[¹⁸F] AlF‑NOTA‑FAPI‑04 PET/CT as a promising tool for imaging fibroblast activation protein in gastrointestinal system cancers: a prospective investigation of comparative analysis with ¹⁸F-FDG

This prospective study was reviewed, approved, and overseen by the institutional review board of Harbin Medical University Cancer Hospital (approval 2021-198-JS) and conducted under the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. The study was registered at ClinicalTrials.gov (ChiCTR2200058108). All subjects signed informed consent forms in line with the local ethics committee's regulations for prospective research. ¹⁸F-FAPI-04 PET/CT was performed for comparative purposes within one week after ¹⁸F-FDG PET/CT. Participants were consecutively recruited for enrollment in this study from October 2021 to December 2022.

The specific inclusion criteria were as follows: 1) adult patients (≥ 18 years old); 2) subjects with newly diagnosed or previously treated gastrointestinal system malignancies (The interval between the completion of anti-tumor treatment and the PET-CT scan was more than 6 months); 3) patients who agreed to undergo paired ¹⁸F-FDG and ¹⁸F-FAPI-04 PET/CT for tumor staging or restaging; 4) according to RECIST1.1, there was at least one measurable target lesion; 5) understand and sign informed consent voluntarily with good compliance. Exclusion criteria included 1) the function of liver and kidney was seriously abnormal; 2) preparation for pregnant, pregnant and lactating women; 3) patients whose treatment had already started between the acquisition of the ¹⁸F-FDG PET/CT scan and ¹⁸F-FAPI-04 PET/CT scan; 4) inability to lie flat for half an hour; 5) suffering from claustrophobia or other mental disorders; 6) other researchers considered it unsuitable to participate in the trial; 7) substandard image quality (such as motion artifacts).

All patients were classified into either an initial staging group or a restaging group. The former refers to those patients who had not received any treatment before examinations. The latter was defined as patients who underwent examinations during treatment (therapeutic effect evaluation) or at least 2 months after completion of treatment (Monitoring tumor recurrence or metastasis). In this study, histopathologic examination of a biopsy or resected surgical specimens served as the gold standard for the final diagnosis. Clinical follow-up information, including results of medical imaging, physical examination, and laboratory tests, was used as the final reference standard when the pathological diagnosis is unavailable. All patients have to receive at least a three-month follow-up period.

¹⁸F-FDG was automatically manufactured at the PET/CT department of Harbin Medical University Cancer Hospital by the standard preparation methods applying the coincidence ¹⁸F-FDG synthesis module (TracerLab Fx FDG; GE Healthcare, Milwaukee, Wis). Radiolabeling of ¹⁸F-FAPI-04 was performed using a previously described protocol. The FAPI precursor (1,4,7-triazacyclononane-1,4,7-triacetic acid) [NOTA] FAPI-04). Quality control of the radiosynthesis was performed by ultraviolet and radio high-performance liquid chromatography (HPLC). Radiochemical purity exceeded 95% for both ¹⁸F-FAPI-04 and ¹⁸F-FDG, and the final product was diluted and sterilized. The sterility tests were conducted in the radiochemistry facility of Harbin Medical University Cancer Hospital. Finally, ¹⁸F-FDG and ¹⁸F-FAPI-04 have to conform to all set criteria prior to their transformation into the clinic for human administration. The synthesis process of ¹⁸F‑FAPI‑04 is described in supplementary data 1, and the corresponding chemical structural formula is displayed in supplementary data 2.

¹⁸F-FAPI-04 PET/CT examination should be performed within one week after ¹⁸F-FDG PET/CT scanning. The intravenous dose of the two agents is calculated according to the patient's weight (3.7 MBq [0.1 mCi]/kg for FDG; 1.8–2.2 MBq [0.05–0.06 mCi]/kg for FAPI). Before the ¹⁸F-FDG PET/CT examination, each patient was required to fast for 4-6 hours, to achieve blood glucose levels < 160 mg/dl. All patients do not need special preparation (e.g., fasting and normal blood glucose level) before the ¹⁸F-FAPI-04 PET/CT scans. Static PET/CT imaging was performed using a hybrid PET/CT system (Discovery MI, GE Healthcare, Milwaukee, WI, USA) 60 min after injection. Firstly, low-dose CT scans (free-breathing state and unenhanced images) were performed before whole-body PET/CT examination. Detailed scanning parameters were as follows: tube voltage 140 kV, tube current 150 mA, slice thickness 3.75 mm, matrix size 512 × 512, and field of view 450 mm. PET examination was immediately performed after the CT examination in three-dimensional acquisition mode with 6–8 bed positions and 2.0–2.5 min/position. Image reconstructions were performed based on the 3D ordered subset expectation-maximization algorithm (2 iterations and 17 subsets), and all PET images were corrected for attenuation correction and reconstructed into a 128 × 128 matrix.

Images were analyzed using a dedicated post-processing software (PET VCAR; GE Healthcare). Two experienced radiologists who have more than 10-year practicing experience, blinded to the surgical and pathological results, independently evaluated the images. Any discrepancy was settled through consensus by discussion. To prevent any bias, ¹⁸F-FDG PET/CT images were evaluated by group 1 (radiologist S.X. and radiologist C.G.), while ¹⁸F-FAPI-04 PET/CT images were assessed in group 2 (radiologist K.W. and radiologist S.C.). All of them were blinded to patient history as well as results of conventional imaging and pathologic results. For visual analysis, based on the understanding of the physiological distribution of the agents in the whole body, compared with the contralateral normal tissues and surrounding soft tissues, if radiotracer uptake is increased or significantly concentrated, which exceeded the average value of adjacent background tissues, then these lesions were considered as positive. For semiquantitative analysis, regions of interest were manually drawn by another radiologist (L.C., with more than 10 years of experience in nuclear oncology) on trans-axial images around the tumor lesions to measure the metabolic parameters. The maximum standard uptake values (SUVmax), the median, and range of standard uptake values were recorded, which were used to quantify the radiotracer uptake in primary tumors, lymph nodes, and metastatic lesions. The tumor-to-background (T/B) ratio was recorded by dividing the tumor SUVmax with the mean SUV of the contralateral normal tissue.

For tumor initial staging, the eighth edition of the AJCC TNM staging system was applied [9]. In brief, the degree of primary tumor invasion defines its T-stage, the number and distribution of metastatic lymph nodes define its N-stage, and the situation of distant metastasis determines its M-stage. For tumor restaging, suspected local recurrence was determined as the occurrence of new lesions at the primary tumor site, and metastasis was defined as the presence of metastatic lymph nodes or/and distant metastases. Lymph nodes mainly involved four regions: neck and supraclavicular, mediastinum, abdomen (including paraaortic, porta hepatic, retroperitoneal, celiac), and pelvis. Any peritoneal or omental or mesenteric metastasis is considered peritoneal carcinoma. Each metastasis of the brain, bone, liver, lung, and peritoneal carcinoma was uniformly defined as distant metastases.

For per-patient analysis, A final diagnosis was made by a dedicated multidisciplinary team (MDT) based on a comprehensive evaluation of clinical symptoms and signs, laboratory tests, radiological findings, pathological results, and clinical follow-up outcome. The MDT is composed of the previously mentioned nuclear medicine experts, two tumor surgery experts, one medical oncologist, and one radiation oncologist, who finally reached a consensus on the final diagnosis as the reference standard. Based on the difference between the two imaging methods and the reference standard in the same patient, final results were classified as the same (correct), overestimated, or underestimated. According to the above findings, we recorded the changes in clinical staging, and subsequent changes in oncological management were assessed by two nuclear medicine experts and two treating physicians. The referring treating physicians have further been consulted on what the therapeutic regimen would be prior to and after ¹⁸F-FAPI-04 PET/CT. The management reference standard was the consensus of the above-mentioned MDT team in line with the final diagnosis and the latest NCCN guidelines. For patients in the initial staging group, the therapeutic regimen included no treatment needed/follow-up, surgical resection, perioperative chemotherapy/radio-chemotherapy plus surgical resection, and non-surgical candidate. For those patients in the re-staging group, the secondary treatment plan was compared with the previous ones and then classified as maintenance or modification treatment.

Between October 2021 and December 2022, sixty-four patients with gastrointestinal system cancers were studied, who underwent both ¹⁸F-FDG and ¹⁸F-FAPI-04 PET/CT scans. Among the finally included 60 patients, 34 patients were female and 26 were males, and the medial age of the patients was 61 years (range, 38–68 years). The characteristics of the patients are summarized in Table 1. The median between ¹⁸F-FDG and ¹⁸F-FAPI-04 PET/CT was 2 days (range 1–5) days. The mean clinical follow-up time was 5.2 ± 1.8 month.

Among these newly diagnosed patients, fourteen patients had gastric cancer, including 12 patients with adenocarcinoma and 2 patients with signet ring cell carcinoma, ten patients had liver cancer, including 8 patients with hepatocellular carcinoma and 2 patients with intrahepatic cholangiocarcinoma, twelve patients had pancreatic cancer, including 3 patients with well-differentiated adenocarcinoma, 5 patients with moderately-differentiated adenocarcinoma and 4 patients with poorly-differentiated adenocarcinoma, eight patients had colorectal cancer, including 6 patients with moderately-differentiated adenocarcinoma and 2 patients with poorly-differentiated adenocarcinoma. All individuals who had previous treatment for gastrointestinal system cancers, in which four instances were gastric cancer (3 patients with adenocarcinoma and 1 patient with signet ring cell carcinoma), two instances were liver cancer (1 patients with hepatocellular carcinoma and 1 patient with intrahepatic cholangiocarcinoma), four instances were pancreatic ductal adenocarcinoma and six instances were colorectal adenocarcinoma.

No drug-related adverse reactions occurred before or after ¹⁸F-FAPI-04 PET/CT scans. PET imaging was tolerated well by all patients and any abnormal symptoms were not observed during injection and the 2 hours of observation period.

A total of 44 primary lesions were examined in patients with newly diagnosed gastrointestinal system cancers. In the depiction of primary tumors, the detection rate was 72.7% (32 of 44) for ¹⁸F-FDG PET/CT and 97.7% (43 of 44) for ¹⁸F-FAPI-04 PET/CT. The false-negative tumors from ¹⁸F-FDG PET/CT were pancreatic cancer (n = 3), gastric cancer (n = 4), and liver cancer (n = 5). ¹⁸F-FAPI-04 PET/CT demonstrated a higher detection rate for primary tumors in the initial staging group (97.7% [43 of 44] vs. 72.7% [32 of 44], P = 0.032). Only one primary lesion from the pancreatic was not detected on ¹⁸F-FAPI-04 PET/CT images, which was covered by diffuse and intense tracer uptake in the whole pancreas. On the ¹⁸F-FAPI-04 PET/CT images, most primary tumors were delineated with clear tumor contour and demonstrated a higher TBR (7.2 vs. 2.4; P < 0.001) than ¹⁸F-FDG, particularly in patients with gastric cancer and liver cancer. For semiquantitative analysis (Table 2), a higher uptake of ¹⁸F-FAPI-04 in pancreatic cancer compared with ¹⁸F-FDG (median SUV max, 10.4 vs. 5.1, respectively; P < 0.001), in gastric cancer (median SUV max, 9.7 vs. 4.6, respectively; P < 0.001), in liver cancer (median SUV max, 11.2 vs. 4.2, respectively; P < 0.001). Although all primary lesions from colorectal tumors were visualized on ¹⁸F-FDG PET/CT, these primary tumors demonstrated a higher uptake of ¹⁸F-FAPI-04 than of ¹⁸F-FDG (median SUV max, 9.1 vs. 5.7, respectively; P < 0.001).

The number of positive lymph nodes detected with ¹⁸F-FAPI-04 PET/CT in the abdomen (154 vs. 69) and pelvis (38 vs. 20) regions was more than that detected with ¹⁸F-FDG PET/CT, and the uptake of ¹⁸F-FAPI-04 was higher than that of ¹⁸F-FDG (median SUV max, 6.6 vs. 3.7, P < 0.001; 6.0 vs. 3.3, P < 0.001, respectively,). The positive lymph nodes in the region of the neck and supraclavicular demonstrated no difference in the number detected on ¹⁸F-FAPI-04 and ¹⁸F-FDG PET/CT (28 vs. 23), whereas the uptake of ¹⁸F-FAPI-04 was higher than that of ¹⁸F-FDG (median SUV max, 5.2 vs. 2.9, respectively; P < 0.001). Lymph nodes in the region of the mediastinum demonstrated no difference in uptake of radiotracers between ¹⁸F-FAPI-04 and ¹⁸F-FDG (median SUV max, 5.4 vs. 4.9, respectively; P= 0.382), whereas ¹⁸F-FAPI-04 PET/CT surpassed ¹⁸F-FDG PET/CT in the detection of mediastinal lymph nodes (19 vs. 9). A sum of 224 suspicious lymph nodes was identified in 36 patients and validated via lymph node dissection (n = 182), biopsy (n = 5), or radiographic follow-up (n = 37). Of these, metastasis was validated in 74 lymph nodes in 33 patients. ¹⁸F-FAPI-04 PET/CT depicted involved lymph nodes with 68 true-positive, 12 false-positive, 6 false-negative, and 138 true-negative. By contrast, ¹⁸F-FDG PET/CT depicted involved lymph nodes with 59 true-positive, 12 false-positive, 15 false-negative, and 122 true-negative. In per-lymph node analysis, the sensitivity, specificity, and accuracy in the diagnosis of metastatic lymph nodes were 91.89% (68 of 74), 92.00% (138 of 150), and 91.96% (206 of 224), respectively, for ¹⁸F-FAPI-04 PET/CT and 79.72% (59 of 74), 81.33% (122 of 150), and 80.80% (181 of 224) for ¹⁸F-FDG PET/CT. Both the sensitivity and the specificity of ¹⁸F-FAPI-04 PET/CT outperformed that of ¹⁸F-FDG PET/CT (91.89% vs. 79.72%, 92.00% vs. 81.33%, respectively; both P values < 0.001). The diagnostic performance in lymph node metastasis based on ¹⁸F-FAPI-04 and ¹⁸F-FDG PET/CT is displayed in Table 3.

Each lesion of the brain, liver, bone, lung, and peritoneal carcinoma was recorded separately. Compared with ¹⁸F-FDG PET/CT, ¹⁸F-FAPI-04 PET/CT depicted more suspected metastases in brain (7 vs. 3), liver (39 vs. 20), bone (79 vs. 51), lung (11 vs. 4) and, peritoneal carcinoma (48 vs. 22). For semiquantitative analysis, the median SUV max values derived from ¹⁸F-FAPI-04 were higher than that from ¹⁸F-FDG in most of the bone (5.2 vs. 3.1, respectively; P < 0.001), liver metastases (8.7 vs. 4.4, respectively; P < 0.001), lung metastases (4.5 vs. 2.7, respectively; P < 0.001) and peritoneal carcinoma (8.1 vs. 3.6, respectively; P < 0.001). It is also worth noting that lesions of brain metastases showed no significant difference in uptake of radiotracers between ¹⁸F-FAPI-04 and ¹⁸F-FDG (median SUV max, 3.8 vs. 3.2, respectively; P= 0.247), whereas the TBRs on ¹⁸F-FAPI-04 PET/CT surpassed that on ¹⁸F-FDG PET/CT (median, 9.6 vs. 2.5).

Regarding diagnostic performance for distant metastases, we evaluated 162 suspicious lesions in 40 patients. Pathological findings via surgery (n = 59), biopsy (n = 45), or radiographic follow-up (n = 58) were used to evaluate suspicious lesions. Of these, 137 lesions were confirmed as positive metastases in 38 patients. ¹⁸F-FAPI-04 PET/CT depicted involved distant metastatic lesions with 125 true-positive, 4 false-positive, 12 false-negative, and 21 true-negative. By contrast, ¹⁸F-FDG PET/CT depicted involved distant metastatic lesions with 102 true-positive, 8 false-positive, 35 false-negative, and 17 true-negative. However, it is also worth noting that biopsy validation of all suspicious distant metastatic lesions in this study was only used to verify PET/CT findings. Therefore, the true-negative and false-positive status of these patients cannot be accurately determined. In per-distant metastases analysis, the sensitivity, specificity, and accuracy in the diagnosis of distant metastatic lesions were 91.24% (125 of 137), 84.00% (21 of 25), and 73.46% (119 of 162), respectively, for ¹⁸F-FAPI-04 PET/CT and 72.34% (102 of 141), 68.00% (17 of 25), and 73.46% (119 of 162) for ¹⁸F-FDG PET/CT. Both the sensitivity and the specificity of ¹⁸F-FAPI-04 PET/CT were superior to that of ¹⁸F-FDG PET/CT (91.24% vs. 72.34%, 84.00% vs. 68.00%, respectively; both P values < 0.001). The summary of diagnostic performance in distant metastatic lesions based on ¹⁸F-FAPI-04 and ¹⁸F-FDG PET/CT is displayed in Table 3.

On patient-based analysis, differential diagnostic accuracies (¹⁸F-FAPI-04 vs. ¹⁸F-FDG PET/CT) were observed in all patients (91.7% vs. 76.7%), the initial staging group (90.9% vs. 79.5%), and the re-staging group (93.8% vs. 68.7%), all P values were less than 0.001. Among these misdiagnosed patients, 80.0% (4/5) were overestimated on ¹⁸F-FAPI-04 PET/CT and 100 % (14/14) were underestimated on ¹⁸F-FDG PET/CT (Table 4). Regarding the diagnostic consistency of ¹⁸F-FAPI-04 and ¹⁸F-FDG PET/CT, ¹⁸F-FAPI-04 PET/CT amended the underestimation of ¹⁸F-FDG PET/CT in 19 patients (31.7%, 19/60), including 10 patients in the initial staging group and 9 patients in the restaging group (Table 5). As demonstrated in Table 6, the overall consistency of the oncological management recommended by ¹⁸F-FAPI-04 PET/CT with the reference standard was much higher than that of ¹⁸F-FDG PET/CT in the re-staging group. However, it is worth noting that the differences were not significant (P=0.098) in the initial staging group. Similarly, regarding the management consistency recommended by ¹⁸F-FAPI-04 and ¹⁸F-FDG PET/CT, among the patients who would have been misdirected according to ¹⁸F-FDG PET/CT, the treatment plans of 13 were corrected by ¹⁸F-FAPI-04 PET/CT (7 in the initial staging group and 6 in the restaging group), which means that ¹⁸F-FAPI-04 PET/CT prompted management changes in 13/60 (21.7%) patients (see details in supplementary data 3).

Our findings have demonstrated that ¹⁸F-FAPI-04 has superiority over ¹⁸F-FDG mostly in the following cases: precise localization of the primary tumor and metastatic lesions, high activity retention with favorable contrast, high tumor-to-background ratio, accurate tumor staging, and potential impact on clinical management. Representative cases of the clinical application of this novel PET radiotracers were displayed in Figs. 1, 2, 3, 4, 5, 6 and Supplementary Figs. 2 to 5.