Prognostic value of anoikis-related genes revealed using multi-omics analysis and machine learning based on lower-grade glioma features and tumor immune microenvironment

Background: The investigation explores the involvement of anoikis-related genes (ARGs) in lower-grade glioma (LGG), seeking to provide fresh insights into the disease's underlying mechanisms and to identify potential targets for therapy. Methods: We applied unsupervised clustering techniques to categorize LGG patients into distinct molecular subtypes based on ARGs with prognostic significance. Additionally, various machine learning algorithms were employed to pinpoint genes most strongly correlated with patient outcomes, which were then used to develop and assess risk profiles. Results: Our analysis identified two distinct molecular subtypes of LGG, each with significantly different prognoses. Patients in Cluster 2 had a median survival of 2.036 years, markedly shorter than the 7.994 years observed in Cluster 1 (P < 0.001). We also constructed a six-gene ARG signature that efficiently classified patients into two risk categories, showing median survival durations of 4.084 years for the high-risk group and 10.304 years for the low-risk group (P < 0.001). Significantly, the immune profiles, tumor mutation characteristics, and drug sensitivity varied greatly among these risk groups. The high-risk group was characterized by a "cold" tumor microenvironment (TME), a lower IDH1 mutation rate (61.7 % vs. 91.4 %), a higher TP53 mutation rate (53.7% vs. 38.9 %), and greater sensitivity to targeted therapies such as QS11 and PF-562271. Furthermore, our nomogram, integrating risk scores with clinicopathological features, demonstrated strong predictive accuracy for clinical outcomes in LGG patients, with an AUC of 0.903 for the first year. The robustness of this prognostic model was further validated through internal cross-validation and across three external cohorts. Conclusions: The evidence from our research suggests that ARGs could potentially serve as reliable indicators for evaluating immunotherapy effectiveness and forecasting clinical results in patients with LGG.