MARKERS OF TUMOR ANGIOGENESIS AND PROTEOLYSIS INDEPENDENTLY DEFINE HIGH-RISK AND LOW-RISK SUBSETS OF NODE-NEGATIVE BREAST-CANCER PATIENTS

Citation
U. Eppenberger et al., MARKERS OF TUMOR ANGIOGENESIS AND PROTEOLYSIS INDEPENDENTLY DEFINE HIGH-RISK AND LOW-RISK SUBSETS OF NODE-NEGATIVE BREAST-CANCER PATIENTS, Journal of clinical oncology, 16(9), 1998, pp. 3129-3136
Citations number
52
Categorie Soggetti
Oncology
ISSN journal
0732183X
Volume
16
Issue
9
Year of publication
1998
Pages
3129 - 3136
Database
ISI
SICI code
0732-183X(1998)16:9<3129:MOTAAP>2.0.ZU;2-6
Abstract
Purpose: To compare the prognostic impact of tumor angiogenesis factor s (vascular endothelial growth factor [VEGF], angiogenin, and basic fi broblast growth factor [bFGF]), tumor proteolysis factors (urokinase-t ype plasminogen activator [uPA] and plasminogen activator inhibitor-1 [PAI-1]), and conventional tumor markers (stage, grade, and steroid re ceptors) in early breast cancer. Patients and Methods: In the primary clinical study, tumor angiogenesis and other factors were detected in frozen biopsies from 305 primary breast tumors. VEGF expression was as sessed by chemiluminescence immunosorbent assay (ICMA); angiogenin, bF GF, uPA, and PAI-1 by enzyme-linked immunosorbent assay (ELISA); and s teroid receptors (estrogen receptor [ER] and progesterone receptor [Pg R]) by enzyme immunoassay (EIA). In the validating clinical study anot her set of 190 node-negative primary breast tumor samples were collect ed at a separate institution. Results: Univariate analysis of the prim ary study showed that VEGF levels were positively correlated with recu rrence (P < .001). Angiogenin levels were positively correlated with d isease relapse (P < .005) for the overall collective group, but not wi thin the node-negative subset. No significant correlations were found between tumor bFGF levels and patient survival. In multivariate regres sion analysis, the only independent predictors of relapse-free surviva l (RFS) were VEGF, uPA, and lymph node status. In the validation set, the distribution of VEGF and uPA values were similar to those in the p rimary study; low expression of both VEGF and uPA identified patients with a less than or equal to 20% likelihood of recurrence within 7 yea rs. Conclusion: Separate primary and validating clinical studies concu r that tumor VEGF level is the most important prognostic parameter amo ng several markers of tumor angiogenesis and proteolysis. I Clin Oncol 16:3129-3136. (C) 1998 by American Society of Clinical Oncology.