Feasibility of early-phase [68Ga]Ga-FAPI-46 PET for preoperative staging in lung cancer

Background Fibroblast activation protein (FAP)-targeting PET has gained attention in oncology for its high diagnostic performance and procedural advantages, including the possibility of early image acquisition. This study investigated the feasibility of early-phase [ ⁶⁸ Ga]Ga-FAPI-46 PET imaging in preoperative staging of non-small cell lung cancer (NSCLC). Results Twenty-six patients with NSCLC scheduled for curative resection and meeting high-risk criteria for mediastinal lymph node metastasis were enrolled. [⁶⁸Ga]Ga-FAPI-46 PET/CT images were acquired at 10 and 60 minutes post-injection, within two days before surgery. Primary tumors and mediastinal lymph nodes were evaluated by visual interpretation and quantitative analyses. Primary tumor detection rates and diagnostic performance for mediastinal nodes, assessed on per-patient and per-station basis, were compared between the two timepoints. Postoperative pathologic results were used as the reference standard. In visual assessment, all primary tumors were detected on 60-minute images, whereas one small tumor showed only faint uptake on 10-minute images (detection rate, 96.2%). For detection of N2 nodal metastasis, both timepoints demonstrated identical high diagnostic performance (per-patient sensitivity/specificity, 0.90/1.00; per-station sensitivity/specificity, 0.79/0.99). SUV _max of primary and nodal lesions did not consistently change between the two timepoints (primary, 8.50 ± 4.15 → 8.68 ± 4.46; lymph node, 5.08 ± 2.99 → 4.84 ± 2.74). However, tumor-to-background ratio increased on 60-minute images (primary, 6.26 ± 3.42 → 8.59 ± 4.95, P  < 0.0001; lymph node, 3.06 ± 1.74 → 4.70 ± 2.71, P  = 0.0001) due to blood pool washout, suggesting potential advantage of delayed imaging for mediastinal lesion discrimination. Conclusions [ ⁶⁸ Ga]Ga-FAPI-46 10-minute imaging exhibited acceptable diagnostic performance compared to 60-minute imaging. However, caution may be warranted when interpreting small lesions and mediastinal nodes, suggesting the need for further validation with larger cohorts.