Comparison of [68Ga]Ga-DOTA-FAPI-04 and [18F]FDG PET/MRI in the Preoperative Diagnosis of Gastric Cancer

Purpose Our objective was to compare the value of positron emission tomography/magnetic resonance imaging (PET/MRI) with the new imaging agent [68Ga]Ga-DOTA-FAPI-04 and the traditional imaging agent [18F]FDG for the preoperative diagnosis of gastric cancer. Methods Forty patients with gastric cancer diagnosed by gastroscopy in gastrointestinal surgery at our hospital from June 2020 to January 2021 were analyzed. All patients underwent simultaneous [68Ga]Ga-DOTA-FAPI-04 and [18F]FDG PET/MRI. The standard uptake value (SUV), fat removal standard uptake value (SUL), and diagnostic sensitivity, specificity, and accuracy for primary and metastatic lesions were compared, and their diagnostic value for different lymph node dissection stages was analyzed. Results The median age of the patients in this cohort was 68 years. Twenty-nine patients underwent surgery, and 11 patients underwent gastroscopic biopsy. The SUVmax of primary lesions in the FDG group and the FAPI group was 5.74 ± 5.09 and 8.06 ± 4.88, respectively (P < 0.01); SULmax values were 3.52 ± 2.80 and 5.64 ± 3.25, respectively (P < 0.01). The SUVmax of metastases in the two groups was 3.81 ± 3.08 and 5.17 ± 2.80, respectively (P < 0.05). The diagnostic sensitivities for primary lesions in the FDG group and the FAPI group were 0.72 and 0.94, respectively (P < 0.05). Combined with postoperative pathological staging, there was no difference in diagnostic sensitivity and specificity of lymph node staging between the FDG and FAPI groups (P > 0.05). Conclusion Compared with the traditional imaging agent, [68Ga]Ga-DOTA-FAPI-04 has better diagnostic efficiency but no substantial advantage for preoperative lymph node staging.


Introduction
Cancer-associated fbroblasts (CAFs) are the main components of the matrix around epithelial cancer cells and can selectively produce fbroblast activation protein (FAP). FAP is highly expressed in a variety of epithelial cancers, such as gastric cancer, colorectal cancer, esophageal cancer, and ovarian cancer, but it is almost not expressed in the matrix of normal tissues [1,2]. Based on this characteristic, fbroblast activation protein inhibitor (FAPI) has been used as the imaging agent for positron emission tomography (PET) in the last few years, for which [ 68 Ga]Ga-FAPI-04 was developed [3]. Because stromal cells account for 90% of the total weight of tumors, cell matrix-based targeted PET may be more sensitive than glucose metabolism PET imaging. Studies have shown that [ 68 Ga]Ga-FAPI-04 has stable performance and can refect some characteristics of diferent solid tumors [4,5]. Other studies have shown that [ 68 Ga]Ga-FPAI-04 positron emission tomography/computed tomography (PET/CT) results in clearer contours and a higher target-to-background ratio than [ 18 F]FDG PET/CT for solid tumors [6]. In addition, [ 68 Ga]Ga-FAPI-04 PET/CT has a higher uptake value and diagnostic accuracy for some suspicious tumors that cannot be characterized by [ 18 F]FDG PET/CT [7].
Gastric cancer is a common disease, and its occurrence is related to many factors, such as HP and garlic [8,9]. PET/ MRI has been widely used in the evaluation of gastric cancer in the last few years. It has the advantages of better soft tissue contrast, functional imaging, and less ionizing radiation. Its disadvantage is also obvious: it is easily afected by respiration and gastric peristalsis during imaging, which results in artifacts. It was found that [ 18 F]FDG PET/MRI has better advantages in preoperative TNM staging than PET/CT [10]. Another study showed that multidetector CT (MDCT) combined with [ 18 F]FDG PET/MRI improves the diagnostic accuracy of a preoperative M-stage in recurrent gastric cancer and has advantages in evaluating the resectability of lesions [11]. At

Imaging
Review. PET/MRI images were analyzed by two nuclear medicine physicians with experience in PET/ MRI for more than 2 years. Te standard uptake value (SUV), fat removal standard uptake value (SUL), diagnostic sensitivity, specifcity, and accuracy of primary and metastatic lesions were measured, and their diagnostic value for diferent lymph node dissection stages was analyzed. Te participants were not given information about the other PET/MRI scan. In case of disagreement in diagnosis, the two doctors discussed and reached a consensus.
Te main function of PET is to detect a lesion, and MRI images are used to confrm whether the hypermetabolic area is the tumor. TNM staging of gastric cancer was determined by referring to the 8th edition of the AJCC gastric cancer staging system. Te criteria for lymph node metastasis of gastric cancer were as follows: shortest diameter > 5 mm, necrotic signs in the center, high DWI signal and low ADC signal, and higher metabolism than the background.

Statistical
Analysis. Data are expressed as the mean-± standard deviation, and statistical analyses were conducted using SPSS 22.0 software. T tests were used to compare measurement data between two groups, and the chi-square test was used to compare count data. Sensitivity, specifcity, and accuracy were compared by the McNemar test. P < 0.05 was considered statistically signifcant.

Diagnostic Efciency of Lymph
Te sensitivity, specifcity, and accuracy of [ 18 F]FDG were 0.33, 0.82, and 0.62, and that of [ 68 Ga]Ga-DOTA-FAPI-04 were 0.58, 0.71, and 0.66, respectively. Te sensitivity, specifcity, and accuracy of the two groups in the N1 phase were 0 and 0.33, 0.77 and 0.81, and 0.69 and 0.76, respectively, those of the N2 phase were 0 and 0, 0.96 and 0.96, and 0.83 and 0.79, and those of the N3 stage were 0 and 0, 0.82 and 0.81, and 0.79 and 0.72. However, there were no substantial diferences between the two groups ( Table 5, P > 0.05).

Discussion
Te standard treatment of gastric cancer depends on accurate preoperative staging [12], which is also important for metastatic gastric cancer. Te purpose of this prospective study was to compare the diagnostic value of the new imaging agent [ 68 Ga]Ga-DOTA-FAPI-04 and the traditional agent [ 18 F]FDG. Our results showed that the maximum uptake value of [ 68 Ga]Ga-DOTA-FAPI-04 was better than that of [ 18 F]FDG, with sensitivity in the diagnosis of primary lesions of gastric cancer being better than that of [ 18 F]FDG. Nonetheless, there was no substantial diference in the sensitivity and specifcity of lymph node staging between the two groups.
Te uptake values of SUV max and SUL max in the FAPI group were higher than those in the FDG group, which was consistent with the results of Chen et al. [13]. Some studies have shown that uptake of [ 18 F]FDG is lower in difuse gastric cancer, gastric mucinous adenocarcinoma, and signet ring cell carcinoma, which afects the diagnosis of gastric cancer [14]. Terefore, our results suggest that [ 68 Ga]Ga-DOTA-FAPI-04 may be able to compensate for this defciency, even though we did not conduct a subgroup analysis of gastric cancer histology type in this study. Further diagnostic analysis found that the sensitivity of [ 68 Ga]Ga-DOTA-FAPI-04 for primary lesions was 0.94, which was considerably higher than that of [ 18 F]FDG. Tis was consistent with the study of Guo et al. [15]. We speculate that this is related to the high uptake rate of [ 68 Ga]Ga-DOTA-FAPI-04. Because all patients were diagnosed with gastric cancer by gastroscopy, specifcity could not be compared for the two groups. In the comparison of diagnostic efcacy for metastases, we found that the sensitivity of the FAPI group tended to increase (0.58 and 0.33, respectively), but there was no substantial diference in the P value. We speculate that this may have been due to the inconsistent judgment criteria for positive lymph nodes. In previous studies, the cutof value of a lymph node's short diameter (usually 5 mm) was used as the criterion for determining positive lymph nodes [16,17]. However, infammation can also lead to lymph node enlargement, and even metastatic lymph nodes do not necessarily show a volume increase. Tese factors may afect the diagnostic efcacy of PET/CT or PET/MRI for lymph node metastasis. In addition, recent studies have shown that lymph node metastasis can be identifed by DWI sequences and ADC images [18]. Terefore, the MRI signal combined with SUV max was used to identify lymph node metastasis in this study. Although our data show that SUV max of the FAPI group and FDG group was not signifcantly diferent, we speculate that this may have been due to the small number of cases or selection bias. However, the increasing trend of SUV max in the FAPI group suggests that [ 68 Ga]Ga-DOTA-FAPI-04 had certain advantages in determining lymph node metastasis in gastric cancer.
We attempted to compare the diagnostic efciency of the two imaging agents for diferent N stages of gastric cancer, though the results showed that the diference in diagnostic sensitivity and specifcity between the two groups was not signifcant at N0 or N1-N3. We speculate the following reasons in addition to the factors of positive diagnostic criteria for lymph nodes. First, infammation in lymph nodes may lead to increased uptake. Indeed, it has been reported that infammation may lead to an abnormal increase in FDG metabolism in lymph nodes, thus increasing the false-negative rate [19]. Second, there may have been selection bias with regard to the patients enrolled. Only 41.4% (12/29) of the patients had lymph node metastasis confrmed by postoperative pathology. Te small number of N1, N2, and N3 stage cases    Of course, this study also had some limitations, such as the small number of enrolled cases and the lack of data on tumor T staging. In addition, the number of patients with dMMR and PDL-1 positivity was relatively small, and there was a lack of subgroup analysis. Prospective studies with a larger number of patients in the future may provide more evidence.

Data Availability
Te data used to support the fndings of this study are available from the corresponding author upon request.

Conflicts of Interest
Te authors declare that there are no conficts of interest regarding the publication of this paper.  Canadian Journal of Gastroenterology and Hepatology 5