Comparative studies in the clinical and molecular features of breast cancer and brain metastases

Abstract

We hope to understand better the unique pattern of CNS metastases from breast cancer, hopefully adding to the field and ultimately increasing the chances of preventing and treating them. In this thesis, we approached the aim using clinical data and material from patients suffering from CNS metastasis from breast cancer from 1994-2014. The methods in the thesis involved a re-evaluation of immunohistochemical subtypes in the material, a genetic study using a 50-gene gene panel, gene expression analysis using a 770-gene panel, and an immunohistochemical analysis of immune cell populations in primary and CNS metastatic breast cancer. In paper I, we found an increasing, but fluctuating, number of CNS metastases per diagnosed primary breast cancer yearly and more patients receiving adjuvant treatments in the second ten-year period. A growing number of patients with CNS metastases, despite increased time from primary tumor to CNS metastasis, might indicate that the CNS is a sanctuary site for breast cancer cells and that despite more prolonged survival, CNS metastases are not prevented. In paper II, we noted discordant receptor subtypes in 25% of paired primary tumors and metastases. Similarly, actionable genes differed, with gains and losses in the matched material. The most commonly mutated genes were TP53 and PIK3CA (44% and 20%, respectively). There was no correlation between the presence of any particular mutation, change in mutation, and change in immunohistochemical subtype or survival parameters. Most gene expression signatures were downregulated in the CNS metastases compared to the matched primary tumors in paper III. The expression signatures involved in immune signaling were all downregulated in CNS metastases, prompting us to investigate immune cell abundance and distribution. There were lower fractions of regulatory T-cells, cytotoxic T-cells, and CD4+ T-helper cells / mm2. Macrophages were present in a higher fraction. To conclude, there is a need to develop noninvasive techniques to determine the subtype of CNS metastasis as it is increasingly occurring. Subtype changes are paramount to treatment decisions, and subtype changes occur in roughly 25% of patients with CNS metastases. The immune landscape of CNS-metastases is altered compared to primary tumors, perhaps explaining the limited efficacy of traditional immunotherapy. Future studies might focus on including patients in clinical and pre-clinical studies and focus on the prevention as well as the treatment of CNS metastasis