An Anoikis Resistance-Based Prognostic Model Reveals Immune Microenvironment Insights in Advanced Lung Squamous Cell Carcinoma

Purpose The prognosis for those afflicted with advanced lung squamous cell carcinoma (LUSC) is daunting, largely due to the malignancy's relentless progression, marked by invasive tumor growth, distant metastasis, a defiant stance against therapeutic interventions, and a profound reshaping of the tumor's microenvironment. The occurrence of anoikis resistance, which is of vital significance in cancer progression, continues to be a perplexing aspect of the LUSC story, especially regarding its impact on the complex interactions within the tumor's immune microenvironment (TIME). Methods LASSO and univariate Cox regression model analysis were used to identify anoikis-resistant genes (ARGs) that can serve as prognostic indicators for LUSC. A prognostic risk model was created, and its predictive accuracy was verified through both internal and external validation datasets. To gain deeper insight into the underlying biological processes, we performed single-cell analysis and enriched pathway analysis for the genes within the model. Additionally, we utilized CIBERSORT, the ESTIMATE algorithm, and the TIDE score to examine the relationship between the model's risk score and various components of the immune microenvironment, including the degree of immune cell infiltration, their characteristics, and their responsiveness to immunotherapy. The significance of key genes was experimentally confirmed. Results Our approach successfully stratified the TCGA-LUSC cohort into two distinct risk subgroups, demonstrating significant prognostic differences. The predictive power of our risk model was corroborated through validation with multiple datasets. The study revealed a substantial association between the TIME characteristics and the potential response to immune checkpoint inhibitors (ICIs) with the identified prognostic genes and risk scores. Notably, three genes (SDCBP, RPS6KA1, ITGA3) were pinpointed as pivotal in anoikis resistance, with SDCBP being experimentally confirmed as an oncogene. Conclusion This study introduces and validates a novel prognostic risk model for LUSC that is informed by anoikis resistance. The model not only provides robust predictions of overall survival for LUSC patients but also uncovers the interrelationship of the risk score and the TIME of LUSC. This contributes novel insights and tools for informing immunotherapy strategies in LUSC treatment.