Cancer Stem Cells, Radiotherapy, and Feedback

Background: Cancer stem cells (CSCs), known as cancer initiating cells, drive tumorigenesis through self-renewal/differentiation and mutations. Multiple studies have shown that xenotransplantation of human cancer initiating cells into healthy mice tissue possess the ability to regrow the tumor. Radiotherapy is a common treatment for many types of cancers including breast, prostate, brain, and skin; however, it has been shown that sublethal doses of radiation can induce the tumor to recur and become much more aggressive. It is believed that regrowth of treated tumors is driven by CSCs which are much more resistant to ionizing radiation compared to the differentiated non-stem cells. Recent experimental data suggests that sublethal radiation can induce dedifferentiation of mature cancer cells into CSCs, thereby increasing the overall number of CSCs. Methods: We investigate the effects of dedifferentiation during radiotherapy using a mathematical model. The model incorporates cell-cell communication and feedback among the different cell types, and the effect of ionizing radiation. We parameterize the model to fit recent experimental, single-dose radiation data. The experiment from which we obtained data describes the identification and tracking of cancer stem cells in their experiments, which is achieved by an imaging system based on the lack of 26S proteasome activity in cancer initiating cells. Results: In parameterizing our model, we find that dedifferentiation rate increased exponentially with the radiation dose. The sensitivity of cancer stem cells to non-stem cells (through feedback regulations) in terms of self-renewal and division capabilities was decreased. The division rate of non-stem cells also decreased due to irradiation; a phenomenon supported by recent findings. High doses of radiation, single or fractionated, eventually yield in an increase in cancer initiating cells that leads to tumor recurrence. In the case of fractionated radiotherapy, various methods were performed to match characteristics of cancer stem cell percentage in a different, recent experiment. Our model makes testable predictions on the outcome of a given radiotherapy protocol. It also suggests an optimal radiation schedule that is predicted to best reduce tumor size and result in a small population of CSCs. Conclusion: While studies have been performed for relatively simple models of tumor growth, there is a lack of studies that have accounted for the reprogramming of differentiated cells that can significantly affect the outcome by facilitating tumor recurrence.