Full text of DOI:10.1080/00313020126325

Pathology (2001) 33, pp. 303–306
A COMPARISON OF THE PATTERNS OF LAMININ EXPRESSION IN
FIBROADENOMA, FIBROCYSTIC DISEASES, PRE-INVASIVE AND
INVASIVE DUCTAL BREAST CARCINOMA
W
EI
-
QIANG
Z
HENG*, LAI
M
ENG
L
OOI
†
AND
P
HAIK
L
ENG
C
HEAH
†
*
†
Department of Pathology, Changhai Hospital, Second Military Medical University, Shanghai, China, and
Department of Pathology, Faculty of Medicine, University of Malaya, Kuala Lumpur, Malaysia
collagens1,2 and non-collagenous glycoproteins. Laminin,
Summary
The basement membrane (BM), of which laminin is a major
glycoprotein component, is an important barrier to tumour
cells which must be breeched before metastatic spread can
occur. We have compared the pattern of laminin expression
in a range of benign and malignant breast lesions to better
understand the process of tumour progression. A total of 162
cases of breast samples, comprising 18 fibroadenomas, 22
cases of fibrocystic disease, 96 cases of invasive ductal
carcinoma and 26 carcinomas with intraductal components,
were evaluated for laminin expression by a standard immu-
noperoxidase method on formalin-fixed, paraffin-embedded
histological sections, using a commercial antibody against
human laminin. The pattern of laminin expression was
charted as follows: Type 1, > 70% of BM complete/con-
tinuous; Type II, > 70% of BM moderately disrupted; Type III,
a non-collagenous glycoprotein component of basement
membranes, has been identified.3 It is a glycoprotein
with a molecular weight of about 600 000 and it is also
important in the cell adhesion process.
Metastasis is one of the most important reasons for
deterioration of patients with breast cancer. The basement
membranes are of particular interest in cancer invasion
because they are regarded as a significant hindrance to the
entry of cancer cells into the matrix and metastatic cells
– 5
which attach to the basement membranes of endothelial
cells in the spread of tumours via circulation.6
– 8
In
mammary carcinomas, the first step of tumoural invasion
is characterised by the loss of basement components,
particularly laminin.9 Basement membranes have been
extensively studied in neoplastic breast lesions.1
Although basement membrane morphology in breast
lesions has been the subject of a large number of studies,
there have been relatively few detailed investigations of
the patterns of basement membrane deposition in various
benign and malignant breast lesions and the consistency of
these patterns. It seemed important to examine the dis-
tribution of basement membrane laminin in non-malignant
and malignant breast lesions in this study.
0–12
>
70% of BM completely disrupted. The Type I pattern was
observed in all cases of fibroadenoma and fibrocystic
diseases, and in 77% of intraductal carcinoma components.
Various patterns of BM disruption were observed in invasive
ductal carcinoma. Severity of BM disruption correlated with
histological grade of the carcinomas (P < 0.001). Small-sized
tumours, those without lymphatic invasion and lymph node-
negative tumours showed more complete patterns of laminin
expression. The current study suggests that tumour cells with
high histological grade possess an enhanced capacity to
disrupt the basement membrane, an important step in the
metastatic process. The detection of BM disruption by
immunohistochemical staining for laminin is technically easy
and may be usefully applied for the differentiation of in situ
and microinvasive carcinoma.
MATERIALS AND METHODS
Patients
A total of 162 breast samples of female patients, comprising 18
fibroadenomas, 22 cases of fibrocystic disease, 96 invasive ductal
carcinomas and 26 lesions with intraductal carcinoma components, were
retrieved from the archives of the Department of Pathology, University of
Malaya. Formalin-fixed, paraffin-embedded histological sections of the
lesions, stained with H&E, were reviewed by three pathologists for
designation into the above diagnostic categories. Invasive ductal carcinoma
were evaluated for histological grade according to the modified Bloom and
Key words: Breast, carcinoma, laminin, immunohistochemistry.
Abbreviations: BM, basement membrane.
Received 21 July 2000; revised 19 February 2001; accepted 26 February
2001
13
Richardson method of Elston.
Demographic and clinical data were obtained from histopathology
request forms accompanying the samples. Information on the size of the
breast carcinomas was extracted from the reports of the pathologists who
performed the original pathological examinations on the lesions. Patients
with invasive ductal carcinomas were divided into two age groups ( < 50
and ꢀ50 years) based on their presumed menopausal status.
INTRODUCTION
Basement membranes occur in connection with epithe-
lium, endothelium, mesothelium, smooth and striated mus-
cle cells, Schwann cells and fat cells. Basement mem-
branes separate the epithelium from the mesenchymal
tissues, and are composed mainly of a specialised generic
type of collagen (type IV collagen) which resembles, in
some respects, the procollagen forms of the interstitial
Immunohistochemical staining
When more than one tissue block was available from a case, the one most
representative of the lesion or, in the case of carcinoma, with the
highest tumour content was selected for immunohistochemistry. Laminin
ISSN 0031–3025 printed/ISSN 1465–3931 online/01/030303–04 © 2001 Royal College of Pathologists of Australasia
DOI:10.1080/00313020120062884

304 ZHENG et al.
Pathology (2001), 33, August
Fig. 1 Linear and continuous laminin staining of the epithelial
basement membrane in a case of breast fibrocystic disease (laminin
immunoperoxidase staining, original magnification, ´100).
Fig. 2 Continuous linear staining of basement membrane around
intraductal components of invasive ductal carcinoma (laminin
immunoperoxidase staining, original magnification ´200).
expression was demonstrated by the standard avidin–biotin complex
immunoperoxidase method on formalin-fixed, paraffin-embedded histo-
logical sections. Five-mm paraffin sections were melted overnight at 42°C,
clear in xylene, and hydrated in a series of decreasing concentrations of
ethanol (100, 80, 70%) and distilled water. The sections were then treated
with 0.2% pepsin (Boehringer Mannheim, Germany) in 0.2 N HCl at 37°C
In intraductal components of invasive ductal carcinoma,
a continuous linear staining of basement membranes
around the involved ducts was again present (Fig. 2).
However, it appeared slightly irregular and thickened, as
shown by darker staining, compared with that observed in
the non-neoplastic breast. In addition, dystrophic basement
membranes around ducts containing tumour cells were
observed (6/26, 23%).
The staining pattern in invasive ductal carcinoma was
quite heterogeneous, even within an individual carcinoma.
The staining was discontinuous around the cancer cell
nests infiltrating the stroma, with gaps and irregular
thickening (Fig. 3). No staining was visible around tumour
cell masses in some areas within an individual carcinoma.
In some focal regions of well-differentiated carcinoma,
partial basement membrane formation by the pseudo-duct
structures was observed.
for 30 min. The reaction was blocked by immersing the sections in distilled
water for 5´3 min. After blocking of endogenous peroxidase activity with
a
0.3% solution of hydrogen peroxide and methanol, slides were
sequentially incubated with a dilution of 1:100 primary rabbit laminin
antibody (Dako, USA) for 30 min, followed by incubation overnight at 4°C
with biotinylated antirabbit immunoglobulin for 45 min, and avidin–biotin
peroxidase complex for 40 min (Dako). Slides were washed three times in
phosphate-buffered saline (PBS) for 5 min between the above steps.
Freshly prepared 0.05 M Tris buffer, pH 7.6, containing 0.01% H
2 2
O
3
,3-diaminobenzidine (DAB) was used as chromogen in the application of
hydrogen peroxide. The slides were then lightly counterstained with
haematoxylin. Several dilutions of the antibody were tested to optimise the
staining (and to avoid background staining) before the total series was
The correlation of immunostaining pattern and patho-
logical features was evaluated. Various sized tumours did
not produce a significant difference in laminin integrity
processed. Positive controls consisted of sections from a breast carcinoma
with known detectable laminin. Omission of the primary antibody
incubation step from the procedure served as negative control.
(
P > 0.05). Although the result of statistical analysis was
not significant (P > 0.05), laminin absence appeared to be
Immunohistochemical evaluation
The pattern of laminin expression was charted as follows: Type I, the
staining pattern of > 70% of basement membranes in the range of one
section was complete/continuous; Type II, > 70% of BM moderately
disrupted; Type III, > 70% of BM completely disrupted.
Statistical analysis
The patterns of laminin positivity of the various breast lesions were
2
compared against each other. The x -test was used to evaluate significance
of correlation between the various parameters under study.
RESULTS
Laminin immunostaining in fibroadenoma and fibrocystic
disease showed a continuous line around lobules and ducts;
the staining was located in basement membranes along the
epithelial cell layers in contact with the stroma (Fig. 1).
Also, regular linear staining was observed in basement
membranes around blood vessels. No intracellular staining
was observed.
Fig. 3 Invasive breast carcinoma with discontinuous laminin staining
around the malignant cells (laminin immunoperoxidase staining, original
magnification, ´200).

LAMININ EXPRESSION IN BREAST CARCINOMA 305
Correlation between laminin immunostaining and morphology in invasive ductal carcinoma of breast
TABLE 1
Immunostaining pattern (%)
Type II Type III
Number
examined
Type I
P
Histological grade
I
14
41
41
9 (64.3)
3 (7.3)
2 (4.9)
4 (28.6)
25 (60.9)
5 (12.2)
1 (7.1)
13 (31.7)
34 (82.9)
II
III
< 0.001
> 0.05
Tumour size (diameter)
<
3 cm
–6 cm
6 cm
45
37
14
7 (15.6)
2 (5.4)
5 (35.7)
13 (28.9)
15 (40.5)
6 (42.9)
25 (55.5)
20 (54.1)
3 (21.4)
3
>
Total
96
14 (14.6)
34 (35.4)
48 (50.0)
correlated with lymphatic invasion and lymph node metas-
tasis because there was a tendency for cases with more
lymphatic invasion and more lymph node metastases to
show the higher-grade staining patterns. It was also
evident that higher-grade patterns were likely to be present
in cases with vascular invasion. Moreover, statistical
significance was found between staining patterns and
histological grade (Table 1); the continuous linear pattern
was more frequently observed in low-grade carcinomas,
whereas basement membrane disruption was more fre-
quently observed in higher-grade carcinomas (P < 0.001).
brane was significantly associated with well-differentiated
carcinomas, whereas basement membrane disruption more
frequently occurred in poorly differentiated carcinomas.
These findings suggest that tumours with a higher histo-
logical grade possess an enhanced disruption of the
basement membrane, an important step in the metastatic
process. The correlation of disintegration of the laminin-
containing basement membranes of carcinomas with
increasingly anaplastic appearance supported the notion
that basement membranes may play a role in cancer
progression.20
It was also noted that there were qualitative changes in
laminin immunostaining of both invasive and intraductal
carcinomas, even within an individual case. Most cancers
originate from the malignant transformation of a single
cell (monoclonal origin of tumours). Nevertheless, the
inherent genetic instability of the malignant phenotype
leads to the appearance of subpopulations with diverse
biological characteristics and profound variations in their
metastatic potential (tumour heterogeneity). The demon-
stration of tumour heterogeneity has led to the concept
that at each step of the metastatic cascade, only the fittest
cells survive. The heterogeneity of laminin distribution in
the current study suggested such demonstration of sub-
populations with diverse biological tumours. This hetero-
geneity of laminin immunostaining distribution indicated
that both invasive and intraductal components of breast
carcinomas were probably composed of heterogeneous
subpopulations with variable abilities to produce laminin.
It is also implied that some intraductal carcinoma cells
lose the ability to synthesise basement membranes, since
dystrophic basement membranes around ducts composed
of tumour cells were exhibited in some intraductal compo-
nents. It is known that intraductal carcinomas are confined
to the epithelium in which they arise. Microinvasion in a
predominantly intraductal carcinoma consists of tumour
penetration through the basement membrane of a duct
involved by intraductal carcinoma. Distortion due to
fibrosis, elastosis, and inflammation may make recognition
of such penetration difficult, and it is also often difficult to
determine whether small tumour glands represent invasion
or merely intraductal carcinoma involving more distal
units of the lobule. It appears that a proportion of
histologically classified intraductal carcinomas of the
breast has a disruption of the laminin layer which may be
indicative of early invasion. Hence, laminin detection by
immunohistochemistry may be a useful and important
DISCUSSION
The visualisation of basement membranes by laminin
staining allowed us to study the organisation and role of
basement membrane integrity in the progression of breast
cancer in more detail.
In the current study, both malignant and non-malignant
breast tissue were stained for laminin. It seems reasonable
to conclude that laminin synthesis also occurs in carci-
noma cells. The significance of this finding linked to the
previous studies on the cell adhesion-promoting properties
of laminin suggests that laminin could be important for the
1
4,15
attachment and growth of tumour cells.
It has been
reported1
6–18
that malignant breast tumours only occasion-
1
9
ally possess basement membranes. Pitelka et al. reported
that well-differentiated malignant mouse mammary tumour
lines had intact basement membranes demonstrable by
electron microscopy, whereas a poorly differentiated
tumour line lacked them. Our study was in agreement with
these previous investigations of basement membrane chan-
ges in breast carcinoma. As visualised by laminin staining,
a progressive lack of basement membranes seemed to
parallel an increasing degree of tumour dedifferentiation.
We considered that this lack of basement membranes may
be because a poorly differentiated tumour would partially
or completely lose its ability to synthesise laminin and
other components. In addition, the occurrence of basement
membrane destruction by proteolytic enzymes released by
stromal cells would be increased in a poorly differentiated
carcinoma, thus the absence of a basement membrane
barrier may facilitate tumour invasion and spread. This has
been confirmed by the correlation found between the
immunostaining patterns and histological grade or differ-
entiated status, i.e., the more complete basement mem-

3
06 ZHENG et al.
Pathology (2001), 33, August
4
. Timpl R, Rohde H, Robey PG, et al. Laminin–a glycoprotein from
basement membranes. J Biol Chem 1979; 254: 9933–7.
marker to differentiate true ‘in situ’ from ‘microinvasion’
breast carcinoma. Moreover, those intraductal components
of carcinoma may be analogous to invasive carcinoma and
it would be of interest to follow up whether they had
5
. Wewer U, Albrechtsen R, Ruoslahti E. Laminin, a noncollagenou s
component of epithelial basement membranes synthesized by a rat
yolk sac tumor. Cancer Res 1981; 41: 1518–24.
obtained an invading propensity.
6. Kraemer RH, Gonzalez R, Nicolson GL. Metastatic tumour cells
adhere preferentially to the extracellular matrix underlying vascular
endothelial cells. Int J Cancer 1980; 26: 639–45.
It has been reported21 that profound changes occur in
the distribution and quantity of the epithelial basement
membrane during the transition from benign to invasive
carcinoma. However, no matter how extensive the archi-
tectural disorganisation, these benign disorders are always
characterised by a continuous basement membrane sepa-
rating the epithelium from the stroma. In contrast, invasive
ductal carcinoma consistently lack a complete extracellular
basement membrane around the invading tumour cells in
the stroma.
This study did not show an accurate correlation between
laminin expression and lymphatic invasion and lymph
node metastasis. Anyway, the possibility that the pattern of
basement membranes in breast cancer might be of prog-
nostic significance needs to be further explored. It will be
important to evaluate in long-term clinical studies whether
absence of laminin represents those subclones of the
tumour with a special metastatic potential.
7
. Murray JC, Liotta L, Rennard SI, Martin GR. Adhesion character-
istics of murine metastatic and nonmetastatic tumor cells in vitro.
Cancer Res 1980; 40: 347–51.
8
. Vlodavsky I, Gospodarowicz D. Respective roles of laminin and
fibronectin in adhesion of human carcinoma and sarcoma cells.
Nature 1981; 289: 304–6.
9. Clavel C, Birembaut P, Adnet JJ, Foidart JM. Breast carcinomas and
the extracellular matrix. Ann Pathol 1988; 8: 107–13.
10. Ekblom P, Miettinen M, Foraman L, Andersson LC. Basement
membrane and apocrine epithelial antigens in different diagnosis
between tubular carcinoma and sclerosing adenosis of the breast. J
Clin Pathol 1984; 37: 357–63.
1
1. Albrechtsen R, Nielsen M, Wewer V, Engvall E, Ruoslahti E.
Basement membrane changes in breast cancer detected by immuno-
histochemical staining for laminin. Cancer Res 1981; 41: 5076–81.
2. Gusterson BA, Warburton MJ, Metchell D, et al. Distribution of
myoepithelial cells and basement membrane proteins in the normal
breast and in benign and malignant breast diseases. Cancer Res
1982; 42: 4763–70.
1
13. Elston CW. Grading of invasive carcinoma of the breast. In: Page
DL, Anderson TJ, editors. Diagnostic histopathology of the breast.
1st ed. Edinburgh: Churchill Livingstone, 1987; 300–11.
A
CKNOWLEDGEMENTS This study was supported by a
research and development grant from the Malaysian Min-
istry of Science, Technology and the Environment: IRPA
14. Carisson RL, Engvall E, Freeman A, Ruoslahti E. Laminin and
fibronectin in cell adhesion: enhanced adhesion to laminin of cells
from regenerating liver. Proc Natl Acad Sci USA 1981; 78:
2403– 6.
0
6–02–03–0181. The authors thank Mr Lee Eng Kee for
1
5. Terranova VP, Rohrbach DH, Martin GR. Role of laminin in the
preparation of immunohistochemical staining and Miss
Angela Ng for assistance in preparation of this
manuscript.
attachment of PAM 212 (epithelial) cells to basement membrane
collagen. Cell 1980; 22: 719–26.
16. Fisher ER. Ultrastructure of the human breast and its disorders. Am
J Clin Pathol 1976; 66: 291–5.
1
7. Gould VE, Miller J, Jao W. Ultrastructure of medullary, intraductal,
tubular, and adenocystic breast carcinomas. Comparative patterns of
myoepithelial differentiation and basal lamina deposition. Am J
Pathol 1975; 78: 401–16.
Address for correspondence: Dr Wei-qiang Zheng, Room 406, No. 89,
Lane 871, Xiang-Yin Road, Shanghai 200433, China. Email:
zhengwqk@online.sh.cn
1
8. Ozzello L. The behaviour of basement membranes in intraductal
carcinoma of the breast. Am J Pathol 1956; 35: 887–900.
9. Pitelka DR, Hamamoto ST, Taggart BN. Basal lamina and tissue
recognition in malignant mammary tumours. Cancer Res 1980; 40:
1600–11.
1
References
1. Bornstein P, Sage H. Structurally distinct collagen types. Annu Rev
Biochem 1980; 19: 1053–62.
. Kefalides NA. Structure and biosynthesis of basement membranes.
Int Rev Connect Tissue Res 1973; 6: 63–104.
. Chung AF, Jaffe R, Freeman IL, et al. Properties of a basement
membrane-related glycoprotein synthesized in culture by a mouse
embryonal carcinoma-derived cell line. Cell 1979; 16: 277–87.
20. Willibrand D, Bosman FT, Goeij AFPM. Patterns of basement
membrane deposition in benign and malignant breast tumours.
Histopathology 1986; 10: 1231–41.
21. Barsky SH, Togo S, Garbisa S, Liotta LA. Loss of basement
membrane components by invasive tumors but not by their benign
counterparts. Lab Invest 1983; 49: 140–7.
2
3