Novel Computed-Tomography-Based Transformer Models for the Noninvasive Prediction of PD-1 in Pre-Operative Settings

SIMPLE SUMMARY: Obtaining the PD-1/PD-L1 status is conducive to observing the patient’s response rate and constructing individualized immunotherapy strategies. However, biopsies are invasive in assessing the PD-1 status, entail sampling bias due to tumor heterogeneity, are expensive and a slow process, and introduce increased risks of complications. Our research explored a new model based on transformer and CT images to predict PD-1 use. We confirmed that our model can accurately predict the expression of PD-1 via the study of a cohort of 93 patients collected in West China Hospital. The promising diagnostic performance shows that our model is an effective and noninvasive classification method, providing a practical tool for predicting various receptors. ABSTRACT: The expression status of programmed cell death protein 1 (PD-1) in patients with hepatocellular carcinoma (HCC) is associated with the checkpoint blockade treatment responses of PD-1/PD-L1. Thus, accurately and preoperatively identifying the status of PD-1 has great clinical implications for constructing personalized treatment strategies. To investigate the preoperative predictive value of the transformer-based model for identifying the status of PD-1 expression, 93 HCC patients with 75 training cohorts (2859 images) and 18 testing cohorts (670 images) were included. We propose a transformer-based network architecture, ResTransNet, that efficiently employs convolutional neural networks (CNNs) and self-attention mechanisms to automatically acquire a persuasive feature to obtain a prediction score using a nonlinear classifier. The area under the curve, receiver operating characteristic curve, and decision curves were applied to evaluate the prediction model’s performance. Then, Kaplan–Meier survival analyses were applied to evaluate the overall survival (OS) and recurrence-free survival (RFS) in PD-1-positive and PD-1-negative patients. The proposed transformer-based model obtained an accuracy of 88.2% with a sensitivity of 88.5%, a specificity of 88.9%, and an area under the curve of 91.1% in the testing cohort.