A 14-yr-old boy with a pathogenic MEN1 variant was diagnosed with asymptomatic insulinoma during routine follow-up

Abstract

Multiple endocrine neoplasia type 1 (MEN1) is an autosomal dominant disorder attributed to mutations in the MEN1 gene and is characterized by multiple endocrine tumors, including insulinoma. Asymptomatic hyperinsulinemic hypoglycemia and pancreatic nodules were incidentally detected in a 14-yr-old male carrying a pathogenic MEN1 variant. Although insulinoma was suspected, it did not meet Whipple’s triad, the classic diagnostic criteria for insulinoma, and some hypoglycemic provocation tests were negative. Selective arterial secretagogue injection (SASI) testing strongly suggested the presence of an insulinoma, and the lesions were surgically excised. The pathological findings were consistent with the SASI test results. Diagnosis of insulinoma based on conventional tests is challenging in some patients with asymptomatic insulinoma, and SASI testing can be useful not only for localization but also for insulinoma diagnosis.

Highlights

● The MEN 1 mutation carries a risk for endocrine tumors, including insulinoma.

● Some patients with insulinoma are asymptomatic.

● The SASI test is useful for insulinoma diagnosis and lesion localization.

Introduction

Multiple endocrine neoplasia type 1 (MEN1) is an autosomal dominant disorder characterized by tumors in the parathyroid glands, pancreatic islets, and anterior pituitary. The etiology of MEN1 involves mutations in the tumor suppressor gene MEN1, which encodes a 610-amino-acid protein called menin. Menin has recently been recognized for its dual role as a tumor suppressor or promoter. It suppresses cholangiocarcinoma, pancreatic ductal adenocarcinoma, and neuroendocrine tumors while functioning as a promoter in leukemia and colorectal, ovarian, and endometrial cancers (1). The penetrance rate of MEN1 is high, with clinical and biochemical manifestations developing in 50% of individuals by age 20 and in 95% by age 40. Clinical manifestations depend on the affected organs.

The prevalence rates of primary hyperparathyroidism, well-differentiated gastroenteropancreatic (GEP) neuroendocrine tumors, and anterior pituitary tumors are 95%, 40–70%, and 30–40%, respectively. The diagnostic criteria of international and Japanese guidelines are as follows: (a) Two or more endocrine tumors, including parathyroid, anterior pituitary, and well-differentiated GEP neuroendocrine tumors; (b) the presence of one endocrine tumor and a first-degree relative with MEN1; and (c) the MEN1 variant is identified through molecular genetic testing. Primary hyperparathyroidism accounts for more than 90% of initial endocrinopathies, whereas anterior pituitary tumors account for approximately 10% (2, 3).

Here, we report the case of a 14-yr-old male diagnosed with asymptomatic insulinoma. His father was diagnosed with MEN1, and the boy was found to carry the same pathogenic MEN1 variant at 11 yr of age. He underwent regular follow-ups every few months, and asymptomatic hyperinsulinemic hypoglycemia and pancreatic nodules were detected upon routine examination at 14 yr of age. Although we considered the possibility of insulinoma, establishing a diagnosis was challenging owing to the absence of symptoms. The selective arterial secretagogue injection test (SASI) was effective in establishing and localizing the diagnosis.

Patients and Methods

Here, we discuss a patient who was born at 34 wk and 6 d of gestation due to premature rupture of the membranes. At birth, his anthropometric measurements were within the normal range (length, 44 cm [+0.44 standard deviation (SD)]; weight, 1,836 g [−1.43 SD]), and Apgar scores were 9 and 10 at 1 and 5 min, respectively. He was treated in the neonatal intensive care unit (NICU) as a premature, low-birthweight infant. Oxygen therapy and intravenous infusion of a 10% glucose solution were initiated postnatally. He required oxygen until postnatal d 3. He was fed milk from the first day but had poor oral intake; therefore, tube feeding was used in combination from postnatal d 4 to 17. He did not experience hypoglycemia during his NICU stay, and no growth-related issues were documented (Fig. 1). At the age of 14, he was diagnosed with attention-deficit/hyperactivity disorder (ADHD) and learning difficulties (LD). Additionally, his mother had been diagnosed with acute lymphoblastic leukemia before conceiving him. Although she initially went into remission, disease relapse occurred during pregnancy, and she died when he was 3 yr old.

Fig. 1.

Growth chart. The patient’s growth pattern was within normal parameters.

When the patient was 2 yr of age, his father experienced chest pain caused by an atypical thymic carcinoid at the age of 47. Based on the results of blood tests, his father was incidentally diagnosed with hypercalcemia, which was attributed to hyperparathyroidism. As his father was self-employed, he had not undergone regular checkups prior to the hypercalcemia diagnosis. Based on these observations, the father was suspected to have MEN1, although the diagnostic criteria were not met at that time. Genetic testing revealed that the father carried the MEN1 variant (c.512_520delGGGATGTCC, [p. Arg171_Val173del]) that had been previously reported (4). This diagnosis prompted an investigation into the father’s family history, revealing that some relatives had MEN1 (Fig. 2). Given that the father died due to malignant thymic carcinoid associated with MEN1 when our patient was 9 yr old, genetic counseling was provided to him and his aunt, his father’s elder sister, who became his guardian following the father’s death. As requested by the guardian, genetic testing was performed when he was 11 yr of age. He was found to carry the same MEN1 variant as his father, and regular follow-up was initiated owing to his status as a mutation carrier.

Fig. 2.

Family tree of his paternal relatives. Squares represent males, circles represent females, and diagonal lines indicate deceased individuals. Black squares and circles indicate individuals with the MEN1 variant.

Around the age of 13, he began experiencing unusual drowsiness during school classes and club activities. Routine blood tests conducted during a randomly scheduled check-up revealed a blood glucose level of 87 mg/dL and an insulin level of 5.6 µU/mL, with no other abnormal findings. Accounting for other possible causes of the symptoms, ADHD, LD, or school-related fatigue were also considered; however, no evidence was detected. Further observation was deemed necessary, as drowsiness may manifest as a symptom of hypoglycemia when an insulinoma develops due to MEN1. At a routine check-up at 14 yr of age, blood tests (Table 1) revealed hypoglycemia (59 mg/dL) and hyperinsulinemia (42.4 µU/mL), despite being asymptomatic. Hypercalcemia (11.0 mg/dL) and inappropriately elevated intact parathyroid hormone (iPTH) level (68.2 pg/mL) were also detected, resulting in a diagnosis of hyperparathyroidism, which was attributed to MEN1. Simultaneously, magnetic resonance imaging (MRI) detected the presence of two small pancreatic nodules, approximately 10 mm in diameter (Fig. 3). A parathyroid imaging study was not conducted because the hypercalcemia was mild, and at present, there are no symptoms such as kidney stones, polyuria, nausea, or vomiting. Given the possibility of a MEN1-associated insulinoma, the patient was admitted to our hospital for a detailed examination.

Table 1. Laboratory findings at admission. This blood test was performed 1 h after a meal.

Fig. 3.

MRI images and yellow arrows indicate the lesions. Both lesions were measured approximately 10 mm. (a) Plain MRI image of the pancreatic head nodule. (b) Contrast-enhanced MRI image of the pancreatic head nodule. (c) Plain MRI image of pancreatic tail nodule. (d) Contrast-enhanced MRI image of pancreatic tail nodule.

Results

Although the 72-h fasting test is the gold standard for insulinoma diagnosis, it was not performed owing to the patient’s inability to tolerate fasting. Therefore, a 48-h fasting test and a 1-mg glucagon stimulation test after 48 h of fasting were performed as surrogates, and no hypoglycemic episodes (hypoglycemia was defined as a glucose level below 55 mg/dL) occurred during these tests (Fig. 4, Table 2). Although the 75-g oral glucose tolerance test (OGTT) was extended to 3 h to assess postprandial hypoglycemia (based on the incidental detection of hypoglycemia 1 h after a meal), no abnormalities were detected (Table 3). Endoscopic ultrasound (EUS)-guided biopsy of both MRI-detected nodules was performed to verify functionality. However, a nodule in the pancreatic tail could not be visualized, leading to biopsy cancellation. No other lesions were identified on contrast-enhanced MRI of the pancreas or contrast-enhanced computed tomography (CT). Because the diagnosis of insulinoma remained uncertain, a SASI test was performed. In this procedure, one catheter was inserted via the femoral artery into the pancreatic nutrient arteries, as described later, while another catheter was inserted via the femoral vein into the common hepatic vein. Calcium gluconate (0.025 mEq/kg) was injected at eight sites along the pancreatic arterial supply: the distal splenic, proximal splenic, right hepatic, left hepatic, distal gastroduodenal, proximal gastroduodenal, common hepatic, and superior mesenteric arteries (Fig. 5). Blood samples were collected to measure insulin levels before injection and every 30 sec thereafter for up to 150 sec. An increase in insulin levels to more than twice the baseline value was considered statistically significant. Injection into the proximal splenic artery resulted in an insulin elevation and was therefore considered a positive result (Table 3). No positive data were obtained at any of the other points. Although the symptoms and biochemistry did not meet the diagnostic criteria for insulinoma, the SASI test results suggested excessive insulin secretion, and the tail nodule was considered the culprit lesion. Somatostatin receptor scintigraphy (SRS) was also performed. The two pancreatic nodules identified on MRI were not detected, and no tracer uptake suggestive of metastasis was observed in other organs, such as the liver. Subsequently, the pancreatic nodules were enucleated. Both nodules were safely excised, and pathological examination was performed. The nodules were assessed by hematoxylin and eosin (H&E) staining and immunohistochemistry (Fig. 6). The nodule on the pancreatic head demonstrated insulin secretion in areas with normal pancreatic tissue, with no insulin secretion observed in the lesioned regions (Figs. 6a, b). In contrast, insulin secretion was observed throughout the lesion in the pancreatic tail nodules (Figs. 6c, d). The results revealed that one tumor in the head was nonfunctional, whereas the other in the tail was an insulinoma. Following surgery, fasting plasma glucose levels were monitored for eight consecutive days with no evidence of hypoglycemia. As a complication, pancreatic fistula leakage occurred 3 d later, and abscess formation occurred 17 d later, requiring drain replacement. The patient was discharged 62 d postoperatively. The drowsiness resolved after surgery. However, he stopped participating in club activities during the previous time period, which may have contributed to the improvement in drowsiness caused by fatigue. Given the risk of insulinoma recurrence, close monitoring is necessary to detect the recurrence of abnormal drowsiness.

Fig. 4.

Results of the 48-h fasting test. There is no significant hypoglycemia.

Table 2. Hypoglycemia provocation test results.

Table 3. Selective arterial secretagogue injection (SASI) test results.

Fig. 5.

The SASI test image. Numbers 1–8 indicate calcium injection sites.

Fig. 6.

Pathological specimen of pancreatic nodules. (a)(b): Head nodule. (c)(d): Tail nodule. The area outlined by the black line in (a) and (b) represents normal tissue, whereas the other region shows a lesion. (a) and (c) are stained with H&E. (b) and (d) are stained for insulin, and the brown area indicates insulin secretion.

Discussion

The classical diagnosis of insulinoma is established by meeting Whipple’s triad: (a) hypoglycemia (plasma glucose <50 mg/dL), (b) neuroglycopenic symptoms due to hypoglycemia, and (c) rapid symptom relief after glucose administration. The gold standard biochemical test is the 72-h fasting test in patients presenting with fasting hypoglycemia. This test can detect up to 99% of tumors (5). If the 72-h fasting test cannot be performed, C-peptide suppression and glucagon stimulation tests after 48 h of fasting have demonstrated efficacy. The mixed-meal test or 75-g OGTT is performed in patients who experience hypoglycemia only after meals (6, 7).

As noninvasive imaging tests for insulinoma localization, contrast-enhanced CT, contrast-enhanced MRI, transabdominal ultrasound, and SRS are known to demonstrate sensitivities of 29–80%, 40–90%, 9–64%, and 53%, respectively (5, 8). Although SRS a has high affinity for somatostatin receptor subtypes 2 (SSTR2), 3, and 5 —particularly subtype 2— indolent insulinomas exhibit lower SSTR2 expression than other pancreatic neuroendocrine tumors, resulting in suboptimal SRS sensitivity for insulinoma detection. Recently, glucagon-like peptide-1 receptor (GLP-1R) PET/CT has been reported to achieve high sensitivity, as GLP-1R is predominantly expressed in pancreatic β-cells and overexpressed in indolent insulinomas; however, this modality is currently unavailable for clinical use in Japan (9). As invasive examinations, EUS and SASI test sensitivities range from 86.6% to 92.3% and 94% to 100%, respectively. However, EUS sensitivity may vary depending on operator expertise and lesion location, particularly for pancreatic tail lesions, where false-negative rates are likely to be higher (5). Furthermore, the SASI test involves the insertion of catheters into the femoral artery and vein, which results in a higher degree of invasiveness than other examinations. Particularly in children, an extended sedation period is required, and safety considerations must be addressed.

In our patient, mild hyperinsulinemic hypoglycemia without obvious symptoms was incidentally detected during a routine health examination; therefore, none of Whipple’s triad criteria were met. Although the possibility that abnormal drowsiness could be a symptom of hypoglycemia cannot be ruled out, no blood glucose measurements were performed at symptom onset, and the cause remained undetected. When hyperinsulinemic hypoglycemia was later identified, given the patient’s known MEN1 variant status and the presence of two pancreatic nodules, insulinoma was strongly suspected, warranting further diagnostic investigation. A 48-h fasting test, followed by a 1-mg glucagon stimulation test, was performed as a hypoglycemic induction test. A 75-g OGTT was performed to evaluate postprandial hypoglycemia because mixed-meal testing was not available at our hospital. However, no hypoglycemic episodes were observed, and a diagnosis of insulinoma was not confirmed. Furthermore, the pancreatic tail nodule was not visualized by EUS; thus, fine-needle aspiration of both lesions could not be performed, and preoperative pathological examination was not possible. In the SASI test, calcium stimulation in the proximal splenic artery resulted in a notable increase in insulin, strongly suggesting an insulinoma, and the perfusion area indicated the pancreatic tail nodule as the culprit lesion. Conversely, calcium stimulation in the common hepatic artery revealed an insulin increase to almost twice the baseline value, and the possibility that the pancreatic head nodule was an insulinoma could not be completely ruled out. However, because no increase in insulin was observed upon stimulating the gastroduodenal artery, which perfuses the pancreatic head, the nodule was considered a nonfunctional lesion.

For MEN1-associated insulinomas, surgery is recommended as the first-line treatment, and surgical methods include enucleation and partial pancreatectomy. To prevent the onset of diabetes mellitus and the decline in exocrine pancreatic function post-surgery, maximum preservation of pancreatic tissue is essential. In cases where surgery cannot be tolerated, EUS-guided alcohol ablation, radiofrequency ablation (RFA), or embolization are considered useful alternatives. Additionally, somatostatin analogs (octreotide), diazoxide, and everolimus have been proven effective in suppressing hypoglycemic attacks (3, 5). In our case, drug therapy was another option, considering the risk of recurrence or reoperation. Conversely, the SASI test could not rule out the possibility that the pancreatic head nodule was an insulinoma, and pathological examination was necessary for a definitive diagnosis. Therefore, surgical intervention was selected. Considering the possibility of undetected culprit lesions on MRI, the pancreatic resection site was determined using intraoperative visual inspection, palpation, and ultrasound (US). During surgery, two firm and elastic nodules were detected macroscopically in the head and tail of the pancreas, respectively, which were clearly distinct from the normal pancreatic tissue. However, intraoperative US revealed multiple microscopic and hyperechoic nodules throughout the pancreas, which complicated the identification of the culprit lesion. Considering the endocrine function and quality of life after surgery, blind pancreatectomy is not recommended. Evidence suggests that larger nodules correlate with increased insulin secretion and higher malignancy potential (10, 11). Therefore, the two macroscopic nodules were enucleated. Postoperative pathological examination revealed that the pancreatic head nodule was nonfunctional, and the nodule in the pancreatic tail was an insulinoma, confirming the accuracy of the SASI test. No postoperative hypoglycemic episodes occurred, and no residual lesions were observed. In the surgical treatment of insulinoma, the resection site should be comprehensively determined, considering not only noninvasive imaging studies but also SASI testing and intraoperative examinations.

The key consideration is that the SASI test and surgery are highly invasive procedures, particularly surgery, which carries risks such as reoperation or adhesion formation. Therefore, their necessity is debatable in patients with asymptomatic hyperinsulinemic hypoglycemia who present with no positive findings on fasting tests, 48-h fasting glucagon stimulation tests, or OGTT. A noninvasive monitoring approach may serve as an alternative method, implemented as follows: regular clinical visits and blood draws are conducted until clinical symptoms emerge, with asymptomatic hypoglycemic data collected through continuous glucose monitoring (CGM). The objective of CGM is to determine the frequency and severity of hypoglycemia while facilitating the acquisition of critical samples. Nevertheless, caution is warranted, as hypoglycemia cannot be prevented. In the current case, genetic testing confirmed MEN1, and blood tests combined with MRI findings indicated a high likelihood of insulinoma in the pancreatic lesions. Selecting observation only is associated with a risk of severe complications due to hypoglycemia; therefore, a thorough evaluation, including invasive testing, was deemed necessary. For patients with asymptomatic hypoglycemia, testing is recommended when underlying causes are suspected, as this may improve patient outcomes. The treatment or monitoring plans should be carefully selected based on the condition, age, and preferences of the patient.

Another key point was that the patient presented with hypercalcemia and elevated iPTH levels. Given the presence of the MEN1 variant, hyperparathyroidism is the most likely cause. However, the hypercalcemia was mild, and there were no evident symptoms such as kidney stones, polyuria, nausea, or vomiting due to hypercalcemia. Therefore, imaging evaluation of the parathyroid glands was not performed. Generally, the symptoms of hyperparathyroidism are thought to develop around the age of 20–25 yr; therefore, regular blood tests should be performed. If worsening hypercalcemia is observed, imaging tests such as 99mTc-hexakis-2-methoxy-2-methylpropyl-isonitrile (99mTc-MIBI) scintigraphy should be conducted (3).

Conclusion

Regular medical monitoring is required for patients with pathogenic MEN1 variants. However, no defined methods exist for managing or monitoring such cases. Given the possibility of insulinoma development in young patients with MEN1, insulinoma should be investigated in patients with asymptomatic hyperinsulinemic hypoglycemia. Therefore, the SASI test should be considered owing to its usefulness and high sensitivity. However, it is not 100% accurate and is highly invasive. Accordingly, it is not recommended for all cases with normal results in other tests. The decision to perform the SASI test and subsequent surgery should be based on the patient’s age or willingness to undergo additional examinations.

Conflict of interests

The authors have nothing to declare.