Affiliation:
1. Department of Radiology The Affiliated Hospital of Qingdao University Qingdao Shandong China
2. Breast Disease Center The Affiliated Hospital of Qingdao University Qingdao Shandong China
3. Department of Pathology The Affiliated Hospital of Qingdao University Qingdao Shandong China
Abstract
BackgroundProgrammed cell death ligand‐1 (PD‐L1) is a promising target for immune checkpoint blockade therapy in breast cancer. However, the preoperative evaluation of PD‐L1 expression in breast cancer is rarely explored.PurposeTo determine the ability of radiomics signatures based on preoperative dynamic contrast‐enhanced (DCE) MRI to evaluate PD‐L1 expression in breast cancer.Study TypeRetrospective.Population196 primary breast cancer patients with preoperative MRI and postoperative pathological evaluation of PD‐L1 expression, divided into training (n = 137, 28 PD‐L1‐positive) and test cohorts (n = 59, 12 PD‐L1‐positive).Field Strength/Sequence3.0T; volume imaging for breast assessment DCE sequence.AssessmentRadiomics features were extracted from the first phase of DCE‐MRI by using the minimum redundancy maximum relevance method and least absolute shrinkage and selection operator algorithm. Three radiomics signatures were constructed based on the intratumoral, peritumoral, and combined intra‐ and peritumoral regions. The performance of the signatures was assessed using area under the receiver operating characteristic (ROC) curve (AUC), sensitivity, specificity, and accuracy.Statistical TestsUnivariable and multivariable logistic regression analysis, t‐tests, chi‐square tests, Fisher exact test or Yates correction, ROC analysis, and one‐way analysis of variance. P < 0.05 was considered significant.ResultsIn the test cohort, the combined radiomics signature (AUC, 0.853) exhibited superior performance compared to the intratumoral (AUC, 0.816; P = 0.528) and peritumoral radiomics signatures (AUC, 0.846; P = 0.905) in PD‐L1 status evaluation, although the differences did not reach statistical significance.Data ConclusionIntratumoral and peritumoral radiomics signatures based on preoperative breast MRI showed some potential accuracy for the non‐invasive evaluation of PD‐L1 status in breast cancer.Level of Evidence4Technical EfficacyStage 2
Subject
Radiology, Nuclear Medicine and imaging
Cited by
1 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献