Assessing the Value of ⁶⁸Ga-FAPI PET/CT in Gastric Mucinous Adenocarcinoma or Signet Ring Cell Carcinoma

Abstract

Purpose

To investigate the clinical impact and prognostic value of gallium 68 (⁶⁸Ga)–labeled fibroblast activation protein inhibitor (FAPI) PET/CT in gastric mucinous adenocarcinoma (MAC) and signet ring cell carcinoma (SRCC).

Materials and Methods

Eighty-six participants with newly diagnosed or recurrent gastric MAC or SRCC were prospectively enrolled from April 2021 to October 2021 and underwent both fluorine 18 (¹⁸F) fluorodeoxyglucose (FDG) PET/CT and ⁶⁸Ga-FAPI PET/CT. The sensitivity, specificity, and accuracy of the two scans in primary and metastatic tumors were evaluated using the McNemar test. Changes of treatment strategies were recorded to compare the treatment management value of the two PET/CT scans. The maximum standardized uptake value (SUV_max) and peritoneal cancer index (PCI) were recorded for survival analysis. Progression-free survival (PFS) was defined as the time interval from the date of PET/CT scans to the date of disease progression.

Results

Eighty-six participants (median age, 62 years [IQR, 45–78 years]; 49 female) were evaluated. ⁶⁸Ga-FAPI PET/CT showed higher diagnostic accuracy in detecting involved lymph nodes (87% [212 of 244] vs 71% [173 of 244], P < .001) and peritoneal metastases (96% [70 of 73] vs 55% [40 of 73], P < .001) than ¹⁸F-FDG PET/CT. Twenty-six participants (30% [26 of 86]) had treatment changes due to more accurate diagnosis with ⁶⁸Ga-FAPI PET/CT. Additionally, the ⁶⁸Ga-FAPI PCI was an independent predictor for PFS (hazard ratio, 6.9; 95% CI: 2.1, 23.1; P = .002).

Conclusion

⁶⁸Ga-FAPI PET/CT had higher accuracy in diagnosis of gastric MAC/SRCC compared with ¹⁸F-FDG PET/CT and demonstrated the potential to improve treatment strategies and predict prognosis.

Keywords: PET/CT, Mucinous Adenocarcinoma, Signet Ring Cell Carcinoma, Oncology, Abdomen/GI, Molecular Imaging

Supplemental material is available for this article.

© RSNA, 2024

Summary

⁶⁸Ga-labeled fibroblast activation protein inhibitor PET/CT showed higher diagnostic accuracy compared with ¹⁸F fluorodeoxyglucose PET/CT in individuals with gastric mucinous adenocarcinoma or signet ring cell carcinoma and demonstrated the potential to improve treatment strategies and predict prognosis.

Key Points

• ■ In this prospective study of 86 participants with gastric mucinous adenocarcinoma or signet ring cell carcinoma, ⁶⁸Ga-labeled fibroblast activation protein inhibitor (FAPI) PET/CT had higher sensitivity in detecting primary tumors (100% vs 46%, P < .001) and peritoneal metastases (100% vs 62%, P < .001) compared with ¹⁸F fluorodeoxyglucose PET/CT.

• ■ Treatment strategies were changed in 26 participants (30% [26 of 86]) after use of ⁶⁸Ga-FAPI PET/CT.

• ■ The ⁶⁸Ga-FAPI PET/CT–derived peritoneal cancer index was an independent predictor of progression-free survival (hazard ratio, 6.9; P = .002).

Introduction

Mucinous adenocarcinoma (MAC) and signet ring cell carcinoma (SRCC) are poorly differentiated cancers of the stomach, and account for 2.4% and 16.8% of patients with gastric cancer (1,2), respectively. These cancers are diagnosed with endoscopic biopsy and staged using CT, MRI, PET/CT, or laparoscopy. Currently, the most common treatment strategies for gastric cancer are surgery, medical therapy, and radiation therapy. Accurate diagnosis is the prerequisite for an appropriate treatment regimen.

CT and MRI are the most commonly used methods for tumor detection, but the two imaging modalities have low sensitivity in detecting peritoneal metastases, especially in small peritoneal seeding lesions (3). Fluorine 18 (¹⁸F) fluorodeoxyglucose (FDG) PET/CT is widely used in tumor staging and recurrence detection, but it is not sensitive enough for detecting MAC/SRCC because of the low cellularity and the presence of mucin in the two types of tumors (4–7).

Fibroblast activation protein (FAP) is highly expressed in cancer-associated fibroblasts (CAFs), which contribute to 90% of the tumor stroma. Gallium 68 (⁶⁸Ga)–labeled FAP inhibitor (FAPI) can specifically target and bind to the FAP. Gastric cancer has abundant CAFs and shows intermediate ⁶⁸Ga-FAPI uptake (8,9). Two case reports of gastric SRCC showed a greater number of metastatic lesions in ⁶⁸Ga-FAPI PET/CT than in ¹⁸F-FDG PET/CT (10,11). Chen et al (12) demonstrated that ⁶⁸Ga-FAPI PET/CT upgraded the clinical TNM stage in three patients with gastric cancer (8.8%) compared with ¹⁸F-FDG–based TNM staging. These findings suggest that ⁶⁸Ga-FAPI PET/CT could be considered an effective imaging method in diagnosing gastric cancer. However, the number of patients with MAC or SRCC enrolled in the above studies on gastric cancer was limited, and the value of ⁶⁸Ga-FAPI PET/CT in modifying clinical management and predicting prognosis has not previously been well established.

We hypothesized that ⁶⁸Ga-FAPI would be used as a method in treatment management in patients with gastric MAC/SRCC. The aim of this study was to compare the diagnostic performance, value in determining appropriate oncologic management, and prognostic value of ⁶⁸Ga-FAPI and ¹⁸F-FDG PET/CT in individuals with gastric MAC/SRCC.

Materials and Methods

Study Participants

This single-center prospective study was approved by the Clinical Research Ethics Committee of our hospital (ID 2012229–2). All participants provided written informed consent. The sample size was calculated according to our preliminary data and a previous study (13). Two correlate proportions (McNemar test) in PASS (version 15.0; NCSS) software was used. With a power of 0.8, a type I error of 0.05, a difference of 0.2, a discordant proportion of 0.4, and a dropout of 10%, a sample size of 85 was required. Participants were consecutively enrolled from our hospital between April 2021 and October 2021. They underwent PET/CT for initial assessment or recurrent detection. The participant inclusion criteria were as follows: (a) had newly diagnosed or previously treated gastric MAC/SRCC or adenocarcinoma with mucinous or signet ring cell component (according to the 2019 World Health Organization classification, MAC is defined as a tumor with >50% of the tumor area covered by extracellular mucin, whereas SRCC is composed of >50% signet ring cells, which contain prominent intracellular mucus secretion and push the nucleus to the cell periphery; adenocarcinoma with mucinous or signet ring cell component refers to carcinomas with mucinous areas or signet ring cells in <50% of the tumor) (14); (b) underwent PET/CT for initial assessment to decide the optimal treatment strategy or for detecting tumor recurrence and metastases; and (c) signed informed consent. Exclusion criteria were (a) pregnant participants, (b) participants with another malignant disease, and (c) participants who refused to provide informed consent.

Treatment plans were decided by oncologists (with >25 years of experience in gastrointestinal oncology) or the multidisciplinary team of gastric cancer, including three senior gastric surgeons, three senior oncologists, two senior radiation oncologists, two senior radiologists, two senior nuclear medicine physicians, and two senior pathologists at our hospital. The final diagnosis was confirmed with pathologic assessment (biopsy or surgery), comprehensive consideration of imaging and laboratory test findings, and radiographic follow-up. The follow-up contrast-enhanced CT or MRI was performed 6 months after treatment. Lesions in the follow-up images were considered malignant based on the following criteria: (a) the morphology, growth pattern, and enhancement pattern of the lesions were in line with the characteristics of malignant tumors and (b) tumor progression or shrinkage after treatment was visible at subsequent follow-up imaging.

Radiopharmaceuticals and PET/CT Imaging Protocol

The synthesis of ¹⁸F-FDG was based on a modified Siemens Explora FDG₄ module with cyclotron (Siemens CTI RDS Eclipse ST; Siemens Healthineers). ⁶⁸Ga-FAPI (Jiangsu Huayi Technology) preparation was performed according to the method reported by Gu et al (15). Briefly, 40 μg DOTA-FAPI-04 was dissolved in 0.6 mL NaAc solution, then an ⁶⁸GaCl₃ eluent was added. The reaction mixture was heated to 95 °C for 10 minutes, and then the product was isolated by solid-phase extraction. Radiochemical purity was greater than 95%.

The paired ¹⁸F-FDG and ⁶⁸Ga-FAPI PET/CT were performed within 7 days. The image data were obtained using a PET/CT system (Biograph mCT Flow; Siemens Healthineers) 1 hour after ¹⁸F-FDG (0.1 mCi/kg) or ⁶⁸Ga-FAPI (0.05 mCi/kg) injection, and images were reconstructed iteratively using an ordered-subset expectation maximization iterative reconstruction by applying CT data for attenuation correction (15).

Image Analysis

Both ¹⁸F-FDG and ⁶⁸Ga-FAPI PET/CT images were interpreted in random order by two experienced nuclear medicine physicians independently. In case of inconsistency, consensus was reached after a comprehensive discussion. All ¹⁸F-FDG PET/CT images were reviewed by Y.S. and J.X. (with 8 and 21 years of experience in nuclear oncology, respectively). All ⁶⁸Ga-FAPI PET/CT images were reviewed by Y.Q. and S.S. (with >15 and >30 years of experience in nuclear oncology, respectively). Reviewers were blinded to clinical data and information from the other PET/CT images. Positive ¹⁸F-FDG and ⁶⁸Ga-FAPI uptake was defined as focal tracer uptake that was greater than the background of normal adjacent tissue. For semiquantitative analysis, the maximum standardized uptake value (SUV_max) normalized according to body weight was measured using a multimodality computer platform (Syngo; Siemens Healthineers). If one participant had multiple positive lesions in one organ, the SUV_max was defined as the average SUV of all lesions (if ≤five lesions) or the five highest uptake lesions (if >five lesions) in each organ/region. To further increase comparability, the tumor to background ratio (TBR) was calculated according to the following formulas: TBR = SUV_max of primary tumor/SUV_max of normal adjacent stomach wall (for primary tumor), or TBR = SUV_max of tumor lesions/SUV_max of normal liver tissue (for metastases).

Participants who were diagnosed with peritoneal metastasis and received therapy (medical therapy or radiation therapy) were included for prognostic analysis. Progression-free survival (PFS) was defined as the time interval from the date of PET/CT examination to the date of disease progression. The follow-up of participants ended at the time of disease progression. Participants without disease progression were followed until May 31, 2023. To analyze survival outcomes in participants with peritoneal metastasis, we calculated the PET/CT-derived peritoneal cancer index (PCI) according to the method reported by Morland et al (16). The PET/CT PCI score was determined according to tumor size and distribution throughout Sugarbaker’s 13 regions; each region was given a lesion size score depending on the size of the largest lesion. The score ranges from 0 to 3 (0, no implant; 1, implants <0.5 cm; 2, implants 0.5–5 cm; 3, implants >5 cm or a confluence of implants). At PET/CT, tumor size estimation was performed on the CT component when the lesion was visible. However, if the lesion had an increased focus of uptake and a miliary pattern, and thus could not be clearly observed on coupled CT images, it was considered a score of 1 (Fig S1).

Statistical Analysis

Data were analyzed using SPSS (version 25.0; IBM) statistical software. The McNemar test was used to compare differences in sensitivity, specificity, and accuracy between ¹⁸F-FDG PET/CT and ⁶⁸Ga-FAPI PET/CT. The 95% CIs were calculated using Prism version 8 (GraphPad; https://www.graphpad.com/quickcalcs/mcNemar1/). The Wilcoxon signed rank test was used to compare the difference between ¹⁸F-FDG and ⁶⁸Ga-FAPI SUV_max and between ¹⁸F-FDG and ⁶⁸Ga-FAPI TBR. Receiver operating characteristic analysis was used to estimate the best cutoff values for PET/CT parameters for prediction with the maximum Youden index. The Kaplan-Meier method was used to estimate survival curves, which were compared using the log-rank test. Cox regression analysis was used for PFS, and P < .05 was considered indicative of a statistically significant difference.

Results

Participant Characteristics

A total of 86 participants (49 female, 37 male; median age, 62 years [IQR, 45–78 years]) with gastric MAC/SRCC or gastric adenocarcinoma with a mucinous or signet ring cell component were included in the study (Fig 1). The clinical characteristics of the 86 participants are summarized in Table 1. Twenty-four participants without treatment underwent paired ¹⁸F-FDG and ⁶⁸Ga-FAPI PET/CT for initial assessment, and 62 participants for recurrence detection. Primary tumors and anastomotic recurrence were confirmed with histopathologic assessment, and metastatic lesions were confirmed with histopathologic assessment or follow-up (Table S1).

Figure 1:

Flow diagram shows study design. ¹⁸F-FDG = ¹⁸F fluorodeoxyglucose, ⁶⁸Ga-FAPI = ⁶⁸Ga-labeled fibroblast activation protein inhibitor.

Table 1:

Participant Characteristics

A total of 55 participants displayed discordance in ⁶⁸Ga-FAPI and ¹⁸F-FDG PET/CT images, including 11 primary tumors, three esophagojejunal anastomotic recurrences, 32 peritoneal/lymph node metastases, four inflammatory lesions, and five others (Table S2). A total of 26 participants had treatment changes due to discrepancies (Fig 2).

Figure 2:

Flow diagram shows an overview of therapeutic changes due to discordant images based on ⁶⁸Ga-labeled fibroblast activation protein inhibitor (⁶⁸Ga-FAPI) and ¹⁸F fluorodeoxyglucose (¹⁸F-FDG) PET/CT. GC = gastric cancer, LM = lymph node metastases, PM = peritoneal metastases.

Comparison of ⁶⁸Ga-FAPI and ¹⁸F-FDG Uptake

⁶⁸Ga-FAPI imaging showed higher SUV_max values than ¹⁸F-FDG imaging in primary tumors (median, 9.3 [IQR, 6.7–11.5] vs 3.1 [IQR, 2.5–6.2]; P = .04) and metastatic peritoneum (median, 6.9 [IQR, 5.2–8.6] vs 3.3 [IQR, 1.6–5.3]; P < .001), but the SUV_max values were similar in metastatic lymph nodes (median, 6.1 [IQR, 4.4–8.2] vs 4.3 [IQR, 3.1–5.9]; P = .09) and visceral metastases (median, 6.2 [IQR, 5.1–7.7] vs 4.3 [IQR, 3.2–6.0]; P = .60). However, both primary and metastatic tumors showed higher TBR values in ⁶⁸Ga-FAPI than in ¹⁸F-FDG PET/CT (Table S3).

Diagnostic Performance of ⁶⁸Ga-FAPI and ¹⁸F-FDG PET/CT

Regarding the performance of the two imaging modalities in the diagnosis of primary tumors in the 24 treatment-naive participants, the detection rates were 100% (24 of 24) for ⁶⁸Ga-FAPI PET/CT and 46% (11 of 24) for ¹⁸F-FDG PET/CT (P < .001) (Fig S2). Regarding the 62 participants with postgastrectomy gastric cancer, the sensitivities for ⁶⁸Ga-FAPI and ¹⁸F-FDG PET/CT in the detection of esophagojejunal anastomotic recurrence were 100% (11 of 11) and 64% (seven of 11), respectively. Conversely, the false-positive rates for ⁶⁸Ga-FAPI and ¹⁸F-FDG were 0% (0 of 51) and 18% (nine of 51), respectively.

For lymph node detection, 244 lymph nodes in 51 participants were detected on the two PET images and the CT component. The status of these lymph nodes was confirmed with pathologic examination (n = 213) and follow-up (n = 31). Of these, 173 nodes in 35 participants were confirmed as metastases. Regarding the 244 lymph nodes detected on PET/CT images, the sensitivity and accuracy of ⁶⁸Ga-FAPI PET/CT was higher than that of ¹⁸F-FDG PET/CT (90% [156 of 173] vs 72% [124 of 173]; 87% [212 of 244] vs 71% [173 of 244], respectively) (Table 2). Regarding the 213 lymph nodes with pathologic findings, the sensitivity and accuracy of ⁶⁸Ga-FAPI PET/CT were also higher than that of ¹⁸F-FDG PET/CT (Table 2).

Table 2:

Comparison of ⁶⁸Ga-FAPI and ¹⁸F-FDG PET/CT in the Evaluation of All Lymph Nodes and Lymph Nodes with Pathologic Findings

For visceral detection, peritoneal lesions were detected in 52 participants with ⁶⁸Ga-FAPI PET/CT and 32 participants with ¹⁸F-FDG PET/CT. At ¹⁸F-FDG PET/CT, peritoneal metastases were missed in 20 participants who had focal or miliary nodular lesions less than 5 mm in diameter (Fig 3). Among these 20 patients, three had serosal intestinal wall implants that were detected with ⁶⁸Ga-FAPI PET/CT but not with ¹⁸F-FDG PET/CT. Table 3 shows the sensitivity and accuracy of the two PET/CT examinations in detecting lesions. For peritoneal metastasis detection, the sensitivity and accuracy of ⁶⁸Ga-FAPI PET/CT were higher compared with ¹⁸F-FDG PET/CT (100% [52 of 52] vs 62% [32 of 52], P < .001; 96% [70 of 73] vs 55% [40 of 73], P < .01, respectively; McNemar test). For other visceral metastasis detection, the sensitivity and accuracy of ⁶⁸Ga-FAPI PET/CT were similar to that of ¹⁸F-FDG PET/CT. ⁶⁸Ga-FAPI PET/CT showed false-positive uptake in abdominal wall incisions and the omentum in three participants.

Figure 3:

Representative PET/CT images in diffuse miliary peritoneal carcinomatosis and serosal colonic implants. A 68-year-old female participant who had a postoperative histopathologically diagnosed gastric mucinous adenocarcinoma 3 years prior was suspected to have relapse because of abdominal pain and an elevated carcinoembryonic antigen level (25.3 ng/mL). ¹⁸F-FDG and ⁶⁸Ga-FAPI PET/CT were recommended for further diagnosis. (A) ¹⁸F-FDG PET/CT images show no suspicious lesions. (B) ⁶⁸Ga-FAPI PET/CT images show diffused and remarkable uptake of ⁶⁸Ga-FAPI by miliary peritoneal carcinomatosis (maximum standardized uptake value, 7.3; arrows). The participant underwent exploratory laparotomy, and miliary peritoneal metastasis was found intraoperatively. ¹⁸F-FDG = ¹⁸F fluorodeoxyglucose, ⁶⁸Ga-FAPI = ⁶⁸Ga-labeled fibroblast activation protein inhibitor.

Table 3:

Comparison of ⁶⁸Ga-FAPI and ¹⁸F-FDG PET/CT in the Evaluation of All Visceral Metastases

Treatment Management Value of ⁶⁸Ga-FAPI and ¹⁸F-FDG PET/CT

Regarding the participants referred for initial assessment, all 24 had FAPI-positive primary tumors, but only 11 were FDG-positive. For N staging, one tumor was upstaged by ⁶⁸Ga-FAPI PET/CT because more regional metastatic lymph nodes were detected. For M staging, disease in two participants was upstaged because peritoneal and distant lymph nodes metastases (confirmed at laparoscopy) were detected at ⁶⁸Ga-FAPI PET/CT, and disease was downstaged in two participants because one left supraclavicular lymphadenitis (confirmed at biopsy) and one mesenteric panniculitis (confirmed at laparoscopy) showed true-negative uptake of ⁶⁸Ga-FAPI but false-positive uptake of ¹⁸F-FDG. Finally, four of 24 participants had changes in treatment strategies, including two from radical surgery to medical therapy and two from chemoradiotherapy to radical surgery. A typical case is presented in Figure 4.

Figure 4:

The impact of ⁶⁸Ga-FAPI PET/CT on treatment in newly diagnosed gastric signet ring cell carcinoma (SRCC). A 56-year-old male participant with newly diagnosed gastric SRCC underwent PET/CT imaging for preoperative assessment. (A) ¹⁸F-FDG PET/CT images show a thickened gastric wall (arrow) with moderate ¹⁸F-FDG uptake (maximum standardized uptake value, 3.8; tumor to background ratio, 1.9) and no regional or distant metastases. Radical surgery was suggested according to the ¹⁸F-FDG PET/CT images. (B) ⁶⁸Ga-FAPI PET/CT images show a thickened gastric wall (white arrow), enlarged lymph nodes in the perigastric and hepatic hilar region (green arrows), and multiple peritoneal nodules in the pelvis (yellow arrows), all of which showed remarkable FAPI uptake; therefore, treatment was changed to chemotherapy. All the lesions were confirmed as malignant at exploratory laparotomy after the ⁶⁸Ga-FAPI PET/CT examination. ¹⁸F-FDG = ¹⁸F fluorodeoxyglucose, ⁶⁸Ga-FAPI = ⁶⁸Ga-labeled fibroblast activation protein inhibitor.

Regarding the 62 participants referred for recurrence detection, two participants had metastatic lymph nodes, 15 had peritoneal metastases, and two had both lymph nodes and peritoneal involvement that were detected with ⁶⁸Ga-FAPI PET/CT, but not with ¹⁸F-FDG PET/CT. These 19 participants received medical therapy (n = 15, multiple metastases) and radiation therapy (n = 4, single metastatic lesion) instead of follow-up; a representative case is shown in Figure 5. In addition, two participants had mesenteric panniculitis (confirmed at laparoscopy), and one participant had inflammatory lymph nodes (>1 cm) in the mediastinum (confirmed with endobronchial US-guided transbronchial needle aspiration). These lesions all showed true-negative uptake of ⁶⁸Ga-FAPI and false-positive uptake of ¹⁸F-FDG. These three participants were finally recommended for follow-up instead of therapy.

Figure 5:

The impact of ⁶⁸Ga-FAPI PET/CT on treatment in recurrent gastric signet ring cell carcinoma (SRCC). A 45-year-old male participant underwent surgery 2 years prior because of gastric SRCC and had anal distention over the past 6 months. (A) ¹⁸F-FDG PET/CT images show a thickened rectal wall (arrow) with negative ¹⁸F-FDG uptake and no other metastasis. Close monitoring was recommended according to the ¹⁸F-FDG PET/CT images. (B) ⁶⁸Ga-FAPI PET/CT images show a thickened rectal wall with intense ⁶⁸Ga-FAPI uptake (maximum standardized uptake value, 9.8; tumor to background ratio; 14.0; arrows); therefore, treatment was changed to radiation therapy. Metastasis in the rectal wall was confirmed at needle biopsy after PET/CT examination. ¹⁸F-FDG = ¹⁸F fluorodeoxyglucose, ⁶⁸Ga-FAPI = ⁶⁸Ga-labeled fibroblast activation protein inhibitor.

Prognostic Value of ⁶⁸Ga-FAPI and ¹⁸F-FDG PET/CT

A total of 52 participants had peritoneal metastasis, and only 49 of 52 received treatment after PET/CT; as a result, these 49 participants were included in the survival analysis. The median PFS was 9.4 months (IQR, 5.2–12.1 months) and the overall disease-progression rate was 67.4%. Areas under the receiver operating characteristic curve for ⁶⁸Ga-FAPI SUV_max, ¹⁸F-FDG SUV_max, ⁶⁸Ga-FAPI PCI, and ¹⁸F-FDG PCI in predicting PFS were 0.697, 0.669, 0.849, and 0.661, respectively (Table S4). Participants with higher ¹⁸F-FDG SUV_max demonstrated a significantly shorter PFS (median PFS, 6.4 months [IQR, 4.2–6.8 months] vs 9.8 months [IQR, 6.1–12.8 months]; P = .03); and participants with higher ⁶⁸Ga-FAPI PCI also demonstrated a significantly shorter PFS (median PFS, 6.4 months [IQR, 2.5–9.8 months] vs median PFS was not reached; P < .001) (Fig 6). In the univariable analysis, higher ¹⁸F-FDG SUV_max and higher ⁶⁸Ga-FAPI PCI were associated with poor PFS (P = .03 and P < .001, respectively). In the multivariable analysis, a higher ⁶⁸Ga-FAPI PCI was an independent predictor for short PFS (hazard ratio, 6.9; 95% CI: 2.1, 23.1; P = .002) (Table 4).

Figure 6:

Kaplan-Meier curves show prediction of progression-free survival (PFS) according to (A) ¹⁸F-FDG maximum standardized uptake value (SUV_max), (B) ⁶⁸Ga-FAPI SUV_max, (C) ¹⁸F-FDG peritoneal cancer index (PCI), and (D) ⁶⁸Ga-FAPI PCI. Participants with higher ¹⁸F-FDG SUV_max demonstrated a significantly shorter PFS (median PFS, 6.4 months [IQR, 4.2–6.8 months] vs 9.8 months [IQR, 6.1–12.8 months]; P = .03); and participants with higher ⁶⁸Ga-FAPI-PCI also demonstrated a significantly shorter PFS (median PFS, 6.4 months [IQR, 2.5–9.8 months] vs median PFS was not reached; P < .001). ¹⁸F-FDG = ¹⁸F fluorodeoxyglucose, ⁶⁸Ga-FAPI = ⁶⁸Ga-labeled fibroblast activation protein inhibitor.

Table 4:

Cox Proportional Hazards Regression Analysis for PFS

Discussion

⁶⁸Ga-FAPI PET/CT has demonstrated a promising role in the detection of tumors in the digestive system, including gastric cancer, but its clinical value in modifying treatment strategy and predicting prognosis remains unclear. In this study, we investigated the value of ⁶⁸Ga-FAPI PET/CT in diagnosis, oncologic management, and prognosis prediction in participants with gastric MAC/SRCC. The results demonstrated that ⁶⁸Ga-FAPI PET/CT had higher accuracy than ¹⁸F-FDG PET/CT for detecting tumors secondary to gastric MAC/SRCC. Furthermore, ⁶⁸Ga-FAPI PET/CT improved treatment strategies in 26 participants (30% [26 of 86]), and PCI derived from ⁶⁸Ga-FAPI PET/CT was an independent predictor of PFS (hazard ratio, 6.9; 95% CI: 2.1, 23.1; P = .002).

Previous studies demonstrated that ⁶⁸Ga-FAPI PET/CT helped detect more lesions than ¹⁸F-FDG PET/CT (17–19), and the minimum diameter of the lesion with ⁶⁸Ga-FAPI uptake was 2 mm (20). Our study demonstrated that more lesions were detected with ⁶⁸Ga-FAPI PET/CT, especially for peritoneal metastasis with a small size. Peritoneal metastasis is the most common metastasis mode of gastric MAC/SRCC. In the current study, 60% of participants had peritoneal metastasis, falling in the range of 40%–80% reported in previous studies (21,22). ¹⁸F-FDG PET/CT sometimes has low sensitivity and accuracy in detecting peritoneal metastasis, especially in small lesions, because the sensitivity of ¹⁸F-FDG PET/CT is influenced by the number of tumor cells with high glycolytic activity. A small lesion that has tumor cells with high glycolytic metabolism less than 10⁷ cannot always be detected with ¹⁸F-FDG (23). In addition, physiologic uptake of ¹⁸F-FDG in the intestinal cavity and diaphragm and/or liver capsule may lead to reducing the diagnostic accuracy of ¹⁸F-FDG PET/CT. Several studies demonstrated that ⁶⁸Ga-FAPI had higher accuracy in diagnosing peritoneal metastasis than ¹⁸F-FDG PET/CT due to no physiologic uptake of ⁶⁸Ga-FAPI in the intestinal wall, resulting in low rates of nonspecific uptake in the peritoneum (24,25). Our results showed that small nodular peritoneal lesions (<5 mm in diameter), including serosal intestinal wall implants, were missed at ¹⁸F-FDG PET/CT because of the small size and physiologic uptake of ¹⁸F-FDG in the intestinal wall, consistent with previous studies (24,25). This result suggests that ⁶⁸Ga-FAPI PET/CT offers an advantage over ¹⁸F-FDG PET/CT in diagnosing peritoneal metastasis. However, in our study, three participants had false-positive ⁶⁸Ga-FAPI uptake in the omentum due to changes in surgery. Previous studies indicated wound repair, arthritis, and some other benign lesions also have false-positive uptake of ⁶⁸Ga-FAPI (26). Thus, ⁶⁸Ga-FAPI PET/CT image interpretation should be made cautiously, requiring information from clinical examination and medical history. Moreover, ⁶⁸Ga-FAPI PET/CT had higher sensitivity and accuracy than ¹⁸F-FDG PET/CT in evaluating lymph nodes.

Better diagnostic accuracy leads to more appropriate therapeutic approaches. Recently, studies showed that ⁶⁸Ga-FAPI PET/CT can improve treatment strategies in pancreatic and ovarian cancer (27,28) due to higher sensitivity than ¹⁸F-FDG PET/CT. Similarly, our results also showed that ⁶⁸Ga-FAPI PET/CT had higher accuracy in detecting gastric MAC/SRCC, leading to changes in treatment approaches in 30% of participants (26 of 86). Additionally, ⁶⁸Ga-FAPI PET/CT might have the potential to differentiate patients who are suitable for regional therapy such as radiation therapy. In our study, four participants who had single metastasis detected at ⁶⁸Ga-FAPI PET/CT received radiation therapy. This suggested that ⁶⁸Ga-FAPI PET/CT might be applied as an alternative to ¹⁸F-FDG PET/CT in managing participants with gastric MAC/SRCC.

Additionally, the extent of peritoneal involvement is a survival factor for individuals with gastric cancer. The PCI, which was used to assess the extent of peritoneal metastasis in our study, is considered to be a common prognostication tool. A previous study reported that patients with a PCI score less than 15 had prolonged overall survival compared with those with a PCI score greater than 15 (29). However, the traditional PCI is obtained at surgery, which is an invasive technique, and not all patients have access to surgery. A previous study indicated that ¹⁸F L-dihydroxyphenylalanine (FDOPA) PET/CT had the potential to predict the surgical PCI in small intestine neuroendocrine tumors (16). According to this finding, PCI derived from PET/CT may be a useful tool to predict survival outcomes in patients with peritoneal metastasis. In the current study, our preliminary finding showed that the PCI derived from ⁶⁸Ga-FAPI PET/CT, not ¹⁸F-FDG PET/CT, was a prognostic factor for gastric MAC/SRCC. It was previously reported that ¹⁸F-FDG PET/CT underestimates the extent of peritoneal metastasis during surgery in ovarian cancer (30) and colorectal cancer (31), and this may be the reason that ¹⁸F-FDG PET/CT PCI was not able to predict prognosis in the current study.

Our study had several limitations. First, the study sample was relatively heterogeneous (gastric MAC or SRCC, or gastric adenocarcinoma with a mucinous or signet ring cell component). Second, tissue sampling was difficult due to tumor location or invasiveness of approach, and thus not all the lesions were confirmed with pathologic examination; however, the final result was reached with imaging and follow-up. Finally, participants included in the prognostic analysis received different therapy according to their individual conditions and the extent of involvement. But all the treatment strategies were decided by oncologists or after discussions in a multidisciplinary tumor board setting. A multicenter study is needed to validate this promising imaging technique.

In conclusion, the study results demonstrated that ⁶⁸Ga-FAPI PET/CT had higher sensitivity and accuracy compared with ¹⁸F-FDG PET/CT in detecting malignant lesions in gastric MAC/SRCC. As a result, it had a significant impact on tumor management and survival prediction. Further studies with larger sample sizes are needed to validate these findings.