[ 68 Ga]Ga-DOTA-FAPI-04 PET/CT in The Staging of Gastric Cancer: Improved Diagnostic Ecacy Conrmed By Postoperative Histopathology

Purpose This study aimed to compare the diagnostic performance of [ 68 Ga]Ga-DOTA-FAPI-04 and [ 18 F]-FDG PET/CT in the primary and metastatic lesions of gastric cancer. Fifty-six patients with histologically proven gastric carcinomas were enrolled in this study. Each patient underwent both [ 18 F]-FDG and [ 68 Ga]Ga-FAPI-04 PET/CT within one week. Activity of tracer accumulation in lesions were assessed by maximum standardized uptake value (SUV max ) and TBR (lesions SUV max / ascending aorta SUV mean ). Histological work-up including immunohistochemical staining for FAP served as a standard of reference. which both showed remission. [ 68 Ga]Ga-FAPI PET/CT can better detect primary gastric cancer and metastatic lesions in peritoneum, abdominal LNs and bone, showing high usefulness in guiding N staging. Furthermore, [ 68 Ga]Ga-FAPI PET/CT provides more information for patients with recurrence detection and also has great potential in monitoring response to treatment. the activity exceeded that of the adjacent background tissues, excluding the possibilities of physiological uptake, trauma, infection, and inammation. The regions of interest (ROIs) were drawn around the lesions on the transverse slices for semiquantitative analysis. The maximum standard uptake value (SUV max ) was automatically calculated, and the target-to-background ratio (TBR) was calculated by dividing the lesion SUV max by the SUV mean of the ascending aorta. Semiquantitative assessment was divided into patient-based and lesion-based investigations. The former included the highest single lesion of the primary tumor or metastases in each organ/region, while the latter referred to the analysis including all lesions ( ≤ 5) or the 5 lesions with highest activity (>5) if there were multiple metastases. The visually interpreted PET/CT results were compared with pathological results obtained from gastroscopy or surgery. If tissue diagnosis was not applicable, the follow-up data including the results of laboratory tests and medical imaging could also serve as a reference for tumor diagnosis. Pathological TNM stage was assigned based on the eighth edition of the American Joint Committee on Cancer staging system [17]. SUV max -FAPI and TBR-FAPI both showed positive correlations with primary tumor depth (r = 0.303, P = 0.043 for SUV max -FAPI; r = 0.471, P = 0.001 for TBR-FAPI). However, primary tumor depth had no correlations with SUV max -FDG or TBR-FDG (r = 0.201, P = 0.190 for SUV max -FDG; r = 0.270, P = 0.077 for TBR-FDG). Pathological results showed 17 patients with signet-ring cell carcinoma (SRCC) and the remaining 28 patients without SRCC. In SRCC group, SUV max and TBR of [ 68 Ga]Ga-FAPI were signicantly higher than those of [ 18 F]-FDG (mean SUV max , 10.38 vs. 6.17, P = 0.001; mean TBR, 11.52 vs. 4.92, P < 0.001). In non-SRCC group, only the TBR of [ 68 Ga]Ga-FAPI was signicantly higher than that of [ 18 F]-FDG (11.69 vs. 6.35, P < 0.001), whereas the difference of SUV max was not signicant (10.16 vs. 9.25, P = 0.248). Seventeen patients pathological TNM stage, who underwent surgery due to early imaging stages. The TBR-FAPI of T3-4, N1-3 and III-IV groups signicantly higher than that of T1-2, N0 and I-II groups, respectively (Z = -2.319, -2.111 and -2.111, P = 0.019, 0.037 and 0.037, respectively).


Introduction
Gastric cancer is the fth most diagnosed malignancy and ranks the third most common cause of cancer-related deaths worldwide [1], with over 1 million estimated new cases annually and 784,000 deaths globally in 2018. Although the diagnosis and treatment of gastric cancer have improved, the prognostic outcome associated with this malignancy remains disappointing [2]. Therefore, early diagnosis, accurate staging and quantitative evaluation are of great importance for the treatment management and prognosis of these patients [3].
In clinical routine practice, gastroscopy and imaging examinations are the main approaches for diagnosing gastric cancer. Moreover, imaging modalities, such as computed tomography (CT) and magnetic resonance imaging (MRI), have been used for the primary staging of gastric cancer. However, CT and MRI are inadequate for staging, especially regarding the distant metastasis. [ 18 F]-Fluorodeoxyglucose (FDG) positron emission tomography-computed tomography (PET/CT) is a potentially valuable imaging modality for providing clinically relevant information on nodal staging and metastatic status of gastric cancer [4]. But [ 18 F]-FDG PET/CT sometimes suffers from low sensitivity in the detection of primary lesions of gastric cancer due to the signi cant variation of [ 18 F]-FDG uptake in different histologic types of gastric cancer [5]. In addition, the low-to-moderate sensitivity for the detection of lymph node metastases and peritoneal metastases further limits the use of [ 18 F]-FDG for determining the clinical stage of gastric cancer. Hence, there is an urgent need for developing more sensitive PET probes to improve the characterization of gastric cancer and to contribute to individualized patient care.
Previous studies have shown that gastric cancer can take up [ 68 Ga]Ga-FAPI, and more abnormal foci can be detected with [ 68 Ga]Ga-DOTA-FAPI-04 than with [ 18 F]-FDG, especially in the liver regions, peritoneum, mesentery, and omentum [14]. However, to the best of our knowledge, no studies regarding the potential advantage of [ 68 Ga]Ga-FAPI in gastric cancer staging and response to chemotherapy have been reported.
Thus, the aim of this study was to investigate the potential usefulness of [ 68 Ga]Ga-DOTA-FAPI-04 PET/CT for the diagnosis of primary and metastatic lesions in gastric cancer by comparison with [ 18 F]-FDG PET/CT, and to evaluate the diagnostic e ciency and N staging of metastatic lymph nodes based on postoperative pathology, in which FAP expression was con rmed by immunohistochemistry.  Primary tumors and metastatic lesions were identi ed as positive if the activity exceeded that of the adjacent background tissues, excluding the possibilities of physiological uptake, trauma, infection, and in ammation. The regions of interest (ROIs) were drawn around the lesions on the transverse slices for semiquantitative analysis. The maximum standard uptake value (SUV max ) was automatically calculated, and the target-to-background ratio (TBR) was calculated by dividing the lesion SUV max by the SUV mean of the ascending aorta. Semiquantitative assessment was divided into patient-based and lesion-based investigations. The former included the highest single lesion of the primary tumor or metastases in each organ/region, while the latter referred to the analysis including all lesions (≤5) or the 5 lesions with highest activity (>5) if there were multiple metastases. The visually interpreted PET/CT results were compared with pathological results obtained from gastroscopy or surgery. If tissue diagnosis was not applicable, the follow-up data including the results of laboratory tests and medical imaging could also serve as a reference for tumor diagnosis. Pathological TNM stage was assigned based on the eighth edition of the American Joint Committee on Cancer staging system [17].

Statistical analysis
Statistical analyses were performed on SPSS software (23.0, IBM Inc.). Continuous variables are expressed as mean ± SD. Categorical variables are expressed as number and percentage. Wilcoxon signed-rank test was used to assess the differences of SUV max and TBR between two groups. The number of positive lesions was compared using the chi-squared test. The McNemar's text and chi-squared test were applied to compare the differences of sensitivity, speci city, accuracy, positive predictive value (PPV) and negative predictive value (PPV) between [ 68 Ga]Ga-FAPI and [ 18 F]-FDG scans. Correlation between the two parameters was determined using Spearman test. A p-value of < 0.05 was de ned as statistically signi cant.

Patients' characteristics
The ow diagram of the study design is presented in Fig. 1. 56 patients with gastric cancer con rmed by surgery or gastroscopy were enrolled in this study. A total of 625 abdominal lymph nodes were resected from 17 patients who underwent surgery, of which 16.6% (104/625) metastatic lesions in 11 patients. Patient characteristics are listed in Table 1.  Seventeen patients had pathological TNM stage, who underwent surgery due to early imaging stages. The TBR-FAPI of T3-4, N1-3 and III-IV groups were signi cantly higher than that of T1-2, N0 and I-II groups, respectively (Z = -2.319, -2.111 and -2.111, P = 0.019, 0.037 and 0.037, respectively). No signi cant differences were observed in other semiquantitative parameters, such as SUV-FAPI, SUV max -FDG and TBR-FDG in these groups (Table 3).    (Fig. 2), which CT showed diffuse thickening gastric wall. Previous studies had shown that [ 18 F]-FDG PET/CT had lower sensitivity and tracer uptake in SRCC and mucinous carcinoma than in conventional adenocarcinoma [25][26][27], which was resulted from the relatively low expression level of glucose transporter 1 (GLUT-1) [28]. Therefore, a subgroup analysis was performed in our study. We found that in 17 patients with SRCC, the tracer uptake of [ 68 Ga]Ga-FAPI (SUV max , 10.38 ± 3.12; TBR, 11.52 ± 4.63) was signi cantly higher than that of [ 18 F]-FDG (SUV max , 6.17 ± 3.94; TBR, 4.92 ± 3.41). In addition, analysis of all primary gastric tumors also showed that [ 68 Ga]Ga-FAPI PET/CT had signi cantly higher tracer uptake and tumor-to-background contrast than those of [ 18 F]-FDG PET/CT. Therefore, [ 68 Ga]Ga-FAPI presented as a promising alternative to [ 18 F]-FDG.
The invasion depth in primary gastric cancer is an important factor on determining the prognosis. However, previous studies pointed out that the SUV max of [ 18 F]-FDG was not correlated with the degree of penetration [29]. Our study indicated that SUV max -FAPI and TBR-FAPI both showed positive correlations with primary tumor depth and the TBR-FAPI of T3-4 and III-IV groups were signi cantly higher than that of T1-2 and I-II groups, which was consistent with the study reported by Jiang et al [21]. As such, our study showed that [ 68 Ga]Ga-FAPI PET/CT could better evaluate the invasion extent of gastric cancer. Additionally, we also found that TBR-FAPI at N0 stage was signi cantly higher than that at N1-3, suggesting that TBR of [ 68 Ga]Ga-FAPI may indicate the possibility of lymph node metastases.
LN staging is crucial in the treatment and prognosis of gastric cancer patients [30,31] [39]. Our study showed that high FAP expression was not correlated with the increased LN metastases, possibly due to the limited sample size. Owing to the unavailability of long-term follow-up, no conclusion on overall survival could be drawn, which will be investigated in our further research.
Several limitations of this study also need to be mentioned. First, although 56 patients with gastric cancer enrolled in this study was the largest number among the resemble published studies, but sample sizes still need to be increased to strengthen our interpretation and our conclusion. Second, there were too few other parenchymal metastases such as ovaries, adrenal glands, lung, and erector spinae in this cohort to conclude on the detection of those lesions. Third, most patients (39/56) in this cohort did not undergo surgery or biopsy for metastatic lesions to con rm the positive lesions, but the follow-up and other imaging data (such as MRI, CT) were available as a reference.
Lastly, the small sample size (17/56) included in the analysis of lymph nodes, which had postoperative pathologic ndings as a reference.     A 83-year-old man with gastric adenocarcinoma underwent PET/CT scans for initial staging.

Figure 5
A 64-year-old man with a moderately differentiated gastric adenocarcinoma was con rmed by postoperative pathology.
[68Ga]Ga-FAPI PET/CT showed high tracer uptake (SUVmax = 11.25) in the primary tumour (a-c, solid arrow). The histological work-up (d, e) including immunohistochemistry for FAP con rmed FAP overexpressed in primary tumour and scored 3. The other patient presented with mucinous adenocarcinoma, which showed mild-moderate uptake (SUVmax = 4.65) in [68Ga]Ga-FAPI PET/CT (f-h, dotted arrow). The pathological results (i, j) showed slight FAP expression in stromal cells and scored 1.