Urokinase-type plasminogen activator gene (PLAU) to predict clinical outcome in invasive lobular carcinoma.

Abstract

e21010 Background: At the molecular level invasive lobular carcinoma (ILC) is mostly composed of luminal A (LA) (≈80%) followed by luminal B (≈15%) and a small fraction of HER2-positive (≈5%) tumors (Metzger et al SABCS2011). In ILC, Genomic Grade (GG) adds prognostic information to clinico-pathological (CP) characteristics (Metzger et al ASCO 2011). In this study we sought to evaluate the prognostic value of different gene signatures/modules in patients diagnosed with LA classic ILC. Methods: Gene expression data were generated from 184 consecutive frozen tumor samples using Affymetrix U133 Plus 2.0 arrays. ILC tumors were classified into molecular subtypes using the PAM50 classifier. ILC tumors characterized as classic ILC by pahtologists and as LA by PAM50 were selected for this analysis. Invasive disease free survival (IDFS) was defined as the interval between date of surgery and date of any invasive recurrence or death. Multivariate analyses for IDFS and overall survival (OS) adjusted for age, tumor size, nodal status and chemotherapy was performed for gene signatures/modules related to proliferation, invasion, immune response and signaling pathways (IGF1, PI3K, MAPK, Src, Wnt). Results: 125 LA classic ILC tumors were identified with a median follow-up of 6.6 years (95% CI 5.6-7.6). The addition of PLAU gene expression as continous variable to CP variables yielded a significant hazard ratio (HR) for IDFS (HR= 2.1 [1.1 – 4.1], Δx2= 4.5, likelihood ratio p = 0.03) and for OS (HR = 3.6 [1.4 – 9.6], Δx2= 6.9, likelihood ratio p = 0.009). GG as a continous variable yielded a significant HR for IDFS (HR= 2.6 [1.2 – 5.7], Δx2= 6.2, likelihood ratio p = 0.01), but not for OS (HR= 2.1 [0.8 – 5.5], Δx2= 2.5, likelihood ratio p = 0.11). In contrast, IGF1, PI3K, MAPK, Src, Wnt, immune response, gene-70 and gene-76 failed failed to add signficant independent prognostic value. Conclusions: In addition to GG, PLAU adds prognostic information to CP factors in LA classic ILC and may guide future drug development in this difficult to treat BC subset. Of interest, PLAU is involved in degradation of extracellular matrix contributing to cell migration and metastasis and is a major component of a validated breast cancer biomarker (UPA/PAI1).