Mann-Whitney U test, P � 0.0013). Score 0: no positively stained tumor
as an independent prognostic indicator for endometrial carcinoma, the first such to report.
To evaluate tumor angiogenesis, we counted the
number of microvessels in the area of the highest neovascularization, using one tissue slide representative
of the invasive edge of the tumor. Consequently, those
patients who showed 110 or more microvessels per
1100 field (0.90 mm2) showed a significant correlation
of this variable with the surgical Stage III / IV, with a
/ 7ba8$$0201
01-13-98 09:28:49
cana
cells detected; Score 1: less than one-third positively stained tumor cells;
Score 2: one-third to two-thirds positive tumor cells; Score 3: more than
two-thirds positively stained tumor cells.
histologic Grade 3 of the tumor, positive lymph node
involvement, and short term survival. In addition,
multivariate analysis showed that the microvessel
count was an independent prognostic indicator for
endometrial carcinoma. These results are consistent
W: Cancer
Angiogenesis in Endometrial Carcinoma/Wagatsuma et al.
527
TABLE 5
Univariate and Multivariate Analysis of Prognostic Variables of Survival in Patients with Endometrial Carcinoma
Univariate analysis
Variable
Age (yrs)
� vs. �
Surgical stage
I / II vs. III / IV
Histologic grade
G1 / 2 vs. G3
Lymph node involvement
0 vs. /
Myometrial invasion
�2 vs. �2
Vascular involvement
0 vs. /
Microvessel count
�0 vs. �0
HGF
Focal vs. diffuse
c-Met
Focal vs. diffuse
Multivariate analysis
RR
95% CI
P value
RR
95% CI
P value
1.758
0.51?6.05
0.3708
?
?
?
48.866
6.53?365.49
0.0002
12.375
1.05?145.46
0.0454
3.544
1.47?8.57
0.0050
0.3.449
1.12?10.65
0.0314
14.165
4.72?42.55
�0001
1.850
0.44?7.70
0.3981
6.983
2.67?18.26
0.0001
3.370
1.13?10.02
0.0289
10.295
2.38?44.50
0.0018
1.064
0.22?5.05
0.9380
9.373
2.16?40.65
0.0028
6.311
1.26?31.63
0.0251
23.539
0.03?20055.70
0.3589
?
?
?
6.891
1.60?29.78
0.0097
2.765
0.54?14.27
0.2244
RR: Relative risk; CI: confidence interval; HGF: hepatocyte growth factor; 0: negative; /: positive; focal: less than one-third of the cancer cells showing positive staining; diffuse: more than one-third of the cancer
cells showing positive staining.
FIGURE 7. Actuarial survival curves for patients with endometrial carcinoma stratified by focal (single line) and diffuse (double line) c-Met expression.
with those showing a significant correlation between
a higher neovascularity and a poor survival for patients
with breast carcinoma,4,5,7,34 squamous cell carcinoma
of the head and neck,11,12 and lung carcinoma.9,10 In
most of the latter studies, and ours, the tumor microvessels were stained with an anti-VWF antibody
and counted in that area of the most intensive neovascularization.
Antibodies that can detect formalin-resistant en-
/ 7ba8$$0201
01-13-98 09:28:49
cana
dothelial antigens include antibodies that act against
VWF and the cell surface markers CD31 (platelet/endothelial cell adhesion molecule)35 and CD34 (human
progenitor cell antigen-1).36,37 Siitonen et al6 evaluated
the differences among these three endothelial cell antibodies in the immunohistochemical stainings of
breast carcinomas and concluded that all three could
be used as endothelial cell markers for breast carcinoma, but that anti-VWF and anti-CD34 showed a better staining performance than anti-CD31. Based on
these results, anti-VWF was used for immunostaining
in the current study.
The issue of whether a random field or the area
of most intense vascularization serves as a better indicator of angiogenesis has been discussed. Using the
method of randomly selecting 25 fields from one section in each tumor and calculating the mean values,
Goulding et al.8 examined the association between tumor vascularity and overall survival and found no correlation. However, the onset of angiogenic activity
does not require that all tumor cells become angiogenic. The highly angiogenic tumor cells are the likely
source of the growing metastatic foci.5 It is thus important to identify the area of most intensive neovascularization in determining tumor behavior.
Conversely, recent studies of patients with breast
carcinoma have shown no significant correlation between tumor vascularity and any prognostic factor or
W: Cancer
528
CANCER February 1, 1998 / Volume 82 / Number 3
overall survival.6,8 In these studies, the authors suggested that, because of the heterogeneity of vascularity
within a tumor, multiple tissue blocks per tumor must
be assessed to find the neovascular hot-spots. This
would diminish any sampling error created by identifying the hot-spots from only one slide per tumor.
However, it is likely that the specific slides chosen for
microvessel staining involved the deepest myometrial
invasion of the carcinoma, and thus represented the
malignant behavior of the tumor. It is also possible
that counting the microvessels in the myometrium of
the margins of the endometrial carcinoma avoids the
variation in the vascular field size seen in breast tissue
due to the presence of fat. Nevertheless, it will be necessary to use multiple tissue blocks in future studies
to address the problem of intratumoral heterogeneity
of vascularity.
Several recent reports have revealed that HGF
and/or c-Met contribute to the progression of other
human cancers, including ovarian carcinoma,38 renal
cell carcinoma,39 bladder carcinoma,40 breast carcinoma,41 pancreatic carcinoma,33 and prostatic carcinoma.42 In the current study of 93 patients with endometrial carcinomas, our results showed that c-Met immunoreactivity was present in a minority of the
patients with normal endometrium, and was found
more often in the patients with endometrial carcinomas. The expression of c-Met in the primary tumor as
assessed by immunohistochemical staining was also
correlated with both prognostic factors and the survival of patients with endometrial carcinoma. In addition, because almost all of the patients who showed
immunoreactivity for c-Met also demonstrated positive immunoreactivity for HGF, the co-localization of
both HGF and c-Met may be a significant factor in
the prognostication of endometrial carcinoma. This
observation would suggest that an autocrine mode of
growth stimulation exists in these tumors. HGF was
expressed in tissues from a large number of patients
with both normal endometrium and endometrial carcinoma, and there was no significant correlation between HGF expression and any prognostic factors or
survival. HGF expression, however, was significantly
correlated with surgical Stage III / IV of the tumor by
semiquantitative analysis. Considering the qualitative
nature of immunohistochemical staining, it may be
useful to investigate HGF expression by a quantitative
analysis such as Western blot or enzyme-linked immunosorbent assay, to examine the relationship between
HGF expression and the prognosis of endometrial carcinoma.
Although the major HGF-producing cell types are
mesenchymal cells, such as fibroblasts and vascular
smooth muscle cells,43,44 both HGF mRNA expression
/ 7ba8$$0201
01-13-98 09:28:49
cana
and protein synthesis of HGF have been reported in
cells from breast carcinoma and lung carcinoma.45,46
In the current study, HGF staining was found in normal endometrial gland cells, adenocarcinoma cells,
and smooth muscle cells. To determine whether the
HGF immunoreactivity detected in the normal endometrial gland cells and the adenocarcinoma cells represents the internalized HGF-c-Met complex, it may
be necessary to study the expression of HGF mRNA.
The current study was limited by its retrospective
design and short follow-up period. It was also restricted by methodologic controversy over whether the
evaluation of the angiogenesis by the ??hot-spots?? microvessel count from one representative tissue block
is appropriate. Despite these limitations, the results
suggest that tumor angiogenesis and the tissue colocalization of HGF and c-Met are important variables
in the prognosis of patients with endometrial carcinoma.
Many attempts have been made to use clinical
and pathologic parameters to arrive at a prognosis for
patients with endometrial carcinoma. Our results suggest that the microvessel count and the presence of
endometrial c-Met may be useful in identifying patients with a poor prognosis. Because endometrial carcinoma can metastasize systemically to such sites as
the lung, the liver, and the para-aortic lymph nodes,
these patients may be candidates for systemic chemotherapy as well as local radiotherapy. The evaluation
of new prognostic indicators, such as the microvessel
count and the c-Met expression, which promotes the
angiogenesis described here, may also be useful in the
induction of new anti-angiogenic drugs to supplement
conventional chemotherapy.
In summary, tumor angiogenesis and the expression of c-Met protein in endometrial tissue were significant prognostic indicators for endometrial carcinoma. Thus, determining the relative expression of
these variables could be useful in identifying patients
who are at high risk for the metastasis and/or recurrence of tumor and for that reason are possible candidates for adjuvant cancer therapy. Prospective studies
involving larger numbers of patients and longer periods of follow-up are needed to confirm these findings.
REFERENCES
1.
2.
3.
Folkman J, Watson K, Ingber D, Hanahan D. Induction of
angiogenesis during the transition from hyperplasia to neoplasia. Nature 1989;339:59?61.
Folkman J. What is the evidence that tumors are angiogenesis dependent? J Natl Cancer Inst 1990;82:4?6.
Ausprunk DH, Folkman J. Migration and proliferation of endothelial cells in performed and newly formed blood vessels during tumor angiogenesis. Microvasc Res 1977;14:53?8.
W: Cancer
Angiogenesis in Endometrial Carcinoma/Wagatsuma et al.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
Weidner N, Semple JP, Welch WR, Folkman J. Tumor angiogenesis and metastasis: correlation in invasive breast carcinoma. N Engl J Med 1991;324:1?8.
Weidner N, Folkman J, Pozza F, Bevilacqua P, Allred EN,
Moore DH, et al. Tumor angiogenesis: a new significant and
independent prognostic indicator in early-stage breast carcinoma. J Natl Cancer Inst 1992;84:1875?87.
Siitonen SM, Haapasalo HK, Rantala IS, Helin HJ, Isola JJ.
Comparison of different immunohistochemical methods in
the assessment of angiogenesis: lack of prognostic value in a
group of 77 selected node-negative breast carcinomas. Mod
Pathol 1995;8:745?52.
Costello P, McCann A, Carney DN, Dervan PA. Prognostic
significance of microvessel density in lymph node negative
breast carcinoma. Hum Pathol 1995;26:1181?4.
Goulding H, Nik Abdul Rashid NF, Robertson JF, Bell JA,
Elston CW, Blamey RW, et al. Assessment of angiogenesis
in breast carcinoma: an important factor in prognosis? Hum
Pathol 1995; 26:1196?200.
Macchiarini P, Fontanini G, Hardin MJ, Squartini F, Angeletti CA. Relation of neovascularization to metastasis of nonsmall cell lung cancer. Lancet 1992;340:145?6.
Yamazaki K, Abe S, Takekawa H, Sukoh N, Watanabe N,
Ogura S, et al. Tumor angiogenesis in human lung adenocarcinoma. Cancer 1994;74:2245?50.
Gasparini G, Weidner N, Malta S, Pazzo F, Boracchi P, Mezzetti M, et al. Intratumoral microvessel density and p53 protein: correlation with metastasis in head and neck squamous
cell carcinoma. Int J Cancer 1993;55:739?44.
Roychowdhury DF, Tseng A Jr., Fu KK, Weinberg V, Weidner
N. New prognostic factors in nasopharyngeal carcinoma:
tumor angiogenesis and c-erbB2 expression. Cancer 1996;
77:1419?26.
Abulafia O, Triest WE, Sherer DM, Hansen CC, Ghezzi F.
Angiogenesis in endometrial hyperplasia and stage I endometrial carcinoma. Obstet Gynecol 1995;86:479?85.
Gasparini G, Bonoldi E, Viale G, Verderio P, Boracchi P,
Panizzoni GA, et al. Prognostic and predictive value of tumour angiogenesis in ovarian carcinomas. Int J Cancer
1996;69:205?11.
Folkman J, Klagsbrun M. Angiogenic factors. Science 1987;
235:442?7.
Polverini PJ, Leibovich SJ. Induction of neovascularization
in vivo and endothelial proliferation in vitro by tumor associated macrophages. Lab Invest 1984;51:635?42.
Kandel J, Bossy-Wetzel E, Radvanyi F, Klagsbrum M, Folksman J, Hanahan D. Neovascularization is associated with a
switch to the export of BFGF in the multistep development
of fibrosarcoma. Cell 1991;66:1095?104.
Nguyen M, Watanabe H, Budson AE, Richie JP, Folkman J.
Elevated levels of the angiogenic peptide basic fibroblast
growth factor in urine of bladder cancer patients. J Natl
Cancer Inst 1993;85:241?2.
Kim KJ, Li B, Winer J, Armanini M, Gillett N, Phillips HS, et
al. Inhibition of vascular endothelial growth factor-induced
angiogenesis suppresses tumor growth in-vivo. Nature
1993;362:841?4.
Liotta LA, Kleinerman J, Saidel GM. The significance of hematogenous tumor cell clumps in the metastatic process.
Cancer Res 1976;36:889?94.
Mandriota SJ, Seghezzi G, Vassalli JD, Ferrara N, Wasi S,
Mazzieri R, et al. Vascular endothelial growth factor increases urokinase receptor expression in vascular endothelial cells. J Biol Chem 1995;270:9709?16.
/ 7ba8$$0201
01-13-98 09:28:49
cana
529
22. Bussolino F, Di Renzo MF, Ziche M, Bocchietto E, Olivero
M, Naldini L, et al. Hepatocyte growth factor is a potent
angiogenic factor which stimulates endothelial cell motility
and growth. J Cell Biol 1992;119:629?41.
23. Grant DS, Kleinman HK, Goldberg ID, Bhargava MM, Nickoloff BJ, Kinsella JL, et al. Scatter factor induces blood vessel
formation in vivo. Proc Natl Acad Sci U S A 1993;90:1937?
41.
24. Miyazawa K, Tsubouchi H, Naka D, Takahashi K, Okigaki
M, Arakaki N, et al. Molecular cloning and sequence analysis
of cDNA for human hepatocyte growth factor. Biochem Biophys Res Commun 1989;163:967?73.
25. Nakamura T, Nishizawa T, Hagiya M, Seki T, Shimonishi M,
Sugimura A, et al. Molecular cloning and expression of human hepatocyte growth factor. Nature 1989;342:440?3.
26. Weidner KM, Arakaki N, Vandekereckhove J, Weingart S,
Hartmenn G, Rieder H, et al. Evidence for the identity of
human scatter factor and human hepatocyte growth factor.
Proc Natl Acad Sci U S A 1991;88:7001?5.
27. Gherardi E, Gray J, Stocker M, Perryman M, Furlong R. Purification of scatter factor, a fibroblast-derived basic protein
which modulates epithelial interactions and movement.
Proc Natl Acad Sci U S A 1989;86:5844?8.
28. Nakamura T, Teramoto H, Ichihara A. Purification and characterization of a growth factor from rat platelets for mature
parenchymal hepatocytes in primary cultures. Proc Natl
Acad Sci U S A 1986;83:6489?93.
29. Tajima H, Mtsumoto K, Nakamura T. Regulation of cell
growth and motility by hepatocyte growth factor and receptor expression in various cell species. Exp Cell Res 1992;202:
423?31.
30. Bottaro DP, Rubin JS, Falleto DL, Chan AML, Kmiecik TE,
Vande Woude GF, et al. Identification of the hepatocyte
growth factor receptor as the c-met proto-oncogene product.
Science 1991;251:802?4.
31. Rosen EM, Nigam SK, Goldberg ID. Scatter factor and the
c-Met receptor: a paradigm for mesenchymal/epithelial interaction. J Cell Biol 1994;127:1783?7.
32. FIGO news. Int J Gynecol Obstet 1989;28:189?93.
33. Furukawa T, Duguid WP, Kobari M, Matsuno S, Tsao MS.
Hepatocyte growth factor and Met receptor expression in
human pancreatic carcinogenesis. Am J Pathol 1995;147:
889?95.
34. Horak ER, Leek R, Klenk N, LeJeune S, Smith K, Stuart N, et
al. Angiogenesis, assessed by platelet/endothelial cell adhesions molecule antibodies, as an indicator of node metastases and survival in breast cancer. Lancet 1992;340:1120?4.
35. Parums DV, Cordell JL, Micklem K, Heryet AR, Gatter CK,
Mason DY. JC70: a new monoclonal antibody that detects
vascular endothelium associated antigen in routinely processed tissue sections. J Clin Pathol 1990;43:752?7.
36. Schlingemann RO, Rietveld FJR, deWaal RMW, Bradley NJ,
Skene AI, Davies AJS, et al. Leukocyte antigen CD34 is expressed by a subset of cultured endothelial cells and on
endothelial abluminal microprocesses in the tumor stroma.
Lab Invest 1990;62:690?6.
37. Fina L, Molgaard HV, Robertson D, Bradley NJ, Monaghan
P, Delia D, et al. Expression of the CD34 gene in vascular
endothelial cells. Blood 1990;75:2417?26.
38. Di Renzo MF, Olivero M, Katsaros D, Crepaldi T, Gaglia P,
Zola P, et al. Overexpression of the MET/HGF receptor in
ovarian cancer. Int J Cancer 1994;58:658?62.
W: Cancer
530
CANCER February 1, 1998 / Volume 82 / Number 3
39. Natali PG, Prat M, Nicotra MR, Bigotti A, Olivero M, Comoglio
PM, et al. Overexpression of the met/HGF receptor in renal cell
carcinomas. Int J Cancer (Pred Oncol) 1996;69:212?7.
40. Joseph A, Weiss GH, Jin L, Fuchs A, Chowdhury S, O?Shaugnessy P, et al. Expression of scatter factor in human bladder
carcinoma. J Natl Cancer Inst 1995;87:372?7.
41. Yamashita J, Ogawa M, Yamashita S, Nomura K, Kuramoto M,
Saishoji T, et al. Immunoreactive hepatocyte growth factor is
a strong and independent predictor of recurrence and survival
in human breast cancer. Cancer Res 1994;54:1630?3.
42. Humphrey PA, Zhu X, Zarnegar R, Swanson PE, Ratliff TL,
Vollmer RT, et al. Hepatocyte growth factor and its receptor
(c-MET) in prostatic carcinoma. Am J Pathol 1995;147:386?
96.
/ 7ba8$$0201
01-13-98 09:28:49
cana
43. Rubin JS, Cahn AML, Battaro DP, Burgess WH, Taylor WG,
Cech AC, et al. A broad spectrum human lung fibroblastderived mitogen is a variant of hepatocyte growth factor.
Proc Natl Acad Sci U S A 1991;88:415?9.
44. Rosen EM, Goldberg ID, Kacinski BM, Buckholz T, Vinter DM.
Smooth muscle releases an epithelial cell scatter factor which
binds to heparin. In Vitro Cell Dev Biol 1989;25:163?73.
45. Tsao MS, Zhu H, Giaid A, Viallet J, Nakamura T, Park M.
Hepatocyte growth factor is an autocrine factor for human
normal bronchial epithelial and lung carcinoma cells. Cell
Growth Differ 1993;4:571?9.
46. Wang Y, Selden C, Morgan N, Stamp GWH, Hodgson HJF.
Hepatocyte growth factor/scatter factor expression in human mammary epithelium. Am J Pathol 1994;144:675?82.
W: Cancer
Документ
Категория
Без категории
Просмотров
4
Размер файла
310 Кб
Теги
771
1/--страниц
Пожаловаться на содержимое документа