AUTHOR=Chang Zhenglin , Li Rongqi , Zhang Jinhu , An Lingyue , Zhou Gaoxiang , Lei Min , Deng Jiwang , Yang Riwei , Song Zhenfeng , Zhong Wen , Qi Defeng , Duan Xiaolu , Li Shujue , Sun Baoqing , Wu Wenqi 

TITLE=Distinct immune and inflammatory response patterns contribute to the identification of poor prognosis and advanced clinical characters in bladder cancer patients

JOURNAL=Frontiers in Immunology

VOLUME=Volume 13 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/immunology/articles/10.3389/fimmu.2022.1008865

DOI=10.3389/fimmu.2022.1008865

ISSN=1664-3224

ABSTRACT=Due to the molecular heterogeneity, the majority of bladder cancer (BLCA) patients show no pathological responses to immunotherapy and chemotherapy yet suffer from their toxicity. In this study, we identified and validated three distinct and stable molecular clusters of BLCA in cross-platform databases based on personalized immune and inflammatory characteristics. H&E-stained histopathology images confirmed the distinct infiltration of immune and inflammatory cells among clusters. Cluster-A was characterized by favorable prognosis, low immune and inflammatory infiltration but showed the highest abundance of prognosis-related favorable immune cell and inflammatory activity. Cluster-B featured the worst prognosis, high immune infiltration, but numerous unfavorable immune cells exist. Cluster-C was characterized by a favorable prognosis and the highest immune and inflammatory infiltration. Based on machine learning, a highly precise predictive model (IIRS), including FN1, IL10, MYC, CD247, and TLR2, was developed and validated to identify the high IIRS-score group that had a poor prognosis and advanced clinical characters. Compared to other published models, IIRS showed the highest AUC in 5 years OS and a favorable predictive value in predicting 1- and 3- year OS. Moreover, IIRS showed an excellent performance in predicting immunotherapy and chemotherapy's response. According to immunohistochemistry and qRT-PCR, IIRS genes were differentially expressed between tumor tissues with corresponding normal or adjacent tissues. Finally, immunohistochemical and H&E-stained analysis were performed on bladder tissues of 13 BLCA patients to further demonstrate that IIRS score is a valid substitute for IIR patterns and can contribute to identify patients who had poor clinical and histopathology characters. In conclusion, we established a novel IIRS depicting IIR pattern which could independently predict OS and acts as a highly precise predictive biomarker for advanced clinical characters and the responses of immunotherapy and chemotherapy.