Neoadjuvant radiotherapy (NeoRT) is used in many types of cancers and aims at improving tumor local control and patient overall survival. The RT schedule and the timing of surgery are mostly empirically based on clinical experiences and side effect occurrence. However, a clinical study revealed that the timing of surgery following neoRT might be crucial for patient overall survival. Therefore, we hypothesized that RT may influence the tumor phenotype as well as the tumor microenvironment (TME) and consequently metastasis formation. As a “proof of concept”, we developed a pre-clinical model to study the impact of different RT schedules and timings of surgery on TME and metastatic dissemination and search for predicting markers thanks to metabolic analyses.
To mimic neoRT used in clinic, we subcutaneously injected human cancer cells (MDA-MB-231) into the flank of SCID mice. When tumors reached an appropriate volume, we locally irradiated the primary tumor with different neoRT schedules (5x2Gy and 2x5Gy) inspired from clinical practice but adapted to mice. We surgically removed carefully tumors 4 or 11 days after the end of RT and kept the mice alive during 6 weeks for metastatic growth. Tumor samples were collected for extractions and histological analyses. Lungs were harvested at the end of the experiments for metastatic burden analysis.
The occurrence of lung metastases was totally different according to the neoRT schedule and the time of surgery. After 2x5Gy, the size and the number of lung metastases were smaller when surgery was performed at 11 days after the end of RT, compared to 4 days. Inversely, in the 5x2Gy schedule, applying surgery at 4 days protected the mice against lung metastases compared to surgery at 11 days. These results demonstrate that the timing of surgery and RT schedules are both important factors that influence the formation of metastases.
To decipher RT-induced TME modifications, we studied tumor features known to be implicated in tumor progression, such as hypoxia, blood vessel density, matrix metalloproteinases, adipocyte-like cells, inflammation. Although none of the observed RT-induced TME modifications was directly related to the metastatic burden, we pointed out a regulation of NK cell recruitment by neoRT, affecting metastatic dissemination. Our second aim was to identify predictive markers of the course of the disease for optimizing treatment. Thanks to nuclear magnetic resonance analysis and powerful statistical tools, the study of primary tumors at the time of surgery showed different metabolic profiles according to the RT schedule. Moreover, a correlation between this metabolic profile and metastatic burden was observed.
Altogether, we provide, for the first time, experimental clues that neoadjuvant RT schedule and the timing of surgery influence the metastatic profile. Although researches are needed to elucidate the underlying mechanisms, our work, which draws the attention on therapy schedule, has clinical implication. Moreover, nuclear magnetic resonance and discriminant analyses showed an impact of neoRT on TME, which could be related to the metastatic profile, and are of interest to be studied in clinic.