Abstract P2-10-36: Analysis of biomarker expression and biological subtype in primary tumour, corresponding lymph node and distant metastasis with 5-year follow-up

Abstract Background: Biological variation in genotype and phenotype of primary breast cancer explains the heterogeneity of the disease and subsequent outcome. Choice of treatment is routinely based on the biomarker expression of the primary tumour with regard to estrogen- (ER) and progesterone (PR) receptor status, human epidermal growth factor receptor 2 (HER2) status and the proliferation marker Ki67, individually. Recently, gene expression profiling and subsequent immunohistochemical studies have identified biologically distinct subtypes by using a combination of biomarkers which have been shown to be of prognostic value beyond the previous known. However, tumour progression theories propose a change in biomarker expression during development from primary tumour to distant metastasis with implications on choice of therapy. The aim of this study was to compare biomarker expression individually and according to biological subtype between tumour samples from primary tumour, lymph node metastasis and distant metastasis from the same patient. Material and methods: From an original cohort of 569 patients with primary breast cancer, tissue microarrays were constructed from archival tissue blocks of primary tumour (n = 521), paired lymph node metastasis (n = 147) and biopsies from distant metastasis (n = 42). The samples were evaluated by two independent pathologists. The individual biomarker expression as well as subtype classification (Luminal A: ER+ and/or PR+, Ki67≤ 20% and HER2−. Luminal B: ER+ and/or PR+, Ki67> 20% and/or HER2+. HER2 type: ER/PR−, HER2+. Basal-like: ER/PR−, HER2−, CK5/6+ and/or EGFR+) were compared between primary tumour, lymph node metastasis and distant metastasis. Survival outcome were estimated using the Kaplan-Meier method and log rank test. The primary end-point was distant disease-free survival (DDFS). Results: Distribution according to subtype. The molecular subgroups in the primary tumours were associated with statistically significant prognostic information by log-rank test (p = 0.005) validating luminal A as a subgroup with excellent prognosis. More detailed survival analysis comparing expression of biomarkers between primary tumour, lymph node metastasis and distant metastasis will be presented. Conclusion: Prognostic information can be obtained by subtype classification using routine-based biomarker analysis in the primary tumour. Preliminary data suggest that the subtype distribution is associated with a worse prognostic profile in the lymph node- and distant metastasis. Detailed survival analysis will be presented at the meeting. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P2-10-36.