Background: Liver hepatocellular carcinoma (LIHC) remains a malignant malignancy with a low cure rate. Anoikis is a newly recognized cancer hallmark. However, an Anoikis-related model has not been clarified in LIHC.
Methods: The Anoikis-related score in the present study was created using Survival Random Forest and least absolute shrinkage and selection operator (LASSO) machine learning algorithms. Anoikis-related scores with respect to mutation analysis, immunological analysis, function annotation, and medication prediction were all thoroughly investigated.
Results: The Anoikis-related score accurately predicted the patients' immunological activity, altered genes, and medication sensitivity. SPP1 immunological analysis, function annotation, medication prediction, and immunotherapy prediction were systematically investigated. SPP1 may effectively predict the outcomes of immunotherapy. SPP1 was revealed to be a mediator of LIHC cell proliferation and migration. A putative axis in LIHC was YBX1/SPP1.
Conclusions: Clinical care and the treatment plan for patients with LIHC were anticipated to benefit significantly from the established Anoikis-related score.
Keywords: Anoikis; LIHC; immunotherapy; machine learning; tumor microenvironment.
© 2023 John Wiley & Sons Ltd.