Harada F, Fukuda T, Uchiyama Y. [Radioproteomics for Discriminating the Activity and Inactivity of Immune Checkpoint Molecules in Breast Cancer].
Nihon Hoshasen Gijutsu Gakkai Zasshi 2023;
79:1136-1143. [PMID:
37587046 DOI:
10.6009/jjrt.2023-1358]
[Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 08/18/2023]
Abstract
PURPOSE
Radioproteomics studies investigating the relationship between lesion phenotype and proteins have been progressed. The purpose of this study was to develop a radioproteomics method for discriminating between active and inactive immune checkpoint molecules based on lesion phenotype.
METHODS
From the public database TCGA-BRCA, mRNA and fat suppression contrast-enhanced T1-weighted images of 49 patients with breast cancer were selected for the experiment. Using mRNA, we defined cases with active (10 cases) and inactive (39 cases) immune checkpoint molecules. To discriminate these cases using lesion phenotype, 275 radiomics features were measured from the tumor area. After selecting 3 radiomics features by using Lasso, logistic regression was employed to discriminate between active and inactive cases of immune checkpoint molecules.
RESULTS
Evaluation of ROC analysis showed that the AUC was 0.81.
CONCLUSION
Patients whose immune cell function is being braked by immune checkpoint molecules are likely to respond to immune checkpoint inhibitors when their activity is inhibited. Therefore, our results may be applied to predict the effects of immune checkpoint inhibitors in breast cancer treatment.
Collapse