Predicting the prognosis, immune response, and immunotherapy in head and neck squamous cell carcinoma using a novel risk model based on anoikis-related lncRNAs

Abstract

Abstract Background Head and neck squamous cell carcinoma (HNSCC) is an extremely heterogeneous and metastatic disease. Anoikis, which is a specific type of programmed apoptosis, is involved in tumor metastasis, tissue homeostasis, and development. Herein, we constructed an anoikis-related long non-coding RNA (lncRNA) signature to predict the prognosis, immune responses, and therapeutic effects in HNSCC patients. Methods A total of 501 HNSCC samples were acquired from the TCGA database and randomly classified into the training and validation groups (1:1 ratio). Thereafter, the results derived from the training set were analyzed with the LASSO regression analysis, and a novel anoikis-related lncRNA risk model was constructed. Time-dependent ROC curves and Kaplan–Meier analysis were carried out to assess the diagnostic value and survival outcomes. A nomogram was utilized to predict the prognostic accuracy. Furthermore, we studied the tumor microenvironment, tumor mutation burden, enrichment pathways, and the response to chemotherapy and immunotherapy. Results Seven anoikis-related lncRNAs (AC015878.1, CYTOR, EMSLR, LINC01503, LINC02084, RAB11B-AS1, Z97200.1) were screened to design a novel risk model, which was recognized as the independent prognostic factor for HNSCC patients. The findings implied that low-risk patients showed significantly longer OS, PFS, and DSS compared to those high-risk patients. The two groups that were classified using the risk model showed significant differences in their immune landscape. The risk model also predicted that low-risk HNSCC patients could attain a better response to immunotherapy, while high-risk patients would be more sensitive to gemcitabine, docetaxel, and cisplatin. Conclusions We constructed a novel risk model that could be employed for effectively predicting patient prognosis with a good independent prognostic value for HNSCC patients. Furthermore, this model could be used for designing new immunotherapeutic and chemotherapeutic strategies, and it helps clinicians establish personalized and detailed strategies for HNSCC patients.