Anoikis is a form of programmed cell death that is common in normal epithelial cells, whereas tumour cells can survive after detachment from solid tissue and can further enhance their ability to invade and metastasis. The "apoptosome" of anoikis consists of caspase-9, anoikis protease activator (APAF), and cytochrome C, which can perform the anoikis process by activating endonuclease enzymes through caspase-3[23], which is regulated by two pathways inside and outside the cell. Endogenous pathways: pro-apoptotic proteins Bax and Bak have pore-forming activity, which can form protein channels after transfer from the cytoplasm to the outer mitochondrial membrane (OMM) under the induction of the integrin pathway, and can also interact with mitochondrial channel proteins to improve their permeability. This eventually causes the release of mitochondrial cytochrome C, forming "apoptosomes"[24]. Exogenous pathways: On the surface of the cell membrane, when the tumour necrosis factor receptor (TNFR) binds to its ligand, it can cause the formation of a death-inducing signaling complex (DISC). DISC recruits caspase-8 and promotes its activation by interacting with Fas-associated death domain proteins (FADDs), which are then released into the cytoplasm to activate caspase-3[25]. At the same time, activated Caspase 8 can also promote the release of mitochondrial cytochrome C and the assembly of apoptotic bodies through T-BID, a mitochondrial outer membrane permeable protein, thereby connecting the two internal and external pathways[20].
Programmed cell death, including anoikis, necrotizing apoptosis, ferroptosis, copper death and cell pyroptosis, is a new therapeutic research direction for the treatment of tumours in recent years, aiming to induce programmed death of tumour cells or establish an immune barrier that restricts metastasis[26–28]. Programmed cell death has been studied in various aspects of LUAD, further exploring the intrinsic mechanism of LUAD, and establishing a predictive model for LUAD based on programmed cell death. For example, Song J, Sun Y, Cao H, et al. used necrotizing apoptosis-related lncRNA to construct a predictive model for LUAD[29–30]. Zhang X, Yu K, Ma L, et al. detected the crucial role of endogenous glutamate and exosomal RNA in iron death in LUAD[31]. Pan S et al. subtyped LUAD by copper death. Li G et al. used machine learning to verify the predictive effect of copper death on LUAD[32–33]. Anoikis is less studied in LUAD, but it can be seen that it is a promising research direction for LUAD.
In this study, based on bulk and single-cell sequencing data, we explored the prognostic significance of anoikis in LUAD, constructed a risk score related to anoikis based on ten key genes, and verified the reliability of the risk score in two independent test datasets. Based on the median value of the risk score, patients were divided into high and low groups, and the prognosis of patients in the high risk group was significantly lower than that in the low risk group. In addition, there were significant differences in immune invasion, tumour mutation, and immunotherapy between the two groups, which may help the treatment and prognosis of patients with clinical LUAD.
The application of immunotherapy in tumour treatment has developed rapidly, as a powerful treatment strategy, immunotherapy can mobilize the body's immune system to fight tumour cells, and has achieved initial results in many solid tumours[34–35]. However, some patients still do not respond well to immunotherapy, which may be related to mechanisms such as low mutational burden and expression of immune checkpoints[36–37]. Moreover, anti-PD1/L1 monotherapy is not effective in the treatment of some patients with LUAD, such as EGFR mutant patients, it is difficult to observe the benefit of immunotherapy[38]. At present, the combination of immunotherapy with other clinical treatment modalities has achieved important results in promoting precision medicine and personalized treatment, which may be the next direction of clinical immunotherapy[39]. In our study, patients in the high-risk group were found to have a significantly poor prognosis, which may be due to lower levels of immune cell infiltration, higher gene mutations, and activation of cell cycle-related pathways such as E2F targets and G2M checkpoints, as well as anti-nesting and anoikis-related pathways PI3K/AKT[40], MAPK pathway[41], and mTOR-related pathways[42]. In addition, the high risk group was associated with a better immunotherapy response, which may be related to the differential expression of immune checkpoint genes such as PDL1, CD276, and CTLA4 between the two groups. In conclusion, our study showed that there were different levels of immune infiltration, gene mutations, and immunotherapy in other risk groups related to anoikis, which provided a reference for the clinical treatment of LUAD patients and the study of anoikis.
Although some people have built some prediction models based on anoikis-related genes, such as ZHAO et al.[43] based on seven anoikis-related genes to construct a risk prediction model for low-grade glioma patients, it was found that the survival rate of patients in high risk groups was reduced, and there were differences in immunity, mutation and immunotherapy in different risk groups. CHI et al.[44] also constructed a prognostic model of head and neck squamous cell carcinoma associated with seven anoikis genes based on bioinformatics methods, which can predict the survival of patients with head and neck squamous cell carcinoma. In addition, Sun et al.[45] explored the genetic signature markers associated with anoikis in glioblastoma to predict the relevant immune, stem, and mutation-related clinical features in glioblastoma patients. Compared with their screening of anoikis related genes based only on bulk data, we analyzed the anoikis-related genes of LUAD patients by combining bulk with single-cell sequencing data, and determined that the predictive model constructed by ten genes may help the treatment and prognosis of clinical LUAD patients.
However, our research still has shortcomings, such as the lack of relevant experiments to prove the expression of ten anoikis-related genes in LUAD and normal tissues, and related cell and animal experiments to explore their molecular regulatory mechanisms. These aspects need to be further researched and explore.