Modeling and targeting metastatic relapse in colorectal cancer

Abstract

Colorectal cancer (CRC) kills around 700,000 people worldwide every year. The majority of these deaths are the result of dissemination of the disease to foreign organs. Despite undergoing curative resection of the primary tumor, 30-40% CRC patients will relapse in the following years. In these patients, residual disseminated tumor cells (DTCs) are undetectable until they regenerate metastatic disease. The identity and features of residual tumor cells and their niches have remained elusive due to the impossibility of analyzing this clinically occult population in patients.

In the first chapter of this thesis, we discovered that genes associated with elevated risk of relapse in human patients are expressed by a defined subset of primary tumor cells that we named High Relapse Cells (HRCs). HRCs are abundant at invasion fronts, retain an epithelial program and express genes involved in cell adhesion, locomotion and extracellular matrix remodeling. To investigate HRCs, we established a human-like CRC mouse model that undergoes metastatic relapse following surgical resection of the primary tumor. We also developed methodology to isolate residual disseminated tumor cells before metastases are detectable. Single cell profiling demonstrated that residual tumor cells occult in mouse livers after primary CRC surgery resembled the HRCs present in patients. Using Emp1 (epithelial morphogenic protein 1) as a marker gene for HRCs, we tracked and selectively eliminated this cell population. Genetic ablation of HRCs prior to extirpation of the primary CRC prevented metastatic recurrence and mice remained disease-free after surgery.

In the second chapter, we tackled how the tumor microenvironment (TME) changes over time during the formation of metastases. We discovered that at the onset of metastasis, pre-existing T cell immunity against the primary CRC can eliminate DTCs as they reach the liver. During this phase, neoadjuvant checkpoint immunotherapy given before surgical removal of the primary tumor is sufficient to eradicate micrometastases. Yet, this curative effect is restricted to a narrow temporal window due to a rapidly evolving TME. By profiling at the single-cell level metastatic lesions at different stages, we found that the TME of metastases becomes complex over time. Thus, to be efficacious, immunotherapeutic treatment must be tailored to these different stages of progression.

The last chapter describes how kinetics of metastatic relapse are influenced by the genetic makeup of CRCs. Tumors with an incomplete set of driver mutations retain niche dependencies. As a result, DTCs transiently enter a latent state resistant to chemotherapy. Eventually, after cessation of therapy, DTCs resume growth and regenerate metastases.

Our findings reveal the features of the tumor cell population responsible for CRC recurrence and anticipate that therapies targeting HRCs may prevent disease relapse. Moreover, we demonstrated that residual disease is a particular state of unique vulnerability that can be targeted using immunotherapies, when applied in a timely manner. The timing of such therapeutic window will be likely determined by the genetics of each CRC.