Abstract
Urokinase-type plasminogen activator (uPA) is a potentially important prognostic factor in breast cancer for identifying patients at high risk of recurrence. This retrospective study assessed two enzyme-linked immunosorbent assay (ELISA) methods measuring uPA antigen levels in 499 primary breast cancer cytosols. Both uPA methods were applied to cytosols used routinely for oestrogen (ER) and progesterone (PgR) receptor assays. uPA was determined using a classical ELISA method (Imubind; American Diagnostica) and a novel automatic immunoluminometric assay (Lia; Sangtec Medical). The uPA Imubind method revealed about twice as much uPA antigen (median 0.75 ng mg(-1) protein) as the uPA Lia method (median 0.38 ng mg(-1) protein). The correlation coefficient between the two methods was acceptable (r = 0.81), but the two techniques are not interchangeable. Univariate analyses confirmed the poor outcome of patients whose tumours contained large amounts of uPA, regardless of the technique used. Multivariate analyses showed that uPA Imubind and uPA Lia values were both strong independent prognostic factors.
Highlights
The study group consisted of 499 breast cancer patients treated at the Centre Rene Huguenin (CRH) between 1981 and 1989
Patients were selected according to the following criteria: (1) primary, unilateral breast tumour; (2) full follow-up at CRH; (3) previously untreated, without evidence of metastatic disease or any other malignant tumour at the time of diagnosis; (4) surgery as the first treatment; and (5) complete clinical, histological and biological information, especially concerning hormone receptors and antigen levels of Urokinase-type plasminogen activator (uPA) measured in cytosols by the two methods
For uPA Imubind and uPA Lia variables and according to each outcome (OS, disease-free survival (DFS) and metastasis-free survival (MFS)), the cut-off points were determined by using the 'minimum P-value' (Hilsenbeck et al, 1996) method, which chooses the cut-off points that minimize the P-value relating the variables to outcome measure
Summary
The aim of our work was to test a new automated ELISA
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
