本前瞻性研究旨在评估[18F]AlF-NOTA-成纤维细胞激活蛋白抑制剂（FAPI）-04正电子发射断层扫描-计算机断层扫描（PET/CT）在预测乳腺癌分子亚型中的价值。该研究连续招募疑似患者来自单一中心的乳腺癌患者，于 2023 年 7 月至 2024 年 5 月前瞻性入组并接受 [ 18 F]AlF-NOTA-FAPI-04 PET/CT。本研究比较了具有不同不良预后因素和分子亚型的乳腺癌示踪剂摄取的差异。使用受试者工作特征（ROC）曲线评估乳腺癌每种分子亚型的分类性能。对 53 名参与者（平均年龄 51±11 岁；52 名女性）进行了评估。具有不良预后因素的乳腺癌病变显示出较高的示踪剂摄取。五种不同的分子亚型表现出不同的吸收水平。 Luminal A 和 Luminal B（HER2 阴性）亚型的摄取量相对较低，而 Luminal B（HER2 阳性）、HER2 阳性和三阴性亚型的摄取量相对较高。 ROC 分析将最大标准化摄取值 (SUVmax) 确定为 Luminal A 亚型的显着分类器 (AUC = 0.912，P = 0.0005)，具有 100% 的敏感性和 83% 的特异性。对于预测 luminal B（HER2 阴性）亚型，SUVmax 的 AUC 为 0.770 (P = 0.0015)。 SUVmax 的 AUC 为 0.781（P = 0.003），用于鉴定三阴性亚型肿瘤，敏感性为 100%，特异性为 51%。最后，ROC曲线显示截止点15.40（AUC = 0.921，P < 0.0001）可以对Luminal A进行分类

This prospective study aims to evaluate the value of [18F]AlF-NOTA-fibroblast activation protein inhibitor (FAPI)-04 positron emission tomography-computed tomography (PET/CT) in predicting molecular subtypes of breast cancer.The study consecutively recruited patients suspected of having breast cancer from a single center who were prospectively enrolled from July 2023 to May 2024 and underwent [18F]AlF-NOTA-FAPI-04 PET/CT. This study compared the differences in tracer uptake among breast cancers with different adverse prognostic factors and molecular subtypes. The classification performance for each molecular subtype of breast cancer was assessed using a receiver operating characteristic (ROC) curve.Fifty-three participants (mean age, 51 ± 11 years; 52 females) were evaluated. Breast cancer lesions with adverse prognostic factors showed higher tracer uptake. The five different molecular subtypes exhibited varying levels of uptake. The luminal A and luminal B (HER2-negative) subtypes had relatively low uptake, while the luminal B (HER2-positive), HER2-positive, and triple-negative subtypes had relatively high uptake. ROC analysis identified the max standardized uptake value (SUVmax) as a significant classifier (AUC = 0.912, P = 0.0005) for the luminal A subtype, with 100% sensitivity and 83% specificity. For predicting the luminal B (HER2-negative) subtype, SUVmax had an AUC of 0.770 (P = 0.0015). SUVmax, with an AUC of 0.781 (P = 0.003), was used to identify the triple-negative subtype tumors, resulting in a sensitivity of 100% and specificity of 51%. Lastly, the ROC curve showed the cut-off 15.40 (AUC = 0.921, P < 0.0001) could classify luminal A & luminal B (HER2-negative), and luminal B (HER2-positive) & HER2-positive & triple-negative, yielding a sensitivity of 94% and specificity of 79%.The uptake of [18F]AlF-NOTA-FAPI-04 is significantly correlated with the molecular subtypes of breast cancer, and [18F]AlF-NOTA-FAPI-04 PET/CT is a potential tool for noninvasive identification of luminal A subtypes and guidance of FAP-targeted therapies.© 2024. The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.