Abstract
Cancer stems cells are cells in tumors that have self-renewing capabilities and proliferation, and are partly responsible for tumor growth, metastasis and drug resistance, and have been associated with multidrug resistance and epithelial-mesenchymal transition. mRNA stemness index or mRNAsi is a machine learning tool that uses the application of algorithms to find associations between cancer stemness and tumor prognostic signatures. mRNAsi predicts gene mutation status and identifies tumor signaling pathways. Clinical tier grading is a common feature for stratifying the presenting features and symptoms of patients in several diseases. This study is a review article that summarizes studies in lung cancer, gastric cancer, hepatocellular carcinoma and glioblastoma that use mRNA stemness index machine learning tools to identify differentially expressed genes, characterize the tumor microenvironment and tumor mutational burden, and determine clinical endpoints. A prognostic signature is shown in this paper as determined by mRNAsi high and low values, and a clinical tier grading system is proposed that categorizes cancer stemness presenting characteristics. This clinical grading tier system demonstrates a relationship between cancer stemness and immune checkpoint inhibitor efficacy. This type of tiered system for cancer patients and the accompanying workflow proposed may prove useful to oncologists, and has not been performed before, and is unique in the literature.