Herein we present the results of two related investigations. The first study determined if concentrations in breast nipple discharge (ND) of two proteins (urinary plasminogen activator, uPA and its inhibitor, PAI-1) predicted the presence of breast atypia and cancer in pre- and/or postmenopausal women requiring surgery because of a suspicious breast lesion. The second study assessed if these proteins increased the predictive ability of a carbohydrate (Thomsen Friedenreich, TF) which we previously demonstrated predicted the presence of disease in postmenopausal women requiring surgery.
In the first study we prospectively enrolled 79 participants from whom we collected ND, measured uPA and PAI-1 and correlated expression with pathologic findings. In the second study we analyzed 35 (uPA and PAI-1 in 24, uPA in an additional 11) ND samples collected from different participants requiring breast surgery, all of whom also had TF results.
uPA expression was higher in pre- and PAI-1 in postmenopausal women with 1) cancer (DCIS or invasive) vs. either no cancer (atypia or benign pathology, p = .018 and .025, respectively), or benign pathology (p = .017 and .033, respectively); and 2) abnormal (atypia or cancer) versus benign pathology (p = .018 and .052, respectively). High uPA and PAI-1 concentrations and age were independent predictors of disease in premenopausal women, with an area under the curve (AUC) of 83-87% when comparing diseased vs. benign pathology. uPA, TF, and age correctly classified 35 pre- and postmenopausal women as having disease or not 84-91% of the time, whereas combining uPA+PAI-1+TF correctly classified 24 women 97-100% of the time.
uPA and PAI-1 concentrations in ND were higher in women with atypia and cancer compared to women with benign disease. Combining uPA, PAI-1 and TF in the assessment of women requiring diagnostic breast surgery maximized disease prediction. The assessment of these markers may prove useful in early breast cancer detection.
Keywords: Breast cancer, Premalignant lesions, Biomarkers, Cancer prediction