Classification of cancer cell lines on their radiosensitivity by machine learning

The outcomes of radiotherapy (RT) of cancer patients significantly depend on the radiosensitivity of tumor to ionizing radiation. The degree of radiosensitivity (RS) and radioresistance (RR) of the tumor is clinical predictor of the therapeutic responce of oncopatients to RT and should be considered as a key factor in RT treatment planning in defining the delivered dose, fractionation, and the duration of the RT course. In this work, we developed a method for determining cancer cell RS/RR based on the analysis of experimental data on clonogenic survival of cancer cells using machine learning. A combination of the clustering methods with the principal component analysis was applied to discriminate clusters of RS and RR cancer cells using parameters of dose dependencies of cancer cell survival. Based on the obtained results, a statistical model was developed and trained on a dataset of experimental data and was successfully validated to determine the radiosensitive and radioresistance cancer cells.

Keywords:Radiotherapy; Radioresistance; Ionizing radiation; Clonogenic survival analysis; Classification; Machine learning