医学
乳腺癌
接收机工作特性
乳房磁振造影
肿瘤科
癌症
阶段(地层学)
列线图
逻辑回归
内科学
放射科
核医学
乳腺摄影术
古生物学
生物
作者
Zengjie Wu,Qing Lin,Haibo Wang,Jingjing Chen,Guanqun Wang,Guangming Fu,Lili Li,Tiantian Bian
摘要
Background Programmed 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. Purpose To determine the ability of radiomics signatures based on preoperative dynamic contrast‐enhanced (DCE) MRI to evaluate PD‐L1 expression in breast cancer. Study Type Retrospective. Population 196 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/Sequence 3.0T; volume imaging for breast assessment DCE sequence. Assessment Radiomics 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 Tests Univariable 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. Results In 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 Conclusion Intratumoral 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 Evidence 4 Technical Efficacy Stage 2
科研通智能强力驱动
Strongly Powered by AbleSci AI