Refining prognostication in early breast cancer: more about the tumor, less about the nodes

Abstract

Breast cancer prognosis has improved over the last two decades for an important part due to the administration of adjuvant systemic therapy. It has become evident that tumor biology, reflected by the different molecular subtypes, plays an important role in predicting breast cancer prognosis and response to treatment. Several gene-expression profiles (GEPs) have been developed to identify the intrinsic molecular subtype and to estimate the recurrence risk. As such GEPs refine prognostication and serve as an aid to guide chemotherapy decision-making. The first part of this thesis focuses on pathological and genomic classifiers of malignancy grade and molecular subtype. Our results illustrate that the distinction of Luminal A- and B-type breast cancers based on pathological examination shows poor concordance with the 70-gene signature. Our results also demonstrate that reproducibility of tumor grade assessment can be improved by using phoshohistone H3 antibodies for marking mitotic figures in breast tumors. The second part of this thesis explores various aspects of lymph node staging and reveals a growing trend towards less aggressive axillary treatment. Biology of the primary tumor has become the main driver in outcome prediction and treatment selection, while information on the extent of local disease becomes increasingly subordinate. The last part of this thesis focuses on patients and doctors’ perceptions of risk and risk assessment in the era of more extensive grading and less invasive staging. Our results illustrate various factors to be associated with variation of the decreasing trend in performing axillary lymph node dissection among Dutch hospitals and an empowering effect of GEPs on patients considering to undergo adjuvant chemotherapy.