PLASMINOGEN-ACTIVATOR INHIBITOR TYPE-1 IN CYTOSOLIC TUMOR EXTRACTS PREDICTS PROGNOSIS IN LOW-RISK BREAST-CANCER PATIENTS

We have reported previously that both urokinase-type plasminogen activ ator (uPA) and its type 1 inhibitor (PAI-1) are statistically signific ant prognostic variables in patients with high risk breast cancer (Gro ndahl-Hansen et al., Cancer Res., 53:2513-2521, 1993), and we recently described that the uPA receptor (uPAR) is a prognostic marker in post menopausal, node-positive breast cancer patients (Grondahl-Hansen et a l., Clin. Cancer Res., 1:1079-1087, 1995). The present retrospective s tudy describes the prognostic impact of uPA, its receptor uPAR, and PA I-I in breast cancer cytosol from 111 low-risk premenopausal patients and 184 low-risk postmenopausal patients with a median follow-up of 6. 0 years (range, 3.8-14.9) and 7.4 (range, 3.7-14.0) years, respectivel y. uPA, uPAR, and PAI-1 levels were determined by sandwich enzyme-link ed immunosorbent assays, and data were dichotomized using the median v alue as the cutoff for calculation of recurrence-free survival and ove rall survival. A correlation was found between the levels of each of t he three molecules. In univariate analysis, high PAI-1 was significant ly associated with short overall survival in postmenopausal patients [ relative risk (RR), 2.3; 95% confidence interval (CI), 1.3-4.3; P=0.00 5] and shorter recurrence-free survival in both premenopausal (RR, 2.5 ; 95% CI, 1.3-4.7; P=0.004) and postmenopausal (RR, 1.8; 95% CI, 1.1-2 .9; P=0.02) patients. Neither uPA nor uPAR reached statistical signifi cance in the univariate analyses. The prognostic value of uPA, uPAR, a nd PAI-1 was then compared with that of other established prognostic v ariables by multivariate analysis. PAI-1 was an independent prognostic variable for recurrence-free survival in premenopausal patients, with a RR of 2.6 (95% CI, 1.3-5.0). For recurrence-free survival (RR, 1.9; 95% CI, 1.1-3.5) and overall survival (RR, 2.6; 95% CI, 1.2-5.7) in p ostmenopausal patients, PAI-I was the only independent variable left i n this group of patients. Neither uPA nor uPAR reached significance in the multivariate analysis. These data, together with previously publi shed data on the prognostic significance of components of the urokinas e plasminogen activation system in breast cancer cytosols, strongly in dicate that PAI-1 is a statistically significant and independent progn ostic marker in both low- and high-risk breast cancer patients.