Angiogenesis Does
Not Correlate with Rectal Cancer Metastasis*
Mokenge P. Malafa, MDa
J. Mark Barnett, MDa
Theresa Karich, MDa
Leslie Neitzel, MSa
Brian Webb, MDb
aDepartment
of Surgery
bDepartment
of Pathology
Southern
Illinois University School of Medicine
PO Box
19638
Springfield,
IL 62794
Key Words: rectal cancer, metastasis, angiogenesis, factor VIII,
survival
Abstract
To determine the usefulness
of angiogenesis in identifying rectal cancer patients with lymph node
metastases, we compared the number of microvessels in the tumors of 22 patients
with lymph node–positive rectal cancer to 22 patients with lymph node–negative
rectal cancer. Patients were matched for tumor depth, histology, and grade.
Microvessel counts in the tumors of patients with rectal cancer did not
correlate with lymph node metastasis. Microvessel counts were 21.86 ±
12.99 for lymph node–positive versus 19.14 ± 14.44 for lymph
node–negative patients (P = .5138).
Furthermore, the disease-free survival of patients with high microvessel counts
was not significantly different from that of patients with low microvessel
counts (P = .5103). Thus our data
reject the hypothesis that angiogenesis is useful in identifying the presence
of lymph node metastases in patients with rectal cancer. Additionally, in
contrast to previous studies, patients with rectal cancer did not demonstrate
an adverse impact of increased microvessel counts on disease-free survival.
*This
study was supported by grants from the William E. McElroy Charitable Foundation
and the Memorial Medical Center Foundation.
Introduction
There
is intense interest in identifying predictive features of lymph node metastasis
in rectal cancer because this tumor is increasingly undergoing minimal surgical
treatment.1-5 Local excision for rectal cancer may spare patients
extensive resection and colostomy. One of the concerns of minimal therapies is
that lymph nodes containing metastases may be left in the patient, thus
resulting in treatment failure. Current clinical methods, such as radiologic
assessment by endorectal ultrasound, have significant limitations with false-
negative rates of 14.3% and an overall accuracy of only 63.5%.6,7
The pathologist is often called on to comment on histologic features of the
primary tumor, which may predict aggressive behavior. Despite many documented
adverse histologic features, there remains no accurate means of predicting
metastatic behavior.8-11
Angiogenesis has been shown
to be predictive of lymph node metastasis.12 Microvessel counts in
tumors reflect the amount of angiogenesis within a tumor and have been proposed
as useful predictors of lymph node metastasis. Previous work by others and
ourselves has demonstrated that microvessel counts are predictive of lymph node
metastasis in breast cancer and melanoma.13,14 There has been a
great deal of interest in this finding as it relates to colorectal cancer.
Several studies have reported mixed results regarding the relationship between
angiogenesis and rectal cancer progression.15-20 These studies were
not specifically designed to examine the impact of angiogenesis in rectal
cancer lymph node metastasis.
In this study, we use a case
control approach to examine the impact of angiogenesis on identifying rectal
cancer patients with lymph node metastases. Our study examined the number of
vessel counts in the tumors of 22 patients with lymph node– positive rectal cancer
compared with 22 patients with lymph node–negative rectal cancer treated from 1984 to 1994.
Materials
and Methods
Patients
Following
approval by the Springfield Committee for Research Involving Human Subjects,
the tumor registry at Memorial Medical Center in Springfield, IL, was used to
obtain a database of patients who underwent surgery for rectal cancer between
the years 1984 and 1994. Cases for which lymph node status, relevant clinical
information, or paraffin blocks could not be obtained were excluded from the
study. Ultimately, 22 patients with rectal cancer and positive lymph nodes and
22 patients with rectal cancer and negative lymph nodes were selected. Their
respective paraffin-embedded tumor specimens were obtained. The two groups were
matched for major risk factors for metastasis, including gender, age, tumor
histology, T stage, tumor grade, and tumor size (Table 1). Information regarding
T stage and lymph node status was obtained from the original surgical pathology
reports. All other information was obtained from either the tumor registry or
the patient’s medical record.
Immunohistochemistry
Formalin-fixed,
paraffin-embedded tissue sections were heated at 60°C for 30 minutes,
deparrafinized, and rehydrated through a graded series of alcohol and water.
The sections were permeabilized with trypsin (1 mg/mL) for 5 minutes and rinsed
thoroughly with water. To block endogenous peroxidase activity, sections were
incubated with hydrogen peroxide diluted with phosphate-buffered saline (PBS)
for 5 minutes and rinsed with PBS. Nonspecific binding of the secondary
antibody was blocked with normal horse serum for 20 minutes. Sections were incubated
with a 1:50 dilution of mouse antihuman factor VIII monoclonal antibody (DAKO
Corp., Carpinteria, CA) for 30 minutes and rinsed with PBS. Immunoperoxidase
staining was performed with an avidin–biotin complex method (VECTASTAIN Elite ABC kit,
VECTOR Laboratories, Inc., Burlingame, CA) and DAB substrate (Research
Genetics, Huntsville, AL). Rinsing with distilled water terminated the
reaction. The sections were placed in a 0.1 M sodium acetate buffer for 10
minutes, counterstained with methyl green for 10 minutes, dehydrated, and
mounted. Omission of the primary antibody was employed as a negative control,
and reaction of the primary antibody with human uterine tissue was employed as
a positive control.
Quantification for Tumor
Vascularity
For
each case, the rectal tumor specimen was scanned microscopically with an
Olympus BH2 microscope to find the area of greatest vascular density. The
number of microvessels per microscopic field (200х and 400х magnification) was counted for the area of
greatest vascularity within the tumor. Structures were counted as microvessels
if they stained positively for the endothelial cell marker, factor VIII, and
appeared as an individual microvessel or endothelial cell cluster. Although
lumens were generally present, this was not a criterion for a structure to be
counted as a microvessel. Vessel counts were assessed without knowledge of
patient outcome and were performed simultaneously by two independent investigators
using a double-headed light microscope. Cases were also scored separately in a
blinded fashion by a third investigator.
Statistics
Independent
group t-tests were used to compare the two patient groups (lymph node positive
versus lymph node negative) on both the continuous and ordinal measures. Chi-
squared tests of independence or Fisher’s exact test, when appropriate, were
used to compare the two groups in regard to the categorical data. A logrank
test was used to compare survival curves. Results are reported as mean ± standard deviation and are considered
significant for P < .05.
Results
Microvessel Counts Do Not
Correlate with Lymph Node Metastasis in Rectal Cancer
Patients
with lymph node metastases at the time of surgery had no statistically
significant difference in number of microvessels compared with patients without
lymph node metastases (Figure 1). Microvessel counts were 21.86 ± 12.99 versus 19.14 ± 14.44 (P
= .5138) per 200х field and 11.55 ± 6.142 versus 9.864 ± 7.415 (P
= .4255) per 400х field for lymph
node–positive versus lymph node–negative groups, respectively. The median
counts were 19.00 and 13.00 per 200х field and 10.00 and 6.50
per 400х field for lymph
node–positive tumors versus lymph node–negative tumors, respectively.
Angiogenesis Is Not
Predictive of Survival in Rectal Cancer
The
median number of microvessel counts for all patients in the study was 18 at 200х magnification. To determine the effect of
microvessel counts on disease-free survival, the patients were stratified into
two groups, those with microvessel counts greater than or equal to 18, and
those with microvessel counts less than 18. This stratification was made
irrespective of a patient’s node status. Data was then plotted using patient’s
survival status at last follow-up. There was no statistically significant
difference in disease-free survival between patients with high versus low
vessel counts (P = .5103; Figure 2).
When patients were stratified by metastatic status and microvessel counts,
angiogenesis still had no impact on disease-free survival (P = .4047 for node-positive patients) and (P = .1360 for node-negative patients; Figure 3).
Discussion
Our study indicates no
correlation between angiogenesis as measured by microvessel counts and lymph
node metastasis in rectal cancer. The literature regarding colorectal cancer
and angiogenesis reports conflicting results in terms of the impact of this
process on lymph node metastasis.15-17,19,20 Our findings support
the conclusions by Saclarides and coworkers16 and Vermeulen and
coworkers20 who did not find a correlation between angiogenesis and
lymph node metastasis. In contrast, our study refutes the findings of
Takebayashi and coworkers who reported a correlation between microvessel counts
and lymph node metastasis.19 It is important to note that the study
by Takebayashi and colleagues did not differentiate between rectal and colon
tumors but instead grouped them together as colorectal cancers. We feel that
this grouping may obscure the unique biology of rectal cancers.
We specifically designed
this study to determine the impact of angiogenesis on identifying patients with
lymph node metastases. The strengths of our study include the focus on rectal
tumors and the use of a control group matched for known factors, which may
influence the metastatic phenotype (tumor depth, grade, histology). The
weakness of our study is its retrospective nature and relatively modest number
of subjects. Prospective studies with larger patient cohorts will be needed to
confirm our findings.
Angiogenesis has been
correlated with survival in patients with diverse malignancies, including
rectal cancer.12,15,16,19,20 We did not observe a significant impact
of angiogenesis on patient disease-free survival. This is in contrast to the
results of other studies.15,16,19,20 Saclarides and colleagues noted
a significant increase in vessel counts in patients who lived less than 5 years
compared with those who lived more than 5 years.16 However, these
patients were not controlled for risk factors associated with rectal cancer
survival. Takebayashi and coworkers, Vermeulen and coworkers, and Ishikawa and
coworkers used univariate and multivariate Cox proportional hazard regression
models to demonstrate that microvessel counts were an independent prognostic
factor in the survival of patients with rectal cancer.15,19,20
However, these studies have a limited value in determining prognosis in
individual cases because there are overlaps and large standard deviations of
microvessel counts related to the presence or absence of other pathologic or
prognostic factors. Our observation of the absence of a correlation between
microvessel counts and survival reflects our matched patient population. The
use of microvessel counts to predict mortality in patients with rectal cancer
should be evaluated in prospective controlled trials.
CONCLUSION
Angiogenesis
is currently an important target for novel antitumor therapy and
prognostication. We had hoped to uncover a predictive value of angiogenesis in
rectal cancer lymph node metastasis in order to include this factor in
selecting patients for local therapy. We discovered significant heterogeneity
in the angiogenic phenotype of rectal tumors. Future clinical studies are
needed to better define the role of angiogenesis for specific applications to
patient care.
REFERENCES
1. Steele Jr GD, Herndon JE, Bleday R, et al: Sphincter-sparing
treatment for distal rectal adenocarcinoma. Ann
Surg Oncol 6:433-441, 1999.
2. Mentges B, Buess G, Effinger G, et al: Indications and
results of local treatment of rectal cancer. Br J Surg 84:348-351, 1997.
3. Mentges B, Buess G, Schafer D, et al: Local therapy of
rectal tumors. Dis Colon Rectum
39:886-892, 1996.
4. Bleday R: Local excision of rectal cancer. World J Surg 21:706-714, 1997.
5. Bleday R, Breen E, Jessup JM, et al: Prospective evaluation
of local excision for small rectal cancers. Dis
Colon Rectum 40:388-392, 1997.
6. Kim NK, Kim MJ, Yun SH, et al: Comparative study of
transrectal ultrasonography, pelvic computerized tomography, and magnetic
resonance imaging in preoperative staging of rectal cancer. Dis Colon Rectum 42:770-775, 1999.
7. Katsura Y, Yamada K, Ishizawa T, et al: Endorectal
ultrasonography for the assessment of wall invasion and lymph node metastasis in
rectal cancer. Dis Colon Rectum
35:362-368, 1992.
8. Blumberg D, Paty PB, Guillem JG, et al: All patients with
small intramural rectal cancers are at risk for lymph node metastasis. Dis Colon Rectum 42:881-885, 1999.
9. Brandao O, Sobrinho-Simoes MA, Serrao D, et al: Prognosis
in colorectal carcinoma: A reassessment of the pathologist's role. Pathol Res Pract 180:506-510, 1985.
10. Brodsky JT, Richard GK, Cohen AM, Minsky BD: Variables
correlated with the risk of lymph node metastasis in early rectal cancer. Cancer 69:322-326, 1992.
11. Tanaka S, Yokota T, Saito D, et al: Clinicopathologic features
of early rectal carcinoma and indications for endoscopic treatment. Dis Colon Rectum 38:959-963, 1995.
12. Folkman J: Angiogenesis research: From laboratory to
clinic. Forum (Genova) 9:59-62, 1999.
13. Neitzel LT, Neitzel CD, Magee KL, Malafa MP: Angiogenesis
correlates with metastasis in melanoma. Ann
Surg Oncol 6:70-74, 1999.
14. Weidner N: Angiogenesis in breast cancer. Cancer Treat Res 83:265-301, 1996.
15. Ishikawa H, Fujii H, Yamamoto K, et al: Tumor
angiogenesis predicts recurrence with normal serum carcinoembryonic antigen in
advanced rectal carcinoma patients. Surg
Today 29:983-991, 1999.
16. Saclarides TJ, Speziale NJ, Drab E, et al: Tumor
angiogenesis and rectal carcinoma. Dis
Colon Rectum 37:921-926, 1994.
17. Saclarides TJ: Angiogenesis in colorectal cancer. Surg Clin North Am 77:253-260, 1997.
18. Takahashi T, Kato T, Kodaira S, et al: Prognostic factors
of colorectal cancer: Results of multivariate analysis of curative resection
cases with or without adjuvant chemotherapy. Am J Clin Oncol 19:408-415, 1996.
19. Takebayashi Y, Aklyama S, Yamada K, et al: Angiogenesis
as an unfavorable prognostic factor in human colorectal carcinoma. Cancer 78:226-231, 1996.
20. Vermeulen
PB, Van den Eynden GG, Huget P, et al: Prospective study of intratumoral
microvessel density, p53 expression and survival in colorectal cancer. Br J Cancer 79:316-322, 1999.
Figure 1. (A) A representative photomicrograph showing immunostaining for factor
VIII in rectal cancer. Many small microvessels positive for factor
VIII were scattered in the tumor stroma. (B) Correlation between microvessel
counts and lymph node metastasis. The number of microvessels in the
rectal tumors of patients with lymph node metastases were not statistically
different from the number in patients without lymph node metastases
(P = .5138 per 200х field and P = .4255 per 400х field).
Figure 2. Prognostic significance of
microvessel counts for disease-free survival in patients with rectal cancer
regardless of lymph node status. Survival analysis for microvessel counts in
relationship to the recurrence of cancer in a total of 44 patients was
performed using a log rank test. A cutoff level of 18 microvessels was used
because this was the median number of microvessels for all patients combined.
There was no significant difference between the curves (P = .5103).
Figure
3. Prognostic significance of
microvessel counts for disease-free survival in patients with (A) and without (B) lymph node metastases. There was no significant difference in
disease-free survival for patients with (P
= .4047) or without (P = .1360) lymph
node metastases.
Table 1
Clinicopathologic Characteristics of 44 Patients
with Rectal Cancer
Lymph Node (positive) Lymph Node (negative) P
Value
Number
of patients 22 22 --
Age
(years) 68.5 ± 9.989 70.05 ± 8.627 .5858
Gender
(male/female) 12/10 11/11 1.0000
Tumor
histology
Adenocarcinoma 21 21 1.5116
Mucinous 1 1 --
T
stage (T2/T3)* 3/19 3/19 1.3360
Tumor
grade (1/2/3)† 3/15/4 8/11/3 .2197
Tumor
size (cm) 4.891 ± 2.104 4.873 ± 1.240 .9723
Nodes
examined 7.619 ± 5.590 6.810 ± 4.262 .6006
Chemotherapy 9/22 3/22 .0904
Radiation
therapy 10/22 7/22 .4310
*T
stage: tumor depth of invasion as defined by AJCC.
†Tumor
grade: 1 = poorly differentiated, 2 = moderately differentiated, 3 = well
differentiated.