Challenging tumour aggressiveness – Individualising patient management

Abstract

Breast cancer therapy is currently undergoing a major challenge: Effective early detection due to more widespread screening efforts worldwide has resulted in an increasing percentage of small early stage breast cancers at initial diagnosis. The majority of these early cancers will be cured by loco-regional therapy and thus do not require any aggressive adjuvant (chemo-) therapy. In the age of molecular medicine, established clinicopathological factors are not sufficient to reliably characterise tumour aggressiveness and estimate an individual patient’s risk for recurrence. Gene or protein profiling of the primary tumour constitute promising options for gaining clinically meaningful prognostic and predictive information. The plasminogen activator (PA) system comprises among others the serine protease urokinase-type plasminogen activator u-PA, its cell surface receptor (u-PAR), as well as its type-1 inhibitor PAI-1. This complex system is not only able to catalyse plasminogen into plasmin on the cell surface, leading to degradation of extracellular proteins, but also has multiple interactions between its members as well as with proteins of the extracellular matrix, such as vitronectin and its integrin receptor. Well balanced production and activation of uPA/PAI-1 system components lead to changes in cell adhesion, degradation of the extracellular matrix and consequent stimulation of proliferation, angiogenesis, and tumour cell invasion and metastasis in a variety of solid tumours. 1 Expression of uPA and PAI-1 is higher in tumours than in the surrounding tissue, which thus indicates their key role in tumour invasiveness and metastasis. With regard to reliable assessment of tumour aggressiveness, uPA and PAI-1 are the only biomarkers whose clinical efficacy has been validated at the highest level of evidence by the prospective clinical Chemo N0 trial 2 and the EORTC Receptor and Biomarker Group (EORTC RBG) pooled analysis comprising more than 8000 primary breast cancer patients. 3 The prognostic impact of uPA and PAI-1 is greatest when both factors are used in combination; 4 it is independent of the HER2 status 5 and currently it is most clinically useful in nodenegative disease in order to avoid over-treatment of lowrisk patients by adjuvant chemotherapy: Node-negative patients with low levels of uPA and PAI-1 in their primary tumour tissue have an excellent chance of surviving breast cancer with about a 95% 5-year survival even without any adjuvant systemic therapy; node-negative patients with high uPA/PAI-1 are at increased risk of relapse, comparable to patients with axillary lymph node involvement. 2 Tumour biological risk assessment using uPA/PAI-1 provides a more accurate estimate of the future course of disease than epidemiological data-based algorithms such as Adjuvant Online!, which presently rely on clinico-pathological factors. 6 The uPA/PAI-1 test has been thoroughly qualityassured by collaborative efforts of the EORTC RBG (now: EORTC Pathobiology Group, EORTC PBG). 7 Yet, widespread international acceptance of uPA/PAI-1 has been hampered by the need for fresh frozen tissue for the enzyme-linked immunosorbent assay (ELISA) test, even though the commercially available test can be used on as little tissue as 2−3 core biopsies or a few 90 mm cryostat sections. Nevertheless, in their annually updated, evidence-based guidelines, the Breast Commission of the German AGO (Working Group for Gynaecological Oncology) has accepted uPA/PAI-1 since 2002 as riskgroup classification markers for routine clinical decisionmaking in node-negative breast cancer, complementing established clinical–pathological factors. 8 Recently, based on the high level of evidence for these markers, uPA and PAI-1 have also been recommended for clinical routine use by the ASCO tumour marker panel. 9 A second confirmatory node-negative breast cancer trial NNBC-3 (AGO/GBG/EORTC/PBG) using uPA/ PAI-1 as stratification criteria and evaluating optimised chemotherapy in high-risk node-negative patients is currently open in Germany and France, with more than 2200 patients recruited at present (www.gbg-crf.com/nnbc3). In addition to NNBC-3, two other major international trials are currently ongoing, which employ molecular tests to optimise patient selection for adjuvant chemotherapy in early breast cancer: The US Trial Assigning Individualized Options for Treatment Rx (TAILORx) uses the 21gene recurrence score, 10 which has also been accepted by the ASCO tumour marker panel. 9 Microarray In Nodenegative Disease may Avoid Chemotherapy (MINDACT), run by TransBIG, uses the Amsterdam 70-gene signature for risk assessment. 11 Other gene signatures, such as