A. Scorilas et al., Overexpression of matrix-metalloproteinase-9 in human breast cancer: a potential favourable indicator in node-negative patients, BR J CANC, 84(11), 2001, pp. 1488-1496
Matrix metalloprotease-9 (MMP-9; 92 kDa type IV collaganase, gelatinase B)
is regarded as, important for degradation of the basement membrane and extr
acellular matrix during cancer invasion and other tissue-remodelling events
. In this study we evaluate the prognostic value of MMP-9, by immunoperoxid
ase staining in a series of 210 breast cancer tissues. The results were qua
ntitated using the HSCORE system, which consider both staining intensity an
d the percentage of cells stained at given intensities. MMP-9 status was co
mpared with the concentration of cytosolic Cathepsin-D and with other estab
lished prognostic factors, in terms of disease free survival and overall su
rvival, The median follow-up period was 62 months. MMP-9 staining was obser
ved primarily in cancer cells, and to a lesser degree in surrounding stroma
[ cells. MMP-9 expression was not detected in normal breast tissue. Levels
of MMP-9 expression below the cut-off point were more frequently observed i
n larger (P= 0.014), invasive ductal histologic (P = 0.037), progesterone r
eceptor (PR)-negative and PR-strong positive tumours (P < 0.001), as well a
s samples belonging to patients with stage III-IV disease (P = 0.009) and a
ge 45-55 years (P = 0.011). in univariate analysis, node-negative breast ca
ncer patients with tumors positive for MMP-9 had a considerable reduction i
n risk for relapse (RR = 0.45; P = 0.039) or death (RR = 0.32; P = 0.009).
Multivariate analysis indicated that MMP-9 status was an independent favour
able predictor of OS (RR = 0.47, P = 0.034) in node-negative but not in nod
e-positive patients. Our results suggest that MMP-9 may be an independent f
avourable prognostic factor in node-negative breast cancer patients. The ov
erexpression of MMP-9 in breast cancer may be also used as a marker to subd
ivide node negative breast cancer patients in order to determine the optima
l treatment modality. (C) 2001 Cancer Research Campaign.