Construction of prognostic risk markers for cervical cancer combined with anoikis-related genes and their clinical significance.

Abstract

Context: Several studies have demonstrated that anoikis affects the development, metastasis and prognosis of cancer.

Aims: This study aimed to identify anoikis-related marker genes in cervical cancer (CC).

Methods: Least absolute shrinkage and selection operator (LASSO) combined with Cox regression analysis was used to construct a prognostic model and analyse the independent prognostic ability of riskscore. Receiver operating characteristic curve (ROC) and survival curves were used to evaluate and verify the performance and accuracy of the model. The nomogram of CC prognostic model was drawn using riskscore combined with clinical information. We analysed the relationship between prognostic riskscore and immune infiltration level and analysed immunophenoscore. Finally, qRT-PCR assay was used to verify the feature genes.

Key results: By Cox analysis, we found that the prognostic risk model could effectively predict the risk of CC in patients independently of other clinical factors. Both the levels of immune infiltration and the immunophenoscore were significantly lower in high-risk CC patients than those in low-risk patients, revealing that high-risk patients were likely to have bad response to immunotherapy. The qRT-PCR results of the feature genes were consistent with the results of gene expression in the database.

Conclusions: The prognostic model constructed, based on anoikis-related genes in CC, could predict the prognosis of CC patients.

Implications: The model described here can provide effective support for assessing prognostic risk and devising personalised protocols during clinical treatment.