Fasmer Kristine Eldevik, Gulati Ankush, Lindås Sunniva, Krakstad Camilla, Haldorsen Ingfrid Salvesen
Mohn Medical Imaging and Visualization Centre, Department of Radiology, Haukeland University Hospital, Bergen, Norway.
Department of Clinical Medicine, University of Bergen, Bergen, Norway.
Eur J Nucl Med Mol Imaging. 2025 Jun 11. doi: 10.1007/s00259-025-07335-7.
To develop a [F]fluorodeoxyglucose ([F]FDG) positron emission tomography (PET) primary tumor radiomic model for predicting disease-specific survival (DSS), and compare it with conventional PET markers in a large endometrial cancer cohort.
Radiomic features were extracted from preoperative [F]FDG PET scans of 489 endometrial cancer patients using a standardized uptake value (SUV) threshold > 2.5 to define primary metabolic tumor volumes (MTVs). A second reader extracted features in 154/489 patients, in which intraclass correlation coefficients (ICCs) were calculated. Radiomic features with ICCs > 0.75 were retained and ComBat harmonization was applied to reduce scanner/protocol effects on the extracted features. Patients were divided into training (n = 343) and test (n = 146) sets. A radiomic DSS score (R) was developed in the training set using least absolute shrinkage and selection operator (LASSO) Cox regression. A combined model (C), incorporating R, PET positive lymph nodes (LN) and preoperative histology risk was constructed using multivariable Cox hazard analyses. Prediction performances were assessed by comparing areas under time-dependent receiver operating characteristic curves (tdROCs AUCs) for R, C, and conventional PET markers: SUV, SUV, MTV, tumor lesion glycolysis (TLG) and LN.
In the test set, AUCs for 2- and 5-year DSS were higher for R (0.855, 0.720) compared to SUV (0.548, 0.572) and SUV (0.549, 0.554) (p ≤ 0.04 for all), while similar to MTV (0.863, 0.696), TLG (0.814, 0.672) and LN (0.802, 0.626) (p ≥ 0.12 for all). C predicted 2-year DSS with AUC of 0.909 in the test set, outperforming all conventional imaging markers (p ≤ 0.04 for all) except MTV (p = 0.29). For 5-year DSS, C (AUC: 0.817) outperformed all conventional imaging markers, including MTV (AUC ≤ 0.696, p ≤ 0.05, for all).
R predicts short-term survival with high accuracy, outperforming tumor SUV, but not MTV, TLG and LN. The combined C model yields high accuracy for predicting both short- and long-term survival, outperforming all conventional PET imaging markers.
建立一种用于预测疾病特异性生存（DSS）的[F]氟脱氧葡萄糖（[F]FDG）正电子发射断层扫描（PET）原发性肿瘤放射组学模型，并在一个大型子宫内膜癌队列中与传统PET标志物进行比较。
使用标准化摄取值（SUV）阈值>2.5从489例子宫内膜癌患者的术前[F]FDG PET扫描中提取放射组学特征，以定义原发性代谢肿瘤体积（MTV）。另一位阅片者在154/489例患者中提取特征，并计算组内相关系数（ICC）。保留ICC>0.75的放射组学特征，并应用ComBat归一化以减少扫描仪/协议对提取特征的影响。患者被分为训练组（n = 343）和测试组（n = 146）。在训练组中使用最小绝对收缩和选择算子（LASSO）Cox回归建立放射组学DSS评分（R）。使用多变量Cox风险分析构建一个综合模型（C），该模型纳入了R、PET阳性淋巴结（LN）和术前组织学风险。通过比较R、C和传统PET标志物（SUV、SUV、MTV、肿瘤病变糖酵解（TLG）和LN）的时间依赖性受试者操作特征曲线下面积（tdROCs AUCs）来评估预测性能。
在测试组中，R的2年和5年DSS的AUC（0.855，0.720）高于SUV（0.548，0.572）和SUV（0.549，0.554）（所有p≤0.04），而与MTV（0.863，0.696）、TLG（0.814，0.672）和LN（0.802，0.626）相似（所有p≥0.12）。C在测试组中预测2年DSS的AUC为0.909，优于所有传统成像标志物（所有p≤0.04），但MTV除外（p = 0.
