Abstract
OBJECTIVE To investigate the relationship between lymphatic vessel density and lymph node metastasis of invasive micropapillary carcinoma (IMPC) of the breast.
METHODS The immunohistochemical study for vascular endothelial growth factor-C (VEGF-C), VEGF Receptor-3 (VEGFR-3) and lymphatic vesseI density of 51 cases of lMPC were performed，and lymph node metastases were examined by microscopic analysis of these cases.
RESULTS In lMPC, VEGF-C was expressed in the cytoplasm and/or on the membrane Of the tumor cells. and the expression of VEGF-C showed a positive correlation with lymph node metastasis (P<0.01). Lymphatic vesseI density was determined by the number of micro-lymphatic vessels with VEGFR-3 positive staining. Lymphatic vessel density was positively correlated with VEGF-C expression (P<0.01) and lymph node metastasis(P<0.01). The percentage of IMPC in the tumor was not associated with the incidence of lymph node metastasis. The metastatic foci in lVmph nodes were either pure or predominant micropapillary carcinoma.
CONCLUSION The results suggested that VEGF-C overexpression stimulated tumor lymphangiogenesis, and the increased lymphatic vessel density may be the key factor that influenced lymph node metastasis of IMPC.
keywords
- breast neoplasm
- invasive micropapillan/carcinoma (IMPC)
- vascular endothelial growth factor(VEGF)
- lymph node metastasis
Invasive micropapillary carcinoma (IMPC) of the breast is a newly recognized subtype of epithelial tumor in the 2003 WHO histological classification of breast tumors.[1] It has a high frequency of axillary lymph node metastases, in keeping with an angioinvasive phenotype. So IMPC is thought to be an aggressive subtype of breast carcinoma. [2, 3] However, we have found a few reports that showed IMPC without lymph node metastases.[4,5] We also observed a case composed of pure micropapillary components but without lymph node metastases in our pathological diagnosis. Therefore we became interested in the mechanism or relevant factors relating to IMPC lymph node metastases.
In this study, we made a retrospective analysis of a series of breast carcinomas with a variable percentage of IMPC. Our objective was to investigate the relationship between lymphatic vessel density and lymph node metastasis of breast IMPC.
MATERIALS AND METHODS
Materials
In a retrospective review, we identified 51 cases of breast IMPC from the Department of Breast Pathology, Tianjin Medical University Cancer Hospital, Tianjin, China. These cases were encountered over approximately a 9-month period in 2003 when 1056 invasive breast carcinomas were diagnosed.
Methods
Hematoxylin and eosin-stained glass slides from each case were reviewed by 3 senior breast pathologists. IMPC in our study included all the cases that had a micropapillary component (according to the morphologic criteria in 2003 WHO histological classification of breast tumors) in the tumor tissue. Grouping was made according to the percentage of micropapillary component in the whole tumor (<25%, 25~49%, 50~75%, >75%, 4 groups).[6] For example, a case of “invasive ductal carcinoma with invasive micropapillary carcinoma (30%)” belonged to the 25~49% group, and a case of “invasive micropapillary carcinoma (80% ) with invasive ductal carcinoma” belonged to the >75% group, and so on. The type and percentage of non-micropapillary components in the tumor tissue were registered at the same time.
Immunohistochemical analysis
Immunohistochemical analysis using the labeled streptavidin biotin (LSAB) method was performed for VEGF-C (Zymed Laboratories) and for VEGFR-3 (Zymed Laboratories). Lymphatic vessel density was evaluated in all 51 cases. Cases were considered positive for VEGF-c if cytoplasmic and/or membranous staining was present in at least 10% oi’the tumor cells tested. Lymphatic vessel density was evaluated by immunohistochemical staining of intratumoral microlymphatic vessels for VEGFR-3. Any positively staining endothelial cells or endothelial cell clusters clearly separated from adjacent clusters and background, with or without lumen, was considered an individual vessel. Vessels that had a thick muscular layer were excluded from the count. The quantification of lymphatic vessel density was assessed according to the method of Weidner [7] by at least two independent observers. The sections were initially screened at low magnifications (×40 and ×100) to identify the most vascular area of the tumor (hot spot). Within the hot spot area, the stained micro-lymphatic vessels were counted in a single high-power (×400) field. Lymphatic vessel densify was expressed as the number of micro-lymphatic vessels/ field. Micro-lymphatic vessel counts were compared between the observers and discrepant results were reviewed together. The consensus reached was used as the final score for analysis.
Lymph node metastasis
The incidence of nodal metastases, and the types of metastatic carcinoma (pure IMPC, mixed IMPC with others, and non-IMPC) in lymph nodes were recorded.
Statistical analysis
Statistical evaluation was performed using the Student’s t-test and Chi-square test.
RESULTS
Incidence of IMPC
Tumors with an IMPC component in this study comprised 4.83% of the reviewed breast carcinomas (1056 cases). The patients were 51 females ranging in age from 38 to 67 years, with a mean age of 52.3 years. IMPC comprised less than 25% of the tumor in 9 cases (17.6%), 25~49% in 11 cases (21.6%), 50~75% in 12 cases(23.5%), and more than 75% in 19 cases (37.3%). The group with more than 75% comprised 1.80% of the reviewed breast carcinomas (1,056 cases).
Tumor components and lymph node metastasis
(1) Tumor components
There were 4 cases of pure IMPC. In the other cases, non-micropapillary components included invasive ductal carcinoma (IDC) (13 cases), IDC and ductal carcinoma in situ (DCIS) (22 cases) and DCIS (6 cases). The DCIS were micropapillary, comedo, solid and cribriform types. The 6 cases with special types of breast carcinoma included 5 mucinous carcinomas and 1 tubular carcinoma.
(2) Lymph node metastasis
Axillary dissections were performed in 51 cases, 10 of which (19.6%) had negative lymph nodes, in particular, including 1 case of pure IMPC. There were 41 cases with positive lymph nodes (41/5leases, 80.4%). The incidence of lymph node metastasis in the groups with micropapillary components <25%, 25~49%, 50~75%, and >75% were respectively 77.8%, 81.8%, 75%, 84.2%. No statistical correlation was found between lymph node metastasis and the percentage of IMPC. The metastatic foci in lymph nodes were pure IMPC in 28 cases (28/41 cases, 68.3%, Fig.1), and predominantly IMPC with little other types of carcinoma in 13 cases (13/41 cases, 31.7%, Fig.2).
The metastatic foci of a lymph node with pure IMPC (Hematoxylin and eosin, × 40 magnification).
The metastatic foci of a lymph node with predominant IMPC with little invasive ductal carcinoma (arrow, Hematoxylin and eosin, × 40 magnification).
Tumor size and lymph node metastasis
Tumor size ranged from 1.5 to 11 cm (mean, 3.9 cm), which was not correlated with lymph node metastasis of the tumors.
Immunohistochemical study
(1) In the immunohistochemical study, 80.4% (41/ 5leases) of the cases stained positively for VEGF-C protein. As mentioned above, VEGF-C expression was positive in the cytoplasm and/or on the membrane of the IMPC tumor cells (Fig.3). The expression of VEGF-C was in positive correlation with lymph node metastasis (P<0.01, Table 1).
VEGF-C expression was positive in the cytoplasm and/or on the membrane of IMPC tumor cells (LSAB immunoperoxidase with hematoxylin counterstain, × 100 magnification).
Relationship between the expression of VEGF-C and lymph node metastasis
(2) Lymphatic endothelial cells with positive VEG-FR-3 staining were dispersedly located in the stroma of IMPC, often in the periphery, and cancer cell emboli could be seen in some lymphatic vessels (Fig.4). In the 51 cases, lymphatic vessel density ranged from 11.45 to 40.21 (mean, 24.35±5.68). Lymphatic vessel density was positively correlated with VEGF-C expression (P<0.01) and lymph node metastasis (P<0.01) (Table 2).
Lymphatic endothelial cells with VEGFR-3 positive staining were dispersedly located in the stroma of IMPC, and cancer cell emboli ∞uld be seen in some lymphatic vessels (arrow, LSAB immunoperoxidase with hematoxylin counterstain, × 100 magnification).
Relationship of lymphatic vessel density with the expression of VEGF-C and lymph node metastasis
DISCUSSION
It has been reported that breast carcinomas with a IMPC growth pattern are associated with a higher histologic grade and a distinctly lymphotropic character,[2, 3, 8-13] but few reports have discussed the mechanism which facilitates IMPC metastasis.
VEGF-C is a member of the VEGF family and is the only ligand that can combine with the VEGFR-3 receptor located on the endothelium of lymphatic vessels and modulates the physiological function of lymphatic vessels. Recent evidence suggests that VEGF-C promotes lymphangiogenesis, and that tumor lymphangiogenesis, in turn, promotes lymphatic metastasis. And studies have shown that the VEGF-C/VEGFR-3 system plays a modulating role in metastasis of many kinds of tumors.’[14-19] Since IMPC have a great tendency to invade lymphatics and spread to regional lymph nodes, we analyzed the relationship between lymphatic vessel density and lymph node status as well as VEGF-C immunoreactivity in IMPC. Our analysis showed that VEGF-C overexpression stimulated tumor lymphangiogenesis, and that the increased lymphatic vessel density was correlated with lymph node metastasis. These findings suggest that tumor-produced VEGF-C bound to VEGF receptors on lymphatic endothelial cells causing proliferation and growth of new lymphatic capillaries. The newly-formed lymphatic capillaries were then more often exposed to cancer cells as most hyperplastic lymphatic capillaries had not formed tubules or only had incomplete tubular structure with thin and frail walls. Thus it was easy for the cancer cells to invade, and increased the possibility of lymph node metastasis of IMPC. So, higher expression of VEGF-C and VEGFR-3, and increased lymphatic vessel density may play an important role in promoting IMPC lymph node metastasis.
Shi et al.[4] reported a case of pure IMPC of the breast (low-grade nuclei) with no lymph node metastasis; Takashi et al.[5] published a case in which 60% of the tumor was composed of IMPC (low histologic grade, with mild lymphatic invasion), admixed with papillotubular and mucinous carcinoma. Only 1 of 25 lymph nodes had tumor metastasis (no IMPC was observed and the metastatic foci were papillotubular carcinoma). We also observed a case of pure IMPC (low histologic grade, VEGF-C positive in only 20% of the tumor cells, and low lymphatic vessel density) with no lymph node metastasis. To analyze these cases, it is suggested that well-differentiated tumor cells, absence of lymphangiogenesis and lymphatic vessel invasion may be the factors resulting in inhibition of IMPC lymph node metastasis. Our study indicated that the malignant biological behavior of tumors with IMPC is not correlated with the extent of the micropapillary pattern, but with the features of tumor histology and stromal reactions. This can also explain why tumors having very little IMPC could spread to regional lymph nodes, while no lymph node metastasis was observed for tumors with pure IMPC.
In conclusion, lymphatic vessel density may be the key factor that influence IMPC lymph node metastasis. Recognition of the pathologic features of IMPC is important in predicting lymph node metastases and poor clinical outcome. Further studies are needed to elucidate the mechanism underlying the lymphotrophism of IMPC.
- Received November 10, 2005.
- Accepted December 28, 2005.
- Copyright © 2006 by Tianjin Medical University Cancer Institute & Hospital and Springer
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.