AUTHOR=Zhang Yuwei , Yang Yichen , Ma Yue , Liu Ying , Ye Zhaoxiang TITLE=Development and validation of an interpretable radiomic signature for preoperative estimation of tumor mutational burden in lung adenocarcinoma JOURNAL=Frontiers in Genetics VOLUME=Volume 15 - 2024 YEAR=2024 URL=https://www.frontiersin.org/journals/genetics/articles/10.3389/fgene.2024.1367434 DOI=10.3389/fgene.2024.1367434 ISSN=1664-8021 ABSTRACT=Background: Tumor mutational burden (TMB) is a promising biomarker for immunotherapy. Challenge of spatial and temporal heterogeneity and high costs weaken its power in clinical routine. The aim of this study is to estimate tumor mutational burden preoperatively using a volumetric CTbased radiomic signature (rMB).Methods: 71 patients with resectable lung adenocarcinoma (LUAD) underwent whole exome sequencing (WXS) from 2011-2014 were enrolled from institutional biobank. 49 LUAD patients with WXS from the Cancer Genome Atlas Program (TCGA) served as external validation cohort. Computed Tomography (CT) volumes were resampled to 1mm-isotropic, semi-automatically segmented, and manually adjusted by two radiologists. 3108 radiomic features were extracted via PyRadiomics then harmonized across cohorts by ComBat. Features with inter-segmentation intraclass correlation coefficient (ICC)>0.8, low collinearity, and significant univariate power were passed to the least absolute shrinkage and selection operator (LASSO)-Logistic classifier to discriminate TMB-high/low at a threshold of 10 mut/Mb. Receiver operating characteristic (ROC) curve analysis, calibration curve determined its efficiency. Shapley Values (SHAP) attributed individual predictions and feature contributions. Clinical variables and circulating biomarkers were collected to find potential associations with TMB and rMB.Results: Top-3 frequently mutated genes significantly differed between Chinese and TCGA cohorts, with median TMB of 2.20 and 3.46mut/Mb, 15(21.12%) and 9(18.37%) cases of TMB-high respectively. After dimensionality reduction, rMB comprised 21 features, which reached an AUC of 0.895 (Sensitivity=0.867, Specificity=0.875, Accuracy=0.873) in discovery cohort, and 0.878 (Sensitivity =1.0, Specificity =0.825, Accuracy =0.857 in a consist cutoff) in validation cohort. rMB of TMB-high patients significantly higher than TMB-low in both cohorts (p<0.01). rMB was wellcalibrated in discovery cohort and validation cohort (p=0.27 and 0.74). Square-filtered gray level concurrence matrix (GLCM) correlation was of most importance in prediction. Proportion of circulating monocytes and monocyte-to-lymphocyte ratio were associated with TMB, whereas circulating neutrophils and lymphocyte percentage, original and derived neutrophil-to-lymphocyte ratio, as well as platelet-to-lymphocyte ratio were associated with rMB.rMB, an intratumor radiomic signature, could predict lung adenocarcinoma patients with higher TMB. Insights from Shapley values may enhance the persuasiveness of purposed signature for further clinical application. rMB would be a promising tool to triage patients who might benefit from a next-generation sequencing test.