Glucagon-like peptide-1 receptor PET/CT in gastrinomas: a case series demonstrating the precise characterization to guide surgical and medical treatment

Abstract

Gastrinoma is a rare type of neuroendocrine tumor with ectopic secretion of gastrin causing Zollinger–Ellison syndrome (ZES). Accurate localization of gastrinomas and their metastases is important for developing precise treatment strategies. We reported three cases of metastatic gastrinomas evaluated with ⁶⁸Ga-exendin-4 and ⁶⁸Ga-DOTANOC PET/CT. Despite most gastrinomas highly expressing somatostatin receptor (SSTR), these patients showed no or heterogeneous uptake on ⁶⁸Ga-DOTANOC PET/CT, whereas ⁶⁸Ga-exendin-4 PET/CT sensitively detected the primary tumor, lymph node metastasis, liver metastasis, and bone metastasis, facilitating accurate tumor assessment and supporting clinical decision-making. The findings were correlated with enhanced CT and MRI and confirmed the location of the hormone-related tumors. This highlights the potential utility of ⁶⁸Ga-exendin-4 PET/CT as a promising imaging modality for evaluating gastrinomas with no or heterogeneous SSTR expression and guiding precise treatment strategies.

Introduction

Gastrinoma is a type of functional neuroendocrine tumor (NET) due to ectopic secretion of gastrin, which causes Zollinger–Ellison syndrome (ZES), characterized by severe peptic ulcer disease, gastroesophageal reflux disease and chronic diarrhea (Chatzipanagiotou et al. 2023, Jensen et al. 2025, Norton et al. 2018). Gastrinomas often occur in the “gastrinoma triangle”, which is formed by joining the following three points: the confluence of the cystic and common bile ducts superiorly, the junction of the second and third parts of the duodenum inferiorly, and the junction of the neck and body of the pancreas medially. Gastrinomas more commonly occur in the duodenal bulb and the head of the pancreas (Kloppel 2011). Despite their relatively indolent biological behavior in many patients, gastrinomas are frequently associated with delayed diagnosis because clinical manifestations often overlap with more common gastrointestinal disorders and the widespread use of long-term proton pump inhibitor (PPI) can induce hypergastrinemia mimicking ZES (Ito et al. 2024).

As with patients with other functional pancreatic neuroendocrine tumors, treatment for gastrinomas focuses on both inhibiting tumor proliferation and suppressing hormone secretion (Norton et al. 2018, Wahba et al. 2024, Hofland and Herder 2023). About 80% of gastrinomas are sporadic, and the rest are caused by multiple endocrine neoplasia type 1 (MEN1) syndrome. Approximately 30% of MEN1 patients develop ZES, in which tumors are commonly multifocal and clinically challenging to localize, especially for functional tumors.

Biochemical diagnosis primarily relies on elevated fasting serum gastrin levels in the presence of inappropriate gastric acid secretion. However, interpretation may be complicated by proton pump inhibitor therapy, chronic atrophic gastritis, renal insufficiency, or achlorhydria (Berna et al. 2006). Consequently, imaging plays a pivotal role not only in tumor localization but also in staging, treatment selection, and response assessment. Conventional anatomical imaging modalities such as contrast-enhanced computed tomography (CT), magnetic resonance imaging (MRI), and endoscopic ultrasound (EUS) remain fundamental components of the diagnostic workup. EUS demonstrates particularly high sensitivity for the tumors of the duodenal pancreatic area (Gincul et al. 2015), whereas multiphasic CT and MRI are useful for evaluating hepatic metastases and locoregional disease. Nevertheless, tiny gastrinomas frequently remain occult on anatomical imaging because of their submucosal location and limited size.

Functional molecular imaging has substantially improved the detection of gastrinomas over the past two decades. Most well-differentiated gastrinomas overexpress somatostatin receptor (SSTR) (Naswa et al. 2013), particularly SSTR2, providing the biological basis for somatostatin receptor imaging. SSTR PET/CT has become the imaging modality of choice for staging and treatment planning in well-differentiated neuroendocrine tumors, including gastrinomas (Wahba et al. 2024). In addition to diagnosis, SSTR imaging provides essential information for determining eligibility for peptide receptor radionuclide therapy (PRRT). However, several limitations remain. Small duodenal lesions may still evade detection due to partial volume effects, physiological bowel activity, or heterogeneous SSTR expression. ¹⁸F-FDG PET/CT may provide complementary biological information, particularly in higher-grade or aggressive disease, where increased glucose metabolism correlates with adverse prognosis and more rapid tumor progression. Recent advances in molecular imaging have further expanded the diagnostic landscape of gastrinomas. Novel tracers targeting the glucagon-like peptide-1 receptor (GLP-1R) and SSTR antagonists are currently under investigation for neuroendocrine tumor imaging. As is well known, exendin-4 PET/CT is primarily used to diagnose insulinomas, with a sensitivity of over 90% (Zhang et al. 2025, Qi-Chang et al. 2025, Aveline et al. 2025). However, a case report describes incidental findings indicating that metastatic gastrinoma also shows intense uptake of exendin-4 (Zhang et al. 2025).

Here, we present three cases of metastatic gastrinomas that underwent both ⁶⁸Ga-exendin-4 and ⁶⁸Ga-DOTANOC PET/CT, confirming the significant value of ⁶⁸Ga-exendin-4 PET/CT in diagnosing and mapping gastrinomas with no or heterogeneous SSTR expression, and describing its potential utility for future radioisotope treatment targeting the GLP-1R.

Case presentations

Case 1

A 46-year-old male patient presented with recurrent diarrhea for 6 months that had progressively worsened, with up to 5–6 watery stools per day at its most severe, which could be improved by self-administering the PPI and would recur when discontinuing the PPI. Gastroscopy revealed erythematous gastric mucosa and multiple ulcerative lesions. Contrast-enhanced CT showed a mass adjacent to the gastric antrum (Fig. 1). Serum gastrin level was 435.0 pg/mL, markedly higher than normal (13.0-115.0 pg/mL). The diagnosis of gastrinoma was suspected based on the clinical manifestations and laboratory findings.

Fig. 1.

Gastroscopy revealed erythematous gastric mucosa with a mottled red-and-white “map-like” appearance (A), and multiple ulcerative lesions densely distributed in the duodenal bulb and descending part (B). Contrast-enhanced CT showed a mass adjacent to the gastric antrum with heterogeneous enhancement (C), and no obvious enhancing lesions in the liver (D)

For further tumor characterization and assessment of tumor stage, ⁶⁸Ga-DOTANOC PET/CT was performed, revealing mild radioactive uptake in the periantral mass, similar to that of normal hepatic tissue. The previous immunohistochemical study indicated that gastrinomas can express GLP-1R (Reubi and Waser 2003). Therefore, we attempted ⁶⁸Ga-exendin-4 PET/CT for tumor detection, which showed intense radiotracer uptake in the periantral mass and an unexpectedly small, high-uptake metastatic lesion. (Fig. 2)

Fig. 2.

⁶⁸Ga-DOTANOC PET/CT (A, MIP image) revealed the periantral mass with mild radioactive uptake similar to the normal hepatic tissue (B, SUVmax = 6.3). No abnormal foci of increased tracer uptake were detected in the liver (C, SUVmax = 6.9). ⁶⁸Ga-DOTANOC PET/CT (D, MIP image) indicated that the periantral mass demonstrated intense radiotracer uptake (E, SUVmax = 10.9), and an unexpectedly small metastatic lesion was identified in hepatic segment 4 (F, SUVmax = 7.5)

Surgical resection of both the periantral tumor and the hepatic metastasis in the left lobe was performed. Pathological examination confirmed the diagnosis of NET G2, with immunohistochemistry showing SSTR (+), Ki-67 (15%+), and gastrin (+), but insulin (-). The serum gastrin level returned to normal (76.0 pmol/mL) in 1 month after the operation. (Fig. 3)

Fig. 3.

Surgical resection of both the periantral tumor and the hepatic metastasis in the left lobe was performed. Pathological examination confirmed the diagnosis of NET G2 (A, H&E). The immunohistochemistry results showed SSTR2 (+) (B), Ki-67 (15%+) (C), gastrin (+) (D), insulin (-)

Case 2

A 50-year-old male patient presented in November 2021 with a 3-month history of diarrhea. Gastroscopy revealed an elevated lesion at the gastric angle and multiple ulcers in the descending duodenum. Pathological examination indicated a G2 NET of the gastric angle. Immunohistochemistry showed Syn (+), CgA (+) and Ki-67 (40% +). Subsequently, the patient underwent laparoscopic distal gastrectomy and resection of liver metastatic lesions. Postoperative pathology demonstrated a G3 gastric NET involving the serosal layer, with no lymph node metastasis (0/3). Immunohistochemistry revealed Syn (+), CgA (+), Ki-67 (60%+), SSTR2 (++), SSTR5 (++). After surgery, the patient started STEM regimen (S1 + Temozolomide) chemotherapy combined with Surufatinib treatment until September 2022. Imaging evaluations every 3 months during treatment demonstrated stable disease. In July 2024, ⁶⁸Ga-DOTANOC PET/CT demonstrated multiple liver metastases that lacked somatostatin receptor expression (Fig. 4).

Fig. 4.

⁶⁸Ga-DOTANOC PET/CT (A, MIP image) revealed multiple liver metastases with no SSTR expression (B-C, CT images; D-E fused images, arrow for tumors)

Thereafter, radiofrequency ablation was performed for liver metastases and maintenance therapy with Surufatinib 300 mg orally once daily was continued.

In July 2025, progression of intrahepatic metastatic lesions was observed, accompanied by recurrence of significant diarrhea symptoms. Serum gastrin level was elevated to 410 pg/mL. Meanwhile, ¹⁸F-FDG PET/CT and ⁶⁸Ga-exendin-4 demonstrated multiple hepatic metastases with low FDG metabolism and markedly increased ⁶⁸Ga-exendin-4 uptake (Fig. 5). The patient was then treated with the CAPTEM regimen (Capecitabine + Temozolomide) chemotherapy. In January 2026, the gastrin level reduced to 316.00 pg/mL. Imaging evaluation in March 2026 demonstrated stable disease (SD).

Fig. 5.

¹⁸F-FDG PET/CT (A: MIP image) re-evaluation for the liver metastasis detected by enhanced CT (B) and MR (C), which showed low FDG metabolism (E: fused image). In contrast, ⁶⁸Ga-Exendin-4 PET/CT (D: MIP image) demonstrated the lesion with markedly increased uptake (SUVmax = 7.8) (F, fused image)

Case 3

A 69-year-old male patient began experiencing diarrhea in 2018, initially intermittent in nature, occurring every 3–4 days on average with 2–3 episodes per day. During severe periods, the frequency increased to more than 10 episodes daily. In July 2022, contrast-enhanced CT revealed hepatic lesions. Liver biopsy pathology demonstrated NET G2. Immunohistochemistry showed: CgA (+), Syn (+), Ki-67 (15%+), p53(+), and SSTR2 (5%+). In August 2022, the patient underwent laparoscopic right hepatectomy, pancreatic tumor resection, and lymph node dissection. Pathology confirmed pancreatic NET, G2, with lymph node metastases (7/7) and multiple liver metastases. Immunohistochemistry demonstrated: Ki-67 (5%+), Syn (+), CgA (+), SSTR2 (focally ++), SSTR5 (+++), INSM1(+), Gastrin (+), Glucagon (-), and Insulin (-). Postoperatively, the patient received long-acting octreotide (Sandostatin LAR) 30 mg every month, after which the diarrhea improved significantly.

In April 2025, contrast-enhanced CT demonstrated enlargement of a metastatic lesion in liver segment S2 and multiple enlarged retroperitoneal lymph nodes. The patient also developed recurrent watery diarrhea, which could be alleviated by Sandostatin LAR 30 mg and oral omeprazole. In June 2025, both ¹⁸F-FDG PET/CT and ⁶⁸Ga-DOTANOC PET/CT revealed scattered metastatic lesions in the remnant liver, lymph nodes and bone metastases with intertumoral heterogeneity of SSTR expression and low FDG uptake (Fig. 6). At the same time, the serum gastrin level was elevated to 655.00 pg/mL. Subsequently, ⁶⁸Ga-exendin-4 PET/CT demonstrated high GLP-1R expression in both hepatic and more osseous metastatic lesions (Fig. 7).

Fig. 6.

¹⁸F-FDG PET/CT (A, MIP image) and ⁶⁸Ga-DOTANOC PET/CT (B, MIP image) revealed metastatic abdominal and retroperitoneal lymph nodes, scattered metastatic lesions in the remnant liver and bones with low FDG metabolism (C-E, fused images, arrows for metastasis). Lymph node uptake of ⁶⁸Ga-DOTANOC was markedly increased (SUVmax = 24.5), while uptake in liver and bone metastases was mildly increased (SUVmax = 10.1), which showed significant intratumoral heterogeneity in SSTR expression (F-H, fused images, arrows for metastasis)

Fig. 7.

⁶⁸Ga-exendin-4 PET/CT (A. MIP image, red circles for additional lesions) showed that high GLP-1R expression of bone (B), liver(C) and lymph nodes metastases (D), and detected more lesions located in the left ilium and femur (E)

Discussion

Gastrinomas are a type of functional neuroendocrine tumor (NET) that most commonly originates from the gastrinoma triangle, which includes the duodenum (60–70%), the pancreatic head (20–30%), and adjacent tissues (10–20%). These tumors with secretion of gastrin lead to Zollinger–Ellison syndrome, characterized by peptic ulcers, gastroesophageal reflux, and chronic diarrhea. Gastrinomas are often associated with MEN1 syndrome (Spada et al. 2025). Diagnosis of gastrinomas relies on clinical presentation, elevated serum gastrin levels, and gastric acid pH ≤ 2. However, diagnosis remains challenging, as hypergastrinemia can also result from factors such as PPI use, renal insufficiency, or Helicobacter pylori infection. In addition, small tumors often escape detection by conventional imaging, particularly in MEN1-associated multiple gastrinomas, where identifying the hormone-secreting lesion can be difficult. Because of hormone-related symptoms and high risk of lymph node metastasis even in very small tumors (< 5 mm), surgical resection is the standard treatment for localized gastrinomas, regardless of primary tumor size. Therefore, accurate localization and staging of gastrinomas are crucial for developing precise therapeutic strategies.

Gastrinomas typically appear as small, round, arterially enhancing lesions on CT imaging within the gastrinoma triangle and show no significant differences from nonfunctional well-differentiated neuroendocrine tumors. Endoscopic ultrasound (EUS) can detect small lesions measuring 2–5 mm in the duodenum and pancreas, and therefore has diagnostic value in identifying gastrinomas. It is well known that more than 80% of gastroenteropancreatic neuroendocrine tumors (GEP-NETs) express somatostatin receptors (SSTRs) at high levels. SSTR PET/CT is considered the gold standard for diagnosing well-differentiated neuroendocrine tumors. An early study demonstrated that ⁶⁸Ga-DOTANOC PET/CT achieved an overall lesion detection rate of up to 92.8% in patients with CT-negative or indeterminate gastrinomas (Naswa et al. 2013). More recent studies have reported that the overall diagnostic sensitivity of SSTR PET for gastrinomas ranges from 68% to 100%. Furthermore, an in-vitro study has shown that gastrinomas express SSTR2, SSTR3, and SSTR5 at rates of 100%, 79%, and 76%, respectively (Hu et al. 2021). However, in clinical practice, some patients exhibit absent or heterogeneous SSTR expression in the different lesions. Thus, accurately localizing the tumor and identifying those lesions that are hormonally active remain crucial clinical challenges.

⁶⁸Ga-exendin-4 PET/CT targets the GLP-1R and is primarily used for detecting insulinomas. Since insulinomas, the most common functional pancreatic NETs, express SSTR2 in only about 50% of cases, SSTR PET/CT often fails to localize these tumors. In contrast, ⁶⁸Ga-exendin-4 PET/CT demonstrates approximately 95% sensitivity and accuracy in diagnosing insulinomas (Zhang et al. 2025). In immunohistochemical analyses (Reubi and Waser 2003), both SSTR2 and GLP-1R of 11 gastrinomas showed high expression, whereas one of 27 insulinomas did not express GLP-1R. Recently, Peking Union Medical College Hospital reported a case (Zhang et al. 2025) of patient presenting with hyperinsulinemia who underwent ⁶⁸Ga-exendin-4 PET/CT to rule out insulinoma. The scan revealed intense uptake in a retroperitoneal lymph node, and biopsy with immunohistochemistry showed gastrin (+) and insulin (-), incidentally identifying a gastrinoma with high ⁶⁸Ga-exendin-4 uptake. Unlike that case, the 3 cases we reported exhibited low uptake on SSTR PET but high uptake on GLP-1R PET, highlighting the importance of ⁶⁸Ga-exendin-4 PET for detecting gastrinomas that lack SSTR expression. Our center previously reported a patient with MEN1 syndrome in whom combined SSTR and exendin-4 PET/CT distinguished insulin-secreting from nonfunctional lesions among multiple pancreatic NETs. Then, surgical removal of the insulin-secreting lesions successfully relieved hormone-related symptoms (Xu et al. 2024). Similarly, as some gastrinomas are associated with genetic syndromes, particularly MEN1 syndrome, which have both functional and non-functional NETs, ⁶⁸Ga-exendin-4 PET/CT can identify gastrin-secreting lesions, allowing for their surgical resection and improvement of hormone-related functional symptoms. Even in cases of metastatic gastrinomas, GLP-1R PET can also detect more lesions, particularly in cases with no or heterogeneous SSTR expression, to determine the tumor burden of gastrin-secreting tumors and provide guidance for individual treatment strategies. Moreover, it may provide a future research direction of this new molecular target for radionuclide-based therapies.

Our series of case reports has certain limitations, including a small sample size and the lack of immunohistochemical confirmation of GLP-1R expression in gastrinomas. In the future, we hope to accumulate more cases of gastrinoma to conduct prospective clinical studies and, through comparison with pathological immunohistochemistry, explore the value of ⁶⁸Ga-exendin-4 PET/CT in the diagnosis of gastrinoma.

Conclusion

⁶⁸Ga-exendin-4 PET/CT is a promising imaging modality for evaluating gastrinomas with no or heterogeneous SSTR expression and for guiding precise treatment strategies.

Abbreviations

ZES

Zollinger ellison syndrome

SSTR

Somatostatin receptor

NET

Neuroendocrine tumor

PPI

Proton pump inhibitor

MEN1

Multiple endocrine neoplasia type 1

GLP-1R

Glucagon-like peptide-1 receptor

CT

Computed tomography

MRI

Magnetic resonance imaging

EUS

Endoscopic ultrasound

Author contributions

All authors contributed to the study conception and design. The clinical material preparation was provided by Liang Yun and Chen Jie. The pathological images were provided by Dan Huang. The review of imagings and clinical information was performed by Xu Junyan. The paper was written by Junyan Xu and then revised by Shaoli Song. And all authors commented on previous versions of the manuscript. The critical revision of the manuscript was provided by Shaoli Song. All authors read and approve the final manuscript.

Funding

No funding source.

Data availability

The datasets of the current study are available from the corresponding author on reasonable request.

Declarations

Ethical approval

Ethical approval was granted by the Ethics Committee of Fudan University Shanghai Cancer Center in line with the principles of the Declaration of Helsinki.

Submission declaration

This is authors original and own work, which has not been submitted elsewhere for publication.

Participant consent for publication

The authors affirm that participant provided informed consent for publication of images in the Figures in manuscript with concealed identification.

Competing interests

There is no relevant financial or non-financial interests to disclose.