Preoperative Radiomic Approach to Evaluate Tumor Infiltrating CD8+ T Cells in HCC Patients Using Contrast-Enhanced CT

Background: Despite great achievement in HCC immunotherapy, only a small portion of HCC patients could respond to the immune treatment. To help identify potential HCC candidate for immunotherapies, we aim to develop and validate a radiomics-based biomarker (Rad score) to predict the infiltration of tumor infiltrating CD8+ T cells in HCC patients and evaluate the correlation of Rad score with tumor immune characteristics. Methods: A total of 142 HCC patients (training set, n = 100; validation set, n = 42) were subjected to radiomic feature extraction (385 features). To develop Rad score of CD8+ T cells, image features and immunochemistry data of patients in the training set were subjected to elastic-net regularised regression analysis to predict the level of CD8+ T cell infiltration. The correlations of Rad score with survival outcomes, tumor infiltrating lymphocytes (TILs), immune phenotypes and PD-L1 expression were also evaluated in all the patients included. Results: We developed a Rad score for CD8+ T cell infiltration that contained seven variables, which was validated in the validation set (area under the curve [AUC]: training set, 0.751 [95%CI 0·656-0·846; validation set, 0.705 [95%CI 0·547-0·863]), and the decision curve indicated the clinical usefulness of Rad score. Higher Rad sore correlated with superior overall and disease-free survival outcomes (P=0.016 and 0.0073, respectively). By using the pathological slides, we found that Rad score positively correlated with the percentage of TILs (Spearman Rho=0.51, P<0.0001). Moreover, Rad score was also able to discriminate inflamed tumors from immune-desert and immune-excluded tumors (Kruskal-Wallis, P<0.0001), and higher Rad scores could be found in patients with positive PD-L1 expression in tumor or immune cells in HCC tissues. Interpretation: The newly developed Rad score was a powerful predictor of CD8+ T cell infiltration, which could be useful in identifying potential HCC patients who can benefit from immunotherapies when validated in large-scale prospective cohorts. Funding: This work was supported by grants from the Natural Science Foundation of China (81872004, 81800564, 81770615, 81700555, 81672882 and 81502441), the Science and Technology Support Program of Sichuan Province (2017SZ0003, 2018SZ0115) and the Science and Technology Program of Tibet Autonomous Region (XZ201801-GB-02). Declaration of Interest: The authors declared no competing interest in this work. Ethical Approval: This retrospective study was approved by the Ethics Committee of Sichuan University, with written informed consent obtained from each study participant according to guidelines from the committee.