A Novel Defined PANoptosis-related Patterns For Predicting the Prognosis and Immune Characteristics in Kidney Clear Cell Carcinoma：Bioinformatics and Machine-learning Approach

Abstract

Background: Kidney clear cell carcinoma (KIRC) exhibits resistance to conventional therapies, and its heterogeneity limits the efficacy of immunotherapy. PANoptosis, as a key in personalized immunotherapeutic approaches, remains unexplored in KIRC. Our aim is to develop a prognostic model utilizing PANoptosis to predict immune response and outcomes in KIRC patients. Methods: Initially, PANoptosis-related genes were identified, and RNA sequencing data along with clinical pathological features from the TCGA-KIRC cohort were collected. Subsequently, consensus clustering analysis was employed to recognize distinct PANoptosis patterns. Then, we screened for prognostic-related genes using univariate Cox regression. After feature selection with LASSO regression, we constructed the PANscore risk assessment system using multivariate Cox regression. The model was validated on the GSE29609 dataset to evaluate its predictive accuracy and generalizability. Additionally, the relationship between PANscore and immune cell infiltration was analyzed, and a pan-cancer analysis was performed. Finally, the predictive capabilities of traditional machine learning algorithms and convolutional neural network algorithms were compared to determine the optimal prediction model. Results: Two distinct PANoptosis patterns were discovered, with PANcluster 2 exhibiting superior potential for immunotherapeutic response. A novel prognostic metric, designated as PANscore, was formulated with the aim of forecasting the clinical trajectories and therapeutic reactions of patients, where patients with high PANscores may benefit more from immunotherapy. High PANscores were negatively correlated with most immune cell populations across various cancers. Patient outcome prediction based on PANoptosis genes demonstrated superior predictive ability of CNN over traditional machine-learning algorithms. Conclusions: PANcluster 2 and PANscore is crucial for assessing the prognosis of patients with KIRC, particularly in identifying high PANoptosis expression who may derive greater benefits from immune checkpoint inhibitor therapy. Furthermore, machine-learning approach has demonstrated its immense potential as an analytical and predictive tool in our research.