Comparison of ¹⁸F-FDG and ⁶⁸Ga-FAPI-04 Uptake in Postoperative Re-evaluation of Gastric, Duodenal, and Colorectal Cancers

Abstract

Purpose 

To compare the potential efficacy of ⁶⁸Ga-FAPI-04 PET/CT with that of ¹⁸F-FDG PET/CT for detecting tumor recurrence and nodal and distant metastases in gastric, duodenal, and colorectal cancers.

Methods 

This single-center retrospective clinical study was performed at the Affiliated Hospital of Southwest Medical University between January 2020 and June 2022. Participants with gastric, duodenal, and colorectal cancers after curative resection underwent both ⁶⁸Ga-FAPI-04 and ¹⁸F-FDG PET/CT. Histopathologic examination, morphologic imaging, and/or follow-up imaging served as the reference standards. The SUV_max of the tumor recurrence and nodal and distant metastases between ¹⁸F-FDG and ⁶⁸Ga-FAPI-04 PET/CT were compared using the paired-sample t test.

Results 

Forty-one participants with gastric, duodenal, and colorectal cancers were enrolled in the study (median age, 51 years; range, 19–75 years). The sensitivity of ⁶⁸Ga-FAPI-04 PET/CT was higher than that of ¹⁸F-FDG PET/CT for detecting tumor recurrence (6 of 6 [100%] vs 4 of 6 [67%]), nodal metastases (92 of 92 [100%] vs 31 of 92 [34%]), and distant metastases (28 of 30 [93%] vs 20 of 30 [67%]).

Conclusion 

⁶⁸Ga-FAPI-04 PET/CT for tumor re-evaluation showed potential for more accurate performance of gastric, duodenal, and colorectal cancers, thereby improving treatment decision-making.

Gastric cancer and colorectal cancer, common types of gastrointestinal cancer, comprised 10.6% (ranked third) and 6.1% (ranked fifth) of all cancers diagnosed in 2018.^1,2 Complete surgical resection of gastrointestinal cancer is considered a potential cure, but its long-term survival is often reported to be poor. The detection of tumor recurrence and metastasis is important for the management of patients. ¹⁸F-FDG PET/CT is a widely accepted tumor imaging method with important implications in various cancers.^3,4 However, ¹⁸F-FDG uptake multiplicity has been noted in PET studies of gastrointestinal patients, especially lower uptake in some specific histological subtypes, such as adenocarcinoma and signet ring cell carcinoma.^5,6 Furthermore, the diagnostic effect of ¹⁸F-FDG is limited, as the variable physiological uptake of ¹⁸F-FDG in the gastrointestinal tract may interfere with the detection of foci and require intensive monitoring from time to time.⁷ Therefore, the development of an effective PET tracer is expected to improve the characterization of gastrointestinal tumors and facilitate the precise diagnosis of patients.

⁶⁸Ga-FAPI has been developed and used for targeting fibroblast-activation protein and visualization of tumor stroma, which showed a higher sensitivity than ¹⁸F-FDG in the detection of primary and metastatic lesions of various types of cancers.^8,9 However, the role of ⁶⁸Ga-FAPI PET/CT in patients with gastric, duodenal, and colorectal cancers after operation is not clear. In this study, we aimed to analyze the value of ⁶⁸Ga-FAPI-04 PET/CT in the follow-up of postoperative patients with gastrointestinal cancer.

PATIENTS AND METHODS

Patients

This retrospective study was conducted at the Affiliated Hospital of Southwest Medical University from January 2020 to June 2022. This study was approved by the ethics committee of our hospital (AHSWMU-2020-035), and informed consent was obtained from each patient. Inclusion criteria were biopsy-proven recurrent gastric, duodenal, and colorectal cancers or suspected disease recurrence, nodal and distant metastases based on other imaging tests, tumor markers, or clinical symptoms or follow-up. Patients were excluded when they had received chemotherapy or radiotherapy within 4 weeks before the PET/CT scan.

Preparation of ¹⁸F-FDG and ⁶⁸Ga-FAPI-04

¹⁸F-FDG was manufactured in accordance with the standard method using the coincidence ¹⁸F-FDG synthesis module (FDG-N; PET Science & Technology, China). We purchased the precursor FAPI-04 from MCE (MedChemExpress, USA). The ⁶⁸Ga-FAPI-04 labeling was carried out according to the method described previously.^10,11 The radiochemical purity of ⁶⁸Ga-FAPI-04 and ¹⁸F-FDG exceeded 95%. The final product is sterile and meets all of our institution's required standards before use.

PET/CT Imaging

For ¹⁸F-FDG PET/CT image acquisition, patients fasted for at least 6 hours, and the plasma glucose level is lower than 11 mmol/L.¹² After intravenous injection of ¹⁸F-FDG (3.7 MBq/kg), the patient rested in a quiet place. The patients were instructed to drink 500 mL of water to stimulate ¹⁸F-FDG excretion in the renal calyx and to urinate before scanning. Data were acquired using a PET/CT scanner (uMI780; United Imaging, China) 45 to 60 minutes after intravenous administration. First, the CT image acquisition was performed from the head to the upper thigh segment (tube current of 120 mA, tube voltage of 120 kV, and slice thickness of 3.00 mm). PET was then performed at the same bed position as the CT scan, 5 to 6 bed positions, and 3D acquisition mode at 1.5 min/position. ⁶⁸Ga-FAPI-04 PET/CT was performed within 7 days of ¹⁸F-FDG PET/CT for comparison, and the dose of ⁶⁸Ga-FAPI-04 injected was calculated based on patient body weight (1.85 MBq/kg). Other parameters of CT or PET were the same as ¹⁸F-FDG PET/CT, except that ⁶⁸Ga-FAPI-04 PET imaging was acquired at 3.0 min/position. After the reconstruction is complete, image analysis is performed using the joint imaging postprocessing fusion software.

PET/CT Imaging Review

Both ¹⁸F-FDG and ⁶⁸Ga-FAPI-04 PET/CT images were interpreted in random order by 2 experienced nuclear medicine physicians. Any differences of opinion will be resolved by consensus. Based on knowledge of the normal biodistribution of ¹⁸F-FDG and ⁶⁸Ga-FAPI-04, lesions were identified as positive if increased tracer uptake was noted in the normal contralateral structure and surrounding soft tissues. Lesions were characterized as either positively or possibly abnormal if the uptake of ¹⁸F-FDG and ⁶⁸Ga-FAPI-04 was markedly to moderately increased. Diffuse mild or no increase in activity (in the absence of an abnormality on CT and no corresponding abnormality on PET) was considered normal or benign disease.

Tumors recurrence, individual lymph nodes, and lesions in the brain, lung, liver, peritoneum, mesentery, omentum, pleura, and bone were coded as positive if the activity exceeded that of adjacent background tissues. Individual lymph nodes were then classified into 8 sites, including the neck, supraclavicular, axilla, mediastinum (including epicardial, internal mammary, diaphragmatic lymph nodes), pulmonary hilum, abdomen (including para-aortic, porta hepatis, retroperitoneal, celiac), iliac, and inguinal regions.⁹ Brain, lung, liver, peritoneum, mesentery, omentum, pleura, and bone were each classified as individual sites. The SUV_max values were recorded for each site for both ¹⁸F-FDG and ⁶⁸Ga-FAPI-04. If multiple positive lesions occurred at any one site, the average SUV was used, which was obtained by averaging the SUVs of all lesions (≤5) or the SUVs of the 5 lesions with the highest activity (>5).⁹

Reference Standard

⁶⁸Ga-FAPI-04 and ¹⁸F-FDG PET/CT findings were validated by pathology as the criterion standard. In the case of the absence of pathological correlation, clinical and radiologic follow-up findings up to at least 3 months were taken into consideration to validate the ⁶⁸Ga-FAPI-04 and ¹⁸F-FDG PET/CT findings.

Statistical Analysis

All statistical analyses were performed using SPSS software (version 22.0; IBM, Armonk, NY). Paired-samples t test was used to compare the SUV_max of tumor recurrence, lymph node, and distant metastases of ¹⁸F-FDG and ⁶⁸Ga-FAPI-04 PET/CT. Results from the visually interpreted PET/CT images were compared with histopathology or follow-up results. We compared the statistical differences in detection rates of tumor recurrence, lymph nodes, and visceral metastases by ¹⁸F-FDG and ⁶⁸Ga-FAPI-04 PET/CT using the McNemar test. P < 0.05 was considered to indicate a statistically significant difference.

RESULTS

Patient Characteristics

Between January 2020 and June 2022, 41 patients (25 men, 16 women; median age, 51 years; interquartile range, 19–75 years) were included in this study. Among these patients, 10 (24%) had gastric cancer, 28 (68%) had colorectal cancer, and 3 (7%) had duodenal carcinoma. Among these 41 patients, 39 (95%) had adenocarcinomas, and 2 (5%) had signet ring cell carcinomas. Patient characteristics are summarized in Table 1.

TABLE 1 - Summary of Patient Characteristics

Characteristic	Value
No. patients	41
Age, y
Median	51
Interquartile range	19–75
Sex
Male	25
Female	16
Diagnosis
Gastric cancer	10
Colon cancer	19
Rectal cancer	9
Duodenal cancer	3
Patient status
Resection surgery	30
Chemotherapy after surgery	11
Histologic findings
Adenocarcinoma	39
Signet ring cell carcinoma	2

Comparative Results for Recurrence Detection

Seventeen of 41 participants had 17 positive anastomotic stoma uptakes that were detected by means of either ⁶⁸Ga-FAPI-04 or ¹⁸F-FDG PET. Tumor recurrence was correctly determined with use of ⁶⁸Ga-FAPI-04 in 6 of 17 participants (false-positive uptake in 7 participants, true-negative uptake in 4 participants), whereas 4 of 17 participants were correctly diagnosed with use of ¹⁸F-FDG PET/CT (false-positive uptake in 10 participants, false-negative uptake in 2 participants, and true-negative uptake in 1 participant). Pathologic examination served as the reference standard for 12 of 17 participants, and morphologic imaging and/or follow-up imaging for the other 5 participants. Finally, 6 positive participants in 17 participants were diagnosed as tumor recurrence.

Regarding the 6 participants in whom recurrence was detected, the true-positive rates for ¹⁸F-FDG PET/CT and ⁶⁸Ga-FAPI-04 PET/CT were 67% (4/6) and 100% (6/6), respectively (on a per-patient basis) (Fig. 1). Conversely, the true-negative rates of ⁶⁸Ga-FAPI-04 and ¹⁸F-FDG PET/CT in the detection of recurrence were 80% (28/35) and 71% (25/35), respectively (on a per-patient basis). For the detection of recurrence, there was no evidence of a higher SUV_max for ⁶⁸Ga-FAPI-04 PET than ¹⁸F-FDG PET (5.4 ± 3.4 vs 3.4 ± 0.4; P = 0.358) (Table 2). Six participants were misdiagnosed as tumor recurrence at both ⁶⁸Ga-FAPI-04 and ¹⁸F-FDG PET/CT, which were finally confirmed as inflammation of anastomosis. In addition, 1 participant with anastomotic fibrosis was misdiagnosed as tumor recurrence on ⁶⁸Ga-FAPI-04 but negative in ¹⁸F-FDG. ⁶⁸Ga-FAPI-04 shows better detection performance in recurrent lesions than ¹⁸F-FDG PET/CT.

FIGURE 1: A 29-year-old man who presented with abdominal pain for more than 10 days. The patient underwent curative resection for transverse colon adenocarcinoma 1 year ago. Images from ¹⁸F-FDG PET/CT (A–C) demonstrated no abnormal uptake throughout the whole body. ⁶⁸Ga-FAPI-04 PET/CT (D–F) showed increased tracer uptake in the anastomotic thickening (curved arrows, SUV_max of 10.4), which was confirmed tumor recurrence by gastroscopy, with multiple nodal (thin arrows, SUV_max of 9.7) and distant metastases (thick arrows, SUV_max of 10.9).

TABLE 2 - SUV_max Comparison Between ¹⁸F-FDG and ⁶⁸Ga-FAPI-04 PET/CT in the Tumor Recurrence and Nodal and Distant Metastases

Index	Tumor Recurrence	Nodal Metastasis	Distant Metastasis
18F-FDG SUVmax	3.4 ± 0.4	6.5 ± 3.6	8.9 ± 5.0
68Ga-FAPI-04 SUVmax	5.4 ± 3.4	8.3 ± 3.8	9.1 ± 5.1
P	0.358	0.002	0.886

Data are means ± SDs.

Diagnostic Performance of ¹⁸F-FDG and ⁶⁸Ga-FAPI-04 PET/CT in Nodal Metastasis

Twenty-one of 41 participants had 103 positive lymph nodes that were detected by means of either ⁶⁸Ga-FAPI-04 or ¹⁸F-FDG PET. Lymph node involvement was correctly determined with use of ⁶⁸Ga-FAPI-04 in 92 of 103 lymph nodes (false-positive uptake in 2 lymph nodes, true-negative uptake in 9 lymph nodes), whereas 31 of 103 lymph nodes were correctly diagnosed with use of ¹⁸F-FDG PET/CT (false-positive uptake in 11 lymph nodes and false-negative uptake in 61 lymph nodes). Pathologic examination served as the reference standard for 8 of 103 lymph nodes, and morphologic imaging and/or follow-up imaging for the other 95 lymph nodes. Finally, 92 positive lymph nodes in 20 participants were diagnosed as nodal metastases.

The nodal metastases were localized in only the abdomen in 7 participants; only the iliac regions in 4 participants; supraclavicular, axilla, mediastinum, abdomen, and iliac regions in 2 participants; mediastinum and pulmonary in 2 participants; the abdomen and supraclavicular fossa in 2 participants; the pulmonary hilum in 1 participant; mediastinum, pulmonary hilum, and abdomen in 1 participant; and neck, supraclavicular, axilla, mediastinum, abdomen, and iliac regions in 1 participant.

Among the 92 nodal metastases, 61 nodal metastases (including 23 abdomen lymph nodes, 11 mediastinum lymph nodes, 12 supraclavicular fossa lymph nodes, 9 iliac regions lymph nodes, 5 neck lymph nodes, and 1 pulmonary hilum lymph node) were missed at ¹⁸F-FDG PET. The sensitivity of ⁶⁸Ga-FAPI-04 PET/CT was higher than that of ¹⁸F-FDG PET/CT for detecting nodal metastases (92 of 92 [100%] vs 31 of 92 [34%]) (Fig. 2). For the metastatic lymph nodes, there was evidence of a higher SUV_max for ⁶⁸Ga-FAPI-04 PET than ¹⁸F-FDG PET (8.3 ± 3.8 vs 6.5 ± 3.6; P = 0.002) (Table 2). Two abdominal lymph nodes were misdiagnosed as nodal metastases at both ⁶⁸Ga-FAPI-04 and ¹⁸F-FDG PET/CT. The 2 lymph nodes were finally confirmed as reactive.

FIGURE 2: A 40-year-old man with rectal mucinous adenocarcinoma had a recent increase in tumor markers (carcinoembryonic antigen, CA-199, CA-724) after radiotherapy and chemotherapy for 9 months after surgery. Images from both ¹⁸F-FDG PET/CT (A–C, solid arrows) and ⁶⁸Ga-FAPI-04 PET/CT (D–F, dotted arrows) showed no increased tracer uptake in the anastomotic stoma, with a slightly increased ¹⁸F-FDG uptake in para-aortic lymph node (solid curved arrows, SUV_max of 1.0). In addition, ⁶⁸Ga-FAPI-04 PET/CT showed multiple para-aortic nodal metastases (dotted curved arrows, SUV_max of 8.4).

Diagnostic Performance of ¹⁸F-FDG and ⁶⁸Ga-FAPI-04 PET/CT in Distant Metastasis

⁶⁸Ga-FAPI-04 PET/CT visualized 30 distant metastases (including 9 hepatic metastases; 6 peritoneal, omentum, and/or mesenteric metastases; 6 pulmonary metastases; 4 bone metastases; 3 abdominal wall metastases; 2 chest wall and pleural metastases) in 18 participants. The distant metastases were localized in only the liver in 3 participants; in the peritoneum, omentum, and mesentery in 4 participants; in the bone in 2 participants; in the abdominal wall in 2 participants; in the lung in 1 participant; in the bone and lung in 1 participant; in the lung and chest wall in 1 participant; in the lung and liver in 1 participant; in the abdominal and chest wall in 1 participant; and in the abdominal wall and peritoneum in 1 participant. ⁶⁸Ga-FAPI-04 PET/CT missed 2 distant metastatic lesions in 1 participant (false-negative uptake in 2 lungs) (Fig. 3).

FIGURE 3: A 51-year-old woman with adenocarcinoma of the descending colon received 5 cycles of chemotherapy 4 months after surgery. ¹⁸F-FDG PET/CT (A–C) demonstrated increased tracer uptake in the iliac region lymph nodes (thin arrows, SUV_max of 8.9), liver masses (thick arrows, SUV_max of 9.5), and lung nodules (curved arrows, SUV_max of 1.8). ⁶⁸Ga-FAPI-04 PET/CT (D–F) showed increased tracer uptake in the iliac region lymph nodes (thin arrows, SUV_max of 8.2) and liver masses (thick arrows, SUV_max of 4.5), without pulmonary nodule uptake (curved arrows). The lesions were finally confirmed to be multiple metastases by follow-up.

¹⁸F-FDG PET/CT missed 10 distant metastatic lesions in 9 participants (false-negative uptake in 4 peritoneum, omentum, and mesentery metastases; 3 liver metastases; 2 bone metastases; and 1 pleural metastasis). Based on all distant metastases analysis, there was no evidence of a higher SUV_max for ⁶⁸Ga-FAPI-04 PET than ¹⁸F-FDG PET (9.1 ± 5.1 vs 8.9 ± 5.0; P = 0.886) (Table 2).

Changes in Oncologic Management After ⁶⁸Ga-FAPI-04 PET/CT

Three patients found more lymph node metastases in ⁶⁸Ga-FAPI-04 than ¹⁸F-FDG; 2 patients found lymph node metastases in ⁶⁸Ga-FAPI-04, but ¹⁸F-FDG was negative; 2 patients had tumor recurrence and lymph node metastases detected by ⁶⁸Ga-FAPI-04, but ¹⁸F-FDG was negative; in 1 patient, tumor recurrence and metastases were detected by ⁶⁸Ga-FAPI-04, and only tumor recurrence was detected by ¹⁸F-FDG, and no lymph node metastases were found.

DISCUSSION

Postoperative follow-up and recurrence detection for gastrointestinal tumors are helpful for patient management. ⁶⁸Ga-FAPI is a very promising tumor imaging agent, which represents a promising alternative to ¹⁸F-FDG.¹³ In this study, ⁶⁸Ga-FAPI-04 PET/CT demonstrated a higher detection efficacy for tumor recurrence (6 of 6 [100%] vs 4 of 6 [67%]), nodal metastases (92 of 92 [100%] vs 31 of 92 [34%]), and distant metastases (28 of 30 [93%] vs 20 of 30 [67%]) in participants with gastric, duodenal, and colorectal cancers compared with the detection efficacy of ¹⁸F-FDG PET/CT. ⁶⁸Ga-FAPI-04 PET/CT resulted in new oncologic findings in 8 of 41 participants compared with ¹⁸F-FDG PET/CT. Thus, PET/CT imaging using ⁶⁸Ga-FAPI may be beneficial for the clinical management for patients with gastric, duodenal, and colorectal cancers. In tumor recurrence detection, 1 signet ring cell carcinoma and 1 adenocarcinoma were negative on ¹⁸F-FDG but positive on ⁶⁸Ga-FAPI-04, which also shows the difference in the imaging principles of the 2 radiopharmaceuticals. ¹⁸F-FDG assessment of signet ring carcinoma subtypes (SRCC) is unreliable; these tumors exhibit lower ¹⁸F-FDG uptake than other tissue tumors and may be missed entirely.¹⁴ Overall, the exact role of ¹⁸F-FDG PET/CT in the detection of recurrent gastrointestinal cancer remains unclear, and our study shows that ⁶⁸Ga-FAPI outperforms ¹⁸F-FDG for recurrence detection.

Lymph node staging is critical for the treatment and prognosis of patients with gastric, duodenal, and colorectal cancers. In our study, compared with ¹⁸F-FDG PET/CT, ⁶⁸Ga-FAPI-04 PET/CT was more efficient in the detection of metastatic lymph nodes because there were many lymph nodes without ¹⁸F-FDG uptake. The high lymph node detection efficiency of ⁶⁸Ga-FAPI-04 PET/CT facilitates accurate lymph node staging and helps guide clinicians to determine reasonable postoperative treatment options. However, we noted that ⁶⁸Ga-FAPI uptake was also shown in reactive lymph nodes in our study, which is consistent with previous studies.^13,15,1668Ga-FAPI is not a more tumor-specific imaging tracer than ¹⁸F-FDG for metastatic lymph node detection. Therefore, it is necessary to consider false-positive uptake in lymph nodes when performing ⁶⁸Ga-FAPI PET/CT imaging.

In detecting distant metastases, ⁶⁸Ga-FAPI-04 PET/CT detected more lesions than ¹⁸F-FDG PET/CT. In our study, ⁶⁸Ga-FAPI-04 PET/CT detected 6 peritoneal metastases, whereas ¹⁸F-FDG was negative in 4 lesions. Recent studies have also demonstrated that ⁶⁸Ga-FAPI-04 is superior to ¹⁸F-FDG for the detection of peritoneal carcinomatosis in patients with gastric cancer.^17,18 Gastrointestinal tumors are prone to liver metastases, especially in more than half of the patients with colorectal cancer.¹⁹ For liver metastases, surgery is the only available curative treatment, increasing the 5-year survival rates up to 40%.²⁰ However, previous studies showed that, on a per-lesion basis, the ¹⁸F-FDG has a sensitivity of 62%, especially for small foci lesions, which may be due to physiological uptake in the liver.²¹ Our study shows that ⁶⁸Ga-FAPI-04 can detect more liver metastasis lesions and facilitates postoperative liver metastasis monitoring. Moreover, in a published study, ⁶⁸Ga-FAPI-04 PET was found to be more sensitive than ¹⁸F-FDG PET in detecting bone metastases, similar to our study.²² Interestingly, 2 lung metastases (diameter in 3 and 7 mm) were negative for ⁶⁸Ga-FAPI-04 uptake, but positive for ¹⁸F-FDG, which is different from previous reports,⁹ and the specific reasons remain to be further analyzed.

Overall, in this study, ⁶⁸Ga-FAPI-04 was found to show good results for the detection of tumor recurrence. For lymph node metastasis and distant metastasis, it is basically similar to the previous reports in the literature. In addition, studies also showed that the ⁶⁸Ga-FAPI-04 has obvious advantages over the ¹⁸F-FDG, including a better tumor-to-background ratio, the independence of blood glucose levels, and the feasibility of rapid image acquisition.¹⁸

Our study has limitations. First, our study sample included a relatively small number of participants (n = 41). Second, because of technical and ethical concerns, histopathological confirmation of all positive lesions could not be performed, as all lymph nodes and distant metastases could not be biopsied; therefore, morphological and/or follow-up imaging results were also used as reference standards for our investigation. Third, because noninvasive imaging is itself a reference standard for tumor staging, the evaluation of possible false-negative lesions is incomplete.

CONCLUSIONS

⁶⁸Ga-FAPI-04 PET/CT for tumor re-evaluation showed potential for more accurate performance of gastric, duodenal, and colorectal cancers, thereby improving treatment decision-making.