Jiang et al. attempted to label NOTA-FAPI-04 with [¹⁸F]AlF, the radioactivity per batch was 9.095 ± 0.587 GBq and the radiochemical yield (RCY) was 26.4 ± 1.5%, and the radiotracer obtained by the automated synthesis equipment was used for PET/CT imaging of a patient with liver metastases from invasive cholangiocarcinoma (IDC). The images showed high tumor-to-background uptake ratios (TBRs) of 8.44 (28). Dahl et al. successfully produced good manufacturing practice (GMP)–compliant ¹⁸F-FAPI-74 in the GE TRACERlab FX2N module and the cassette-based module Trasis AllInOne (AIO) in radioactivity per batch of 6410 MBq and 6900 MBq, respectively, with the RCY of about 20% (29). Naka et al. performed a proof-of-concept automated synthesis of [¹⁸F] AlF-FAPI-74 with one-pot and one-step by a multi-purpose synthesizer CFN-MPS200; the radioactivity per batch was 11.3 ± 1.1 GBq, and the RCY was 37.0% ± 4.3% (30).

FAPIs coupled to radionuclide 68Ga via DOTA chelator are mainly FAPI-04, FAPI-46, FAPI-74, DOTA-2P(FAPI)₂, DOTA.SA.FAPI, and DATA5m.SA.FAPI.

FAPI-04 is a small-molecule FAPI derivative that has been chemically modified to provide better imaging results, with effective doses of 1.64E-02mSv/MBq. Kratochwil et al. performed ⁶⁸Ga-FAPI-02 and ⁶⁸Ga-FAPI-04 PET/CT imaging on 80 patients and obtained uptake data for tracers of 28 types of tumors (31). Among them, sarcoma, esophageal, breast, cholangiocarcinoma, and lung cancer showed high uptake rates with Maximum Standardized Uptake Value (SUVmax) >12; liver tumors and colorectal tumors, head and neck tumors, pancreatic tumors, ovarian tumors, and prostate tumors had moderate uptake rates (6 < SUVmax < 12). Pheochromocytoma, renal cell carcinoma, differentiated thyroid cancer, adenoid cyst, and gastric cancer had the lowest uptake rates (SUVmax < 6). The images from this study were named “Image of the Year” at the 2019 North American Nuclear Medicine Annual Meeting ( Figure 1 ). Giesel et al. analyzed the biodistribution and preliminary dosimetric data of ⁶⁸Ga-FAPI-02 and ⁶⁸Ga-FAPI-04 in 50 patients containing 14 different types of tumors (32). Compared with ¹⁸F-Fludeoxyglucose (¹⁸F-FDG), ⁶⁸Ga-FAPI showed lower uptake in the brain, liver, and oral/pharyngeal mucosa and no significant difference in tumor and other normal tissues, showing higher TBRs. Wang et al. performed the first biodistribution and internal radiation dosimetry after ⁶⁸Ga-FAPI-04 injection in six Chinese adults, and the results were in general agreement with the findings of Giesel et al. (32, 33).

FAPI-46 is another FAPI derivative with highly desirable imaging results, with effective doses of 7.80E-03mSv/MBq. The initial clinical imaging study work with FAPI-46 included eight patients, and compared with FAPI-04, ⁶⁸Ga-FAPI-46 had a higher TBR and has a longer tumor retention time, making it ideal for tumor imaging and targeted therapy (18). After that, a study by Meyer et al. showed that ⁶⁸Ga-FAPI-46 PET/CT has good dosimetric properties with high TBR increasing over time (34).

Zhao Let al. synthesized DOTA-2P(FAPI)₂ with two FAPI-46 monomers (22). ⁶⁸Ga-DOTA-2P(FAPI)₂ has an effective dose of 1.19E-02mSv/MBq, higher than that of ⁶⁸Ga-FAPI-46 and comparable to ⁶⁸Ga-FAPI-02 (1.80E-02mSv/MBq) and ⁶⁸Ga-FAPI-04. Intra-tumoral uptake of ⁶⁸Ga-DOTA-2P(FAPI)₂ is higher than that of ⁶⁸Ga-FAPI-46 (SUVmax of 8.1–39.0 vs. 1.7–24.0) in most tumor lesions in nasopharyngeal carcinoma, papillary thyroid carcinoma, and hepatocellular liver cancer. The above studies suggest that FAPI dimers exhibit better in vivo pharmacokinetics and higher tumor uptake than FAPI monomers, leading to a more apparent contrast of primary lesions and metastases with normal tissue, thus potentially improving the sensitivity of lesion detection. However, the tracer showed higher physiological uptake in the thyroid and pancreas and prolonged retention time in the blood pool. The data from this study need to be validated with a larger sample size.

Moon et al. reported FAPI derivatives DOTA.SA.FAPI and DATA5m.SA.FAPI with IC₅₀ of 0.7–1.4 nM; SA replaces the heterocyclic nitrogen portion of FAPI-04 to form a squaramide unit. The chelators used were macrocyclic DOTA and the hybrid chelator DATA5m (21). In animal studies, ⁶⁸GaDOTA.SA.FAPI had good TBR, comparable to the previously reported animal data for FAPI-04 (21). In clinical studies, ⁶⁸Ga-DATA5m.SA.FAPI PET/CT showed specific uptake in focal nodular hyperplasia in the liver (35), and there was no significant difference in tumor uptake between 68Ga-DOTA.SA.FAPI and ¹⁸F-FDG compared with ¹⁸F-FDG in 14 types of tumor imaging in a total of 54 patients, with the exception of brain metastatic lesions (27). Next, the team found that radiolabeling SA.FAPI with AAZTA5 as a chelator (i.e., AAZTA5.SA.FAPI) was superior to both DOTA and DATA5m in terms of labeling process and drug yield, and the IC₅₀ of the product was in the low nanomolar range (0.55–0.57 nM), but the drug is still in preclinical studies (26).

With the development of ¹⁸F production equipment and labeling technology, the advantages of ¹⁸F clinical application will gradually increase: first, the maximum β⁺ energy of ¹⁸F is significantly lower than that of ⁶⁸Ga, so the sensitivity and spatial resolution of ¹⁸F-PET imaging is better than ⁶⁸Ga-PET imaging, which can improve the image contrast to a certain extent; second, ⁶⁸Ga has become expensive due to a sharp increase in global demand and thus a shortage of supply (36); in addition, the ⁶⁸Ge/⁶⁸Ga generator is not yet capable for mass production of ⁶⁸Ga due to technical limitations, which means that the automated production of large quantities of FAPIs radiotracers is limited. In an automated synthesis study of ⁶⁸Ga-FAPI-46 by Spreckelmeyer et al., it was found that the radioactivity at the end of the synthesis was only 700–1,700 MBq, which was only sufficient for the examination of four to six patients (37). In comparison, the long half-life of ¹⁸F (1.13 h) supports centralized commercial production and distribution to customers with no ¹⁸F production facilities, resulting in a significant reduction in production costs (38). The development of low-power benchtop cyclotrons has made ¹⁸F dispersion production easier (automated, low cost, and on-demand production); the ¹⁸F labeling of peptides has become more efficient due to the development of chelator and labeling technologies. Currently, in FAPI radiotracer research, an increasing number of researchers are experimenting with ¹⁸F labeling of FAPIs to develop radiopharmaceuticals that can meet the global demand for large-scale assays, especially in developing countries. For example, ¹⁸F-FAPI-74 and ¹⁸F-NOTA-FAPI-04 showed good visualization in clinical studies, and ¹⁸F-FGlc-FAPI-04 showed superior performance in in vitro experiments (20, 23, 24).

Giesel et al. applied ¹⁸F-FAPI-74 and ⁶⁸Ga-FAPI-74 for the first time in clinical practice and successfully attached [¹⁸F]AlF to FAPI via 1,4,7-triazacyclonane-N,N′,N′-triacetin acid (NOTA) for the first time and showed satisfactory imaging results. A total of 10 patients with non–small cell lung cancer in this study received ⁶⁸Ga-FAPI-74 and ¹⁸F-FAPI-74 imaging within 1 week. The data showed that the best tumor-to-background contrast presents 1 h after injection, and the estimated radiation dose of ¹⁸F-FAPI-74 was lower than that of other ¹⁸F-labeled PET scanner tracers, and there was no difference in tumor uptake between lung adenocarcinoma and lung squamous cell carcinoma. This study shows the value of ¹⁸F-FAPI-74 for lung cancer diagnosis and detection of small lung cancer lesions, but the number of cases is too small to analyze whether it is useful for staging and radiotherapy planning guidance (20) ( Figure 2 ). Others have attempted to label multiple FAPIs with [¹⁸F] AlF and found that ¹⁸F-FAPI-74 gave the most favorable results in an in vitro study, and they performed a visualization study in a patient with metastatic non–small cell lung cancer and obtained conclusions similar to those of Giesel et al. (20, 39).

¹⁸F-NOTA-FAPI-04 (¹⁸F-FAPI-42) was obtained by replacing the DOTA with chelator NOTA in FAPI-04 and Aqueous¹⁸F-Labeling by the Al¹⁸F chelation technique (21, 40). ¹⁸F-FAPI-42 was initially used to diagnose lung, pancreatic, colorectal, prostate, and lymphoma and was compared with ¹⁸F-FDG (21). The results showed that the excretion route of ¹⁸F-FAPI-42 was mainly in the urinary and biliary systems, similar to the previously reported tracers ¹⁸F-FAPI-74, ⁶⁸Ga-FAPI-04, and ⁶⁸Ga-FAPI-46; compared with ¹⁸F-FDG imaging, ¹⁸F-FAPI-42 had lower SUVmean in the liver, brain, spleen, and bone marrow; in addition, ¹⁸F-FAPI-42 bone uptake was low, indicating that this tracer was not defluorinated at the time of detection. Another study including a total of 22 patients with hepatocellular carcinoma, lung cancer, colorectal cancer, gastric cancer, esophageal cancer, cheek carcinosarcoma, abdominal neuroendocrine carcinoma, and clear cell carcinoma of the bladder conducted a human biodistribution study of ¹⁸F-FAPI-42 and compared it with ⁶⁸Ga-FAPI-04 (40). Data from this study showed that the optimal image acquisition time for ¹⁸F-FAPI-42 was 1 h after injection; ⁶⁸Ga-FAPI-42 showed higher physiological uptake in the parotid, salivary gland, thyroid, and pancreas compared with ⁶⁸Ga-FAPI-04; both had equal detection rates for lesions; in addition, ¹⁸F-FAPI-42 showed higher TBRs in the liver, bone, lymph nodes, pleura, and peritoneal metastases.

¹⁸F-FGlc-FAPI-04 is an [¹⁸F]-6-fluoro-6-deoxyglycosylated FAPI radiotracer with an IC₅₀ of 167 nM, much higher than ⁶⁸Ga-FAPI-04 (32 nM), showing higher tumor uptake and longer tracer clearance times were shown in HT1080-xenografts and U87MG-xenografts. However, because the tracer is mainly eliminated via the hepatobiliary pathway and sees intraosseous and intra-articular uptake, this has hindered the entry of the tracer into clinical studies (24).

FAPI-34, QCP01[1-(6-((3-((4-((2-((S)-2-cyanopyrrolidin-1-yl)-2-oxoethyl)-carbamoyl)quinolin-6-yl)oxy)propyl)amino)-6-oxohexyl)-2-((E)-2-((E)-3-(2-((E)-3,3-dimethyl-5-(trioxidaneylthio)-1-(4-(trioxidaneylthio)butyl)indolin-2-ylidene)ethylidene)-2-(4-(trioxidaneylthio)phenoxy)cyclohex-1-en-1-yl)vinyl)-3,3-dimethyl-3H-indol-1-ium-5-sulfonate)] and QCP02(2,2′,2″-(10-(1-carboxy-4-((3-((4-((2-((S)-2-cyanopyrrolidin-1-yl)-2-oxoethyl)carbamoyl)quinolin-6-yl)oxy)propyl)amino)-4-oxobutyl)-1,4,7,10-tetraazacyclododecane-1,4,7-triyl)triacetic acid) both demonstrated low nanomolar inhibition (19, 25). FAPI-34, with an IC₅₀ of 6.4–12.7 nM, has been used for SPECT imaging of patients with ovarian and pancreatic cancer metastases, showing good imaging results. When PET imaging is not possible due to equipment, ^99mTc-FAPI-34 can be used as the imaging agent for SPECT imaging, which can also be used for diagnosis and efficacy evaluation (19). QCP01 and QCP02 were applied in tumor transplantation models with higher tumor and normal organ uptake than FAPI-02 and FAPI-04, indicating that QCP01 and QCP02 have high affinity for FAP. In addition, ¹¹¹In-QCP02 SPECT imaging showed higher tumor uptake and lower kidney uptake, which is still in the preclinical study stage (25).

With FAPI-specific PET imaging, patients do not require dietary preparation, and high-quality images can be obtained early (10 min to 1 h) after tracer injection (20, 41) ( Figure 3 ).

In head and neck tumors, compared with ¹⁸F-FDG, it was able to improve the staging, target area contour, and prognostic assessment of oral squamous carcinoma, nasopharyngeal carcinoma, and adenoid cystic carcinoma. For example, in 45 patients with nasopharyngeal carcinoma, the detection rates of primary and metastatic foci by FDG were 97.0% and 75.2%, respectively, and the detection rates of primary and metastatic foci by FAPI-04 were 100% and 95.0%, respectively (42); FAPI-PET/CT detected more details of primary foci and lymph node metastases, resulting in a change in staging in 5 of 12 cases of adenoid cystic carcinoma (43). Among 18 patients with head and neck tumors in whom primary lesions were not detected by FDG PET/CT imaging, FAPI PET/CT detected primary lesions in 5 of them (44). However, FAPI may not be advantageous in detecting lymph nodes and distant metastases (41, 44) ( Figure 4 ).

In malignant brain lesions, cranial MRI is currently the primary imaging method for diagnosis, staging, and outlining radiotherapy target areas for high-grade gliomas in the brain. However, PET/CT and PET/MRI often provide more tumor progression and prognostic assessment (45). Conventional ¹⁸F-FDG tracers have high physiological uptake in the brain, resulting in low TBRs and obscuring accurate information about the tumor. FAPI has low physiological uptake in the brain and is highly expressed on the cell surface of high-grade gliomas (excluding diffuse astrocytomas). Clinical studies have shown high uptake of FAPI imaging agents in high-grade gliomas (SUVmax = 2.8 ± 0.6), whereas low-grade gliomas showed only mild uptake (SUVmax = 0.35 ± 0.10) (46, 47). Therefore, FAPI imaging agents may be helpful in the detection and target area sketching of high-grade gliomas in the brain, non-invasive grading of gliomas, and assessment of prognosis after radiotherapy. In addition, it showed higher TBR for brain metastases from gastric, breast, lung, and liver cancers, with a higher detection rate than¹⁸F-FDG (47–50).

FAPI PET/CT shows high uptake and high TBRs in primary and lymph node metastases of non–small cell lung cancer, and the detection rate of lymph node metastases is higher than ¹⁸F-FDG and thus may improve the staging of lung cancer and contribute to the outline of radiotherapy target areas (20, 50).

¹⁸F-FDG PET and PET/CT are less sensitive for detecting regional lymph node metastases in esophageal squamous cell carcinoma (about 66%) (51). Because of the high expression of FAP in esophageal cancer, FAPI PET shows a high uptake rate (SUVmax > 12) and high TBR in esophageal cancer and its metastases, and lymph node metastases of esophageal cancer with a false-negative ¹⁸F-FDG test are strongly positive by FAPI test, which improves the detection rate of lymph node lesions (31). The high TBR and high sensitivity of FAPI for primary foci and lymph node metastases provide more information for clinical purposes, both to improve the staging of esophageal cancer and to help outline the target volume (GTV) in precision radiotherapy. ¹⁸F-FDG has limitations in detecting certain pathological types of gastric malignancies, such as non-mesenchymal diffuse carcinoma, mucinous carcinoma, and indolent cell carcinoma (52, 53). In contrast, in some small sample studies, although FAPI showed mostly low uptake in gastric cancer, it also showed high TBR, which may still bring a turnaround for PET/CT detection of the abovementioned tumors (31). For example, in a study that included 10 patients with primary gastric tumors, the detection rates of FDG and FAPI were 50% (5/10) and 100% (10/10), respectively (47). In addition, it has been suggested that FAPI appears to be the best performing imaging agent for PET imaging of liver malignancies at present. Because of the high non-specific uptake of ¹⁸F-FDG in the liver, the false-negative rate of ¹⁸F-FDG PET in detecting HCCs is high, about 40%–50% (54). In a study of 25 patients with suspected primary liver malignancies (hepatocellular carcinoma/intrahepatic cholangiocarcinoma) on ⁶⁸Ga-FAPI-04 imaging, the detection rate of primary liver malignancies was higher with FAPI than with ¹⁸F-FDG (96% vs. 65%) (55). Another study that included 20 patients with suspected primary malignancy of the liver also showed a higher sensitivity of FAPI for the diagnosis of malignancy (100% vs. 58.8%) (56).

Breast carcinoma (15%), ovarian carcinoma (5%), and uterine carcinoma (4%) are the most common gynecologic malignancies, and accurate staging is essential for treatment guidance (1). The most commonly used radiotracer for PET/CT diagnosis of gynecologic malignancies is ¹⁸F-FDG. A multicenter prospective study including 31 gynecologic malignancies showed higher TBRs in breast, ovarian, cervical, endometrial, and smooth muscle metastases; uterine smooth muscle sarcoma; and fallopian tube cancer. FAPI PET had higher TBRs in metastatic lesions than ¹⁸F-FDG [regional lymph node metastases (31.9 vs. 27.4) and distant metastases (13.0 vs. 5.7)]. However, there were significant differences in FAPI uptake between premenopausal and postmenopausal patients in the endometrium (mean SUVmax of 11.7 vs. 3.0) and breast (mean SUVmax of 1.8 vs. 1.0), which may affect the detection of lesions at these sites (57) ( Figure 5 ).

In a case report of liver metastasis from malignant melanoma, the lesion showed moderate ⁶⁸Ga-FAPI-04 uptake (SUVmax of 6.5) but was not compared with the current PET standard tracer ¹⁸F-FDG (58). In another retrospective analysis of FAPI PET/CT imaging of rare malignancies, patients with melanoma (one case) also showed more favorable imaging results (59).

In conclusion, FAPI-specific tracers simplify the clinical workflow and have significant advantages in detecting malignancies in the brain, liver, and other sites.

IgG₄-related disease (IgG₄-RD) is an autoimmune inflammatory disease that occurs mainly in the pancreas, biliary tract, salivary glands, kidneys, aorta, and other organs, and the disease is often accompanied by infiltration of IgG₄+ cells or (and) activated fibroblasts (60). A study that included 26 patients with IgG₄-RD showed that ⁶⁸Ga-FAPI PET/CT detected an additional ¹⁸ (13.2%) involved organs out of 136 involved organs in 26 patients compared with ¹⁸F-FDG PET/CT and had a higher positive rate than ¹⁸F-FDG PET/CT in detecting pancreatic, bile duct/liver, and lacrimal gland involvement, but the former was at a disadvantage in detecting lymph node involvement (60). Co-imaging with ⁶⁸Ga-FAPI-04 and ¹⁸F-FDG-PET may enable non-invasive tracking of the evolution of IgG₄-RD from inflammation to fibrosis (60, 61). In a FAPI PET/CT imaging of patients with early-stage immune checkpoint inhibitor (ICI)–associated myocarditis, the heart showed specific uptake (62). High tracer accumulation was seen in the left ventricular myocardium on FAPI PET/CT imaging in a patient undergoing chemotherapy for pancreatic cancer. These study data may provide new ideas for early diagnosis of myocardial injury and intervention (63). In a case of thyroiditis with ⁶⁸Ga-FAPI PET/CT, the uptake rate of inflammatory lesions (SUVmax of 4.2) was similar to that of hypofractionated thyroid cancer (SUVmax < 6), suggesting that FAPI PET/CT has difficulty in differentiating hypofractionated thyroid cancer from benign lesions (31, 64). Monitoring the activity of aortitis often requires a joint assessment of imaging data, clinical data, and biological data, which are not always correlated, making the work challenging. A patient with aortitis underwent ⁶⁸Ga-FAPI-PET/CT, showing mild thickening of the walls of the bilateral carotid, subclavian, and thoracoabdominal aorta, where there is no specific uptake of ¹⁸F-FDG. Therefore, ⁶⁸Ga-FAPI-PET/CT may be superior to ¹⁸F-FDG in monitoring the activity of aortitis (65). In two case reports, the TB lesions showed high uptake of the FAPI tracer with higher intensity than ¹⁸F-FDG; this helps in assessing extra-pulmonary tuberculosis lesions (66, 67).

The intensity of FAPI imaging agent uptake by various tumors is closely related to the degree of FAP expression, e.g., breast cancer and esophageal cancer, where FAP expression levels are high and tumor uptake of FAPI imaging agent is very high (SUVmax >12) (9, 31). The non-specific uptake of FAPI visualizers also correlates with the FAP expression profile, with FAP expression levels often upregulated during stromal remodeling, as in wound healing, and these sites tend to exhibit higher FAPI visualizer uptake (68, 69). Interestingly, cancers such as colon and pancreatic cancers, which have the highest desmoplastic reaction by histopathology, exhibit only moderate 68Ga-FAPI uptake (SUVmax of 6–12), this may be due to the variability of cancer, i.e., cancer development is complex and FAP expression varies at different stages of stromal remodeling, a conjecture that requires large sample studies to verify (31, 70).

Stromal remodeling occurs not only in malignancies but also in development, wound healing, chronic inflammation (e.g., arthritis, atherosclerotic plaques, and fibrosis), and certain physiologic processes where FAP is often highly expressed, posing challenges for differentiating benign from malignant, tumor and peritumor chronic inflammation in FAPI imaging (71). For example, a clinical study including 92 patients found high nonspecific uptake of FAPI-04 and FAPI-46 at sites where scarring (SUVmax of 7.7 ± 3.3) or degenerative lesions (SUVmax of 7.7 ± 2.9) occurred in patients, i.e., these sites may show false-positive results on tumor detection (69); one patient with pancreatic cancer who developed chronic peritumor inflammation due to radiotherapy was unable to identify the target volume for radiotherapy with FAPI tracer (72). In addition, some sites with low or moderate FAP expression, such as the uterus and breast, also showed higher FAPI tracer uptake in a recent study with SUVmax of 12.2 ± 7.3 and 4.5 ± 1.5, respectively, and this difference may be related to the biological age of women, and large sample studies are needed to confirm this finding (69, 73). Furthermore, FAPI-specific PET is still controversial in diagnosing bone metastases and lymph node metastases in different tumors (44, 74, 75), whether it is superior to ¹⁸F-FDG in diagnosing primary colorectal cancer and pancreatic cancer combined with chronic inflammation of the pancreas also needs further study (76). In another way, changes in the stroma during tumor development may lead to changes in the expression of FAP, so whether different aspects of tumor variability may have an impact on FAPI imaging results needs to be further explored. In brief, FAPIs cannot yet replace the work of FDG, but they can be used as a complement to it.