Metastases are responsible for the vast majority of deaths in breast cancer patients. The goal of this project was to identify drivers of this process. To do this we are using patient-derived xenograft models where we obtained biopsies from the primary breast tumours of patients with triple-negative breast cancer. The majority of these biopsies are from patients that have not been treated with chemotherapy, and many of these patients do go on to develop metastatic disease. We implant these tumours into the mammary fat pads of immune-compromised mice and allow the tumours to grow and some of them do metastasise within the mice. We’ve been able to label these tumour cells with bioluminescent reporters so that we can track their metastasis in real time and this has allowed us to generate transcriptional signatures so we can look at the gene expression profile of cells in the metastatic site compared back to the primary tumour. When we do this, we see that several genes are upregulated in metastatic sites and several are downregulated in metastatic sites. We’re interested especially in the genes that are upregulated in the metastatic sites because these might be responsible for driving this process.

We undertook a gain-of-function high throughput screen in mice where we overexpressed the genes that were upregulated in the metastatic site, put the tumours into mice and then looked for what genes caused the tumour cells to metastasise faster. When we did this we identified a gene called CEACAM 5, which stands for carcinoembryonic antigen cell adhesion molecule five, and is also known as CEA in the clinic. It is used as a serum biomarker for colorectal cancer and so we thought that it might have some clinical relevance in breast cancer too. There’s also evidence that CEACAM 5 can serve as a tumour and serum biomarker in metastatic breast cancer.

We wanted to know if CEACAM 5 actually had a functional role in metastasis, so we overexpressed it in our breast cancer cells and we injected the tumour cells into the tail veins of mice. This models the last stage of metastasis. We found that when we overexpressed CEACAM 5, these tumours grew faster in the metastatic site, so this suggests that CEACAM 5 can drive metastatic outgrowth. To look at this process a little bit deeper, we did some intracellular or in vivo mechanistic studies, where we took our cells that overexpressed CEACAM 5 and compared to the parental cells and we found that the signalling cascades that induce a process called epithelial-mesenchymal transition or EMT are inhibited when CEACAM 5 is overexpressed. This is important because EMT is thought to be important for escape of tumour cells from the primary site, which allows them to get into the blood and go through the initial stages of metastasis. But then when these tumour cells get to the metastatic site, they upregulate expression of CEACAM 5 and this inhibits the process of EMT. This is important because when EMT is turned on tumour cells proliferate much more slowly. When CEACAM 5 is turned on this inhibits EMT and allows the tumour cells to proliferate more quickly in the metastatic site, which can lead to tumour progression in metastatic sites which is eventually what leads to patient death.

What we’ve found is that CEACAM 5 can serve as a biomarker for breast tumour cells in metastatic sites. It looks like CEACAM 5 may be useful as a biomarker for metastatic progression in breast cancer patients.