Abstract P2-06-04: Clinical Association of Invasion Factors UPA and PAI-1 and Classical Clinicopathological Parameters in Early Breast Cancer Patients: Implication for Individualized Therapy Decisions

Autor: H. Saadoun, William Jacot, P-J Lamy, S Pouderoux, S. Thezenas, Gilles Romieu
Rok vydání: 2010
Předmět:
Zdroj: Cancer Research. 70:P2-06
ISSN: 1538-7445
0008-5472
Popis: Purpose: A strong prognostic impact of urokinase type plasminogen activator (uPA) and its inhibitor plasminogen activator inhibitor type 1 (PAI-1) as individual factors is well established, with a level of evidence I according to the ASCO recommendations, and can be used in early breast cancer (EBC) adjuvant treatments discussions. However, the association or independence of these factors with other determinant clinicopathological parameters remains unclear in this setting. The independence or correlation of these factors in the adjuvant population is evaluated here. Patients and methods: 556 EBC patients operated between January 2006 and December 2009 (mastectomy, 99 patients; breast conservative surgery, 457 patients) were analyzed in this retrospective study. Inclusion criteria were: (i) patient curatively operated for EBC, (ii) no previous diagnosis of EBC or another malignant disease, (iii) no neoadjuvant treatment, (iv) no evidence of distant metastasis, (v) unilateral disease. Classical clinicopathological prognostic and predictive factors were considered, namely age, sex, type of surgery, tumor size, nodal status, histological subtype, SBR grade, mitotic activity, peritumoral vascular invasion, hormonal receptor status (ER, PgR), HER-2 expression, KI67 staining and follow-up informations. The levels of uPA and PAI-1 were centrally measured in our specialized laboratory using the Femtelle ELISA Kit. The positivity limits were those commonly used (uPA >3ng/mg of cytosolic protein and/or PAI-1 >14ng/mg of cytosolic protein). Correlations of the classical clinicopathological parameters with the levels of uPA and PAI-1 were analyzed in order to find statistical associations precluding their concomitant use in adjuvant decision making. Two-sided P-values below 0.05 were considered to be statistically significant. Results: A total of 556 patients were analyzed. Median patients age was 54 years (range 26-85 years). 424 patients (76.3 %) were node negative. Peritumoral vascular invasion was found in 30% of the population (162 patients). uPA and PAI-1 levels were elevated in 254 (45.7%) and 302 (57.6%) of the samples respectively. 206 tumors (37.1%) presented an elevation of UPA and PAI-1 levels. In multivariate analysis, SBR grade, mitotic count, ER status, HER-2 over-expression and triple negative status were found to be significantly associated with the uPA and/or PAI-1 levels. No statistically significant association was found between uPA and/or PAI-1 levels and other classical prognostic factors such as age, tumour size, menopausal status and nodal status. Interestingly, uPA and PAI-1 levels were not statistically associated with the presence of peritumoral vascular invasion, with P-values of 0.39, 0.60 and 0.66 for uPA, PAI-1 and the association uPA and PAI-1 levels respectively. Conclusion: To our knowledge, this is the first clinical study reporting the absence of statistical association of UPA and/or PAI-1 levels and peritumoral vascular invasion. Considering the prognostic and decision-making impact of these two parameters, the absence of statistical association allow to considered these parameters altogether in adjuvant treatment discussion. Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P2-06-04.
Databáze: OpenAIRE
Externí odkaz: