Ensemble deep learning enhanced with self-attention for predicting immunotherapeutic responses to cancers
作者全名："Jin, Wenyi; Yang, Qian; Chi, Hao; Wei, Kongyuan; Zhang, Pengpeng; Zhao, Guodong; Chen, Shi; Xia, Zhijia; Li, Xiaosong"
作者地址："[Jin, Wenyi] Wuhan Univ, Renmin Hosp, Dept Orthoped, Wuhan, Peoples R China; [Yang, Qian; Chen, Shi; Li, Xiaosong] Chongqing Med Univ, Affiliated Hosp 1, Clin Mol Med Testing Ctr, Chongqing, Peoples R China; [Chi, Hao] Southwest Med Univ, Clin Med Collage, Luzhou, Peoples R China; [Wei, Kongyuan] Heidelberg Univ, Dept Gen Visceral & Transplantat Surg, Heidelberg, Germany; [Zhang, Pengpeng] Nanjing Med Univ, Affiliated Hosp 1, Dept Thorac Surg, Nanjing, Peoples R China; [Zhao, Guodong] Chinese Peoples Liberat Army PLA Gen Hosp, Med Ctr 1, Fac Hepatopancreatobiliary Surg, Beijing, Peoples R China; [Xia, Zhijia] Ludwig Maximilians Univ Munchen, Dept Gen Visceral & Transplant Surg, Munich, Germany"
通信作者："Li, XS (通讯作者)，Chongqing Med Univ, Affiliated Hosp 1, Clin Mol Med Testing Ctr, Chongqing, Peoples R China.; Xia, ZJ (通讯作者)，Ludwig Maximilians Univ Munchen, Dept Gen Visceral & Transplant Surg, Munich, Germany."
来源：FRONTIERS IN IMMUNOLOGY
关键词：deep learning; immunotherapy; cancer; PD1; PD-L1; ELISE
摘要："IntroductionDespite the many benefits immunotherapy has brought to patients with different cancers, its clinical applications and improvements are still hindered by drug resistance. Fostering a reliable approach to identifying sufferers who are sensitive to certain immunotherapeutic agents is of great clinical relevance. MethodsWe propose an ELISE (Ensemble Learning for Immunotherapeutic Response Evaluation) pipeline to generate a robust and highly accurate approach to predicting individual responses to immunotherapies. ELISE employed iterative univariable logistic regression to select genetic features of patients, using Monte Carlo Tree Search (MCTS) to tune hyperparameters. In each trial, ELISE selected multiple models for integration based on add or concatenate stacking strategies, including deep neural network, automatic feature interaction learning via self-attentive neural networks, deep factorization machine, compressed interaction network, and linear neural network, then adopted the best trial to generate a final approach. SHapley Additive exPlanations (SHAP) algorithm was applied to interpret ELISE, which was then validated in an independent test set. ResultRegarding prediction of responses to atezolizumab within esophageal adenocarcinoma (EAC) patients, ELISE demonstrated a superior accuracy (Area Under Curve [AUC] = 100.00%). AC005786.3 (Mean [|SHAP value|] = 0.0097) was distinguished as the most valuable contributor to ELISE output, followed by SNORD3D (0.0092), RN7SKP72 (0.0081), EREG (0.0069), IGHV4-80 (0.0063), and MIR4526 (0.0063). Mechanistically, immunoglobulin complex, immunoglobulin production, adaptive immune response, antigen binding and others, were downregulated in ELISE-neg EAC subtypes and resulted in unfavorable responses. More encouragingly, ELISE could be extended to accurately estimate the responsiveness of various immunotherapeutic agents against other cancers, including PD1/PD-L1 suppressor against metastatic urothelial cancer (AUC = 88.86%), and MAGE-A3 immunotherapy against metastatic melanoma (AUC = 100.00%). DiscussionThis study presented deep insights into integrating ensemble deep learning with self-attention as a mechanism for predicting immunotherapy responses to human cancers, highlighting ELISE as a potential tool to generate reliable approaches to individualized treatment."
基金机构：National Natural Science Foundation of China; Natural Science Foundation of Chongqing; Chongqing Science and Health Joint Medical High-end Talent Project; Science and Technology Research Program of Chongqing Municipal Education Commission; CQMU Program for Youth Innovation in Future Medicine; ; [cstc2020jcyj-msxmX0159]; [2022GDRC012]; [KJZD-K202100402]; [KJQN201900449]; [W0073]
基金资助正文："Funding This study was funded by the National Natural Science Foundation of China (grant number 81871653), the Natural Science Foundation of Chongqing (cstc2020jcyj-msxmX0159), Chongqing Science and Health Joint Medical High-end Talent Project (2022GDRC012), Science and Technology Research Program of Chongqing Municipal Education Commission (KJZD-K202100402, KJQN201900449) and CQMU Program for Youth Innovation in Future Medicine (W0073)."