Head-to-head intra-individual comparison of biodistribution and tumor uptake of ⁶⁸Ga-FAPI and ¹⁸F-FDG PET/CT in cancer patients

Altered metabolism is one of the hallmarks of cancer [1]. The metabolic switch to utilize glucose (often equated to the Warburg effect) has been successfully exploited by ¹⁸F-FDG PET imaging since the 1980s. ¹⁸F-FDG is the most frequently used tracer for oncological PET imaging. However, its known limitations are high uptake in many normal tissues, low glucose transporter and hexokinase activity in certain tumor types, and a lack of specificity [2, 3].

PET imaging of the fibroblast activation protein (FAP) expression in cancer was recently introduced [4, 5]. The first generation of radiolabeled FAP inhibitors (FAPI) is peptidomimetic quinoline derivatives that bind with high affinity to FAP expressed on cancer-associated fibroblasts (CAFs). CAFs are a key component of the tumor stroma or tumor microenvironment. The tumor stroma consists of a variety of benign cells which interact with the tumor cells to promote growth, invasion, and metastasis. However, these cancer-associated cells, which are thought to promote tumor growth, often express unique receptors not found in the same cells outside of the tumor microenvironment. Therefore, an extensive stroma is often associated with metastatic spread and poor prognosis [6, 7].

While many activated CAFs overexpress FAP, ubiquitous resting fibroblasts in healthy tissue have no or low FAP expression. FAP is a type II transmembrane serine protease with post proline dipeptidyl peptidase as well as endopeptidase activity [8]. FAP is overexpressed by stromal fibroblasts in over 90% of epithelial carcinomas. However, FAP expression is not cancer specific: many activated fibroblasts express FAP in wound healing and in nonmalignant diseases such as chronic inflammation, rheumatological diseases, myocardial infarction, lung fibrosis, or liver cirrhosis, among others [8, 9].

The objective of this multicenter retrospective analysis was to perform a head-to-head comparison of ⁶⁸Ga-FAPI to ¹⁸F-FDG in patients with a variety of tumor types to establish generalizable differences between the two agents. Therefore, we intra-individually compared the biodistribution in normal tissue and tumor uptake of ⁶⁸Ga-FAPI and ¹⁸F-FDG.

This is an international multicenter retrospective analysis for the intra-individual comparison of ¹⁸F-FDG with newly developed ⁶⁸Ga-labeled FAP inhibitors [4, 5]. Regardless of the various equipment and imaging protocols used, we report high tumor-to-background ratios for ⁶⁸Ga-FAPI, comparable or superior to ¹⁸F-FDG that may result in high diagnostic performance for cancer staging and restaging.

Our findings suggest that ⁶⁸Ga-FAPI may improve tumor delineation in organs with physiologically high ⁸F-FDG background activity. This may be relevant in cancers such as pancreatic, ovarian, head-and-neck, hepatocellular and cholangiocarcinoma, colon cancer, as well as brain tumors and lung cancer [12‐15]. Due to low ⁶⁸Ga-FAPI uptake in most normal parenchyma, favorable tumor delineation was found in head-and-neck region (Fig. 4), liver metastases (Fig. 5), and abdominal cancer (Fig. 6). Furthermore, in the mediastinal region, ⁶⁸Ga-FAPI uptake in the cardiac muscle is very low compared to ¹⁸F-FDG (Fig. 7). However, we also examined a diffuse large B-cell lymphoma which showed relatively low ⁶⁸Ga-FAPI uptake. FAP-negative tumor phenotypes have also previously been described in the literature, e.g., in differentiated thyroid carcinoma and renal cell cancer [16‐18]. Unfortunately, in this work, we can only report the mean values over a relatively heterogenic group of patients. Further research will be needed to explore dedicated kinds of tumor where staging per FAPI PET might provide a clinical advantage in comparison to FDG PET. In addition to tumor entity, it would also be interesting whether FAPI uptake depends on tumor aggressiveness or proliferation rate as it is for ¹⁸F-FDG. However, due to small sample sizes within the respective histopathological subtypes, we could not reasonably perform a correlation between grading and FAPI uptake, yet. This investigation would have to be carried out in a future evaluation.

According to previous reports about their respective biodistribution and pharmacokinetics, the ligands FAPI-02, FAPI-04, and FAPI-46 do not differ in early uptake phase 10 min to 1 h p.i. and were thus considered widely exchangeable with regard to our diagnostic evaluation. In comparison to FAPI-02, the ligands FAPI-04 and FAPI-46 have longer tumor retention time beyond 1 h p.i., which however would only be relevant in the perspective of FAP-targeted radionuclide therapy [18, 19]. FAPI-74 either can be used as ⁶⁸Ga-FAPI-74 (this work) or alternatively could also be labeled with ¹⁸F providing an advantage in large batch production. However, biodistribution at 1 h p.i. was reported comparable to previous ligands [20].

⁶⁸Ga-FAPI and ¹⁸F-FDG reflect two different aspects of tumor behavior. ¹⁸F-FDG directly targets tumor cell metabolism. In contrast, ⁶⁸Ga-FAPI targets FAP on CAFs in the stroma of tumors, which is an indicator of desmoplastic reaction and was reported to be one of the key determinants of tumor immunity [19] and multidrug resistance [21] possibly related to reductions in transtumoral transport of cells and drugs. Thus, in addition to its function as a staging modality, FAPI PET/CT may be helpful in understanding tumor biology related to the tumor microenvironment.

Due to the limited number of studies to date, the tumor entities with low or high ⁶⁸Ga-FAPI uptake are also not yet sufficiently known.

Furthermore, FAPI could also play a role in the examination of myocardial infarction and IgG4-related diseases [22, 23].

Our retrospective analysis has several limitations: ⁶⁸Ga-FAPI PET/CT was performed with four different ligands, which however share a common backbone and their early phase biodistribution as well as tumor uptake are comparable [18‐20]. Lacking gold standard validation of discrepant lesions, we are not able to evaluate the respective sensitivity, specificity, and accuracy of ¹⁸F-FDG and ⁶⁸Ga-FAPI, which was beyond the scope of the current work but has already been addressed by other researchers [24].

The main limitation is the long (up to 3 month) interval between the examinations, which may cause “interval progression” that might be non-neglectable in aggressive cancers. However, we had to find a good tradeoff between test vs. re-test reliability and inclusion of a sufficient large patient cohort. For future clinical trials, more stringent inclusion criteria—and alternating order of tracer administration—should are encouraged.

In this international multicenter retrospective analysis, 71 patients with various cancers underwent both ¹⁸F-FDG and ⁶⁸Ga-labeled FAP inhibitors. ⁶⁸Ga-FAPI showed equal or higher TBR at lower radiation doses than ¹⁸F-FDG. These findings suggest that ⁶⁸Ga-FAPI may demonstrate higher diagnostic performance for cancer staging and restaging in various indications. In addition, the high target to background ratios and the low uptake in normal organs also suggest potential use of FAPI ligands as a potential means of treating tumors with therapeutic radioisotopes.