Insulinoma presenting with anti-insulin antibodies

Background

Insulinoma is the most common cause of hypoglycemia due to endogenous hyperinsulinemia and occurs in 4 cases per 1 million person-years (1). Surgical removal is the fundamental treatment (2). However, when patients are not considered suitable surgical candidates or decide against surgery, alternative pharmacotherapies for the management of hypoglycemia include the administration of diazoxide, somatostatin analogs such as octreotide, and mammalian target of rapamycin (mTOR) inhibitors (everolimus) (2).

Insulin autoimmune syndrome (IAS, also known as Hirata disease) also results in spontaneous hypoglycemia, caused by anti-insulin antibodies. A recent study estimated a prevalence for this pathology of 0.017 cases per 100,000 capita in the Japanese general population, and IAS represents the third leading cause of hypoglycemia following insulinoma and non-insulinoma pancreatogenous hypoglycemia syndrome (3). No consensus has yet been reached on the optimal treatment for IAS, so patients have been given various treatment plans including drug withdrawal, frequent small meals, avoidance of monosaccharide intake, alpha-glucosidase inhibitors, glucocorticoids, plasma exchange, and immunosuppressive agents (4).

Insulinoma presenting with concomitant anti-insulin antibodies appears rare. To the best of our knowledge, only one case of insulinoma with the concomitant anti-insulin antibodies has been reported in the English language literature previously, with insulinoma pathologically confirmed by a biopsy specimen obtained under endoscopic ultrasonographic guidance (5). We report a rare case of insulinoma in a patient showing anti-insulin antibodies in the serum.

Case presentation

An 89-year-old woman presented with a 6-year history of occasional episodes of impaired consciousness that were relieved upon eating a snack. Hypoglycemia had been noted in a local clinic where she was being treated for hypertension, but no further examinations were performed. Three months prior to this presentation, she had been hospitalized at another local hospital with subdural hematoma resulting from a head contusion, and previously undocumented hypoglycemia (40–60 mg/dL, 2.2–3.3 mmol/L) had been discovered incidentally. The head contusion was thus attributed to a fall during a hypoglycemic episode. Serum insulin concentration at this point was relatively high (34.9 µU/mL) with low plasma glucose concentration (37 mg/dL). Computed tomography (CT) showed a 14 mm hyperenhancing tumor in the tail of the pancreas (Fig. 1) and she was referred to the Department of Endocrinology and Metabolism at our hospital for further investigation.

View Full Size

Computed tomography of insulinoma in the pancreas. The arrowhead indicates a 14 mm hyperenhancing tumor in the tail of the pancreas (red arrow).

Citation: Endocrinology, Diabetes & Metabolism Case Reports 2024, 3; 10.1530/EDM-24-0062

• Download Figure
• Download figure as PowerPoint slide

Blood pressure was 115/78 mmHg, height was 152 cm, weight was 71.8 kg, and body mass index was 31.1 kg/m². Physical examination showed no abnormalities other than mild edema of the legs. Right mastectomy had been performed at 38 years of age for breast cancer, and right thyroid lobectomy had been performed at 84 years of age due to thyroid cancer. She had no history of diabetes mellitus and had never taken exogenous insulin injections, supplements containing alpha-lipoic acid, or sulfhydryl group-containing agents. She was taking a calcium antagonist (amlodipine) and an angiotensin receptor blocker (candesartan) for hypertension, and atorvastatin for dyslipidemia. No contributory family history was noted.

Investigation

A blood examination performed at 08:00 h in a fasted state in our hospital showed a plasma glucose concentration of 67 mg/dL (3.7 mmol/L) (Table 1). Neither adrenal insufficiency nor growth hormone deficiency was suspected from laboratory data. The concentration of thyroid-stimulating hormone was mildly elevated with normal concentrations of free triiodothyronine and free thyroxine, which was considered to be associated with the history of right thyroid lobectomy. Whereas the concentration of intact parathyroid hormone was mildly elevated to 77 pg/mL, concentrations of corrected calcium and phosphate were 8.7 mg/dL and 3.0 mg/dL, respectively, which were at the lower limit of the reference values. These findings were considered to be associated with aging. No parathyroid nodules were observed on thyroid echography. Therefore, it was unlikely that the patient suffered from primary hyperparathyroidism, which is most frequently observed in multiple endocrine neoplasia type 1. Interestingly, the concentration of anti-insulin antibodies was elevated at 21.1 U/mL (Table 2). Scatchard analysis was performed to characterize the anti-insulin antibodies but failed due to the low concentration of antibodies. The concentrations of anti-glutamic acid decarboxylase and anti-insulinoma-associated protein-2 antibodies were < 0.5 U/mL and < 0.6 U/mL, respectively, which were below the threshold of measurement sensitivity. Human leukocyte antigen (HLA) typing revealed HLA-DRB1*0407 and HLA-DRB1*1405 alleles.

In a 72-h prolonged fasting test, plasma glucose concentration fell to 43 mg/dL (2.4 mmol/L) 8 h after the last meal. Serum concentrations of insulin, proinsulin, and C-peptide were 21.1 µU/mL, 16.9 pmol/L, and 2.72 ng/mL, respectively. Subsequent intravenous administration of 1 mg of glucagon increased the plasma glucose concentration to 76 mg/dL (4.2 mmol/L) (Table 3). These data indicated the presence of endogenous hyperinsulinemia. The calculated insulin-to-C-peptide molar ratio was 0.14, suggesting insulinoma. Taking these results and the images from CT together, insulinoma was diagnosed (Fig. 1). Selective arterial calcium stimulation testing was not performed due to the advanced age of the patient. Daily blood glucose profiles were assessed using a flash glucose monitoring (FGM) system (FreeStyle Libre Pro (FSL-Pro); Abbot, Tokyo, Japan) (Fig. 2), revealing hypoglycemia in the fasting and postprandial states. Notably, the patient was unaware of hypoglycemia during the prolonged fasting test and glucose monitoring by FSL-Pro. For mild leg edema, echocardiography showed a normal ejection fraction (68%) with no valvular heart disease. Moreover, deep vein thrombosis was not observed on ultrasound examination. The mild leg edema was therefore attributed to her remaining sedentary for large portions of the day due to her advanced age.

View Full Size

Flash glucose monitoring (FGM) using the FreeStyle Libre Pro system. FGM was performed before (A) and after (B) diazoxide initiation. (A) Hypoglycemia was observed in both fasting and postprandial states. (B) Diazoxide prevented the recurrence of hypoglycemia.

Citation: Endocrinology, Diabetes & Metabolism Case Reports 2024, 3; 10.1530/EDM-24-0062

• Download Figure
• Download figure as PowerPoint slide

Treatment

As the patient was not a candidate for surgery owing to her advanced age, treatment with diazoxide at 200 mg/day (3 mg/kg/day) was initiated to prevent severe hypoglycemia.

Outcome and follow-up

Hypoglycemia disappeared after starting diazoxide administration (Fig. 2). However, diazoxide induced weight gain and exacerbation of leg edema and was discontinued after 2 months. She was given supplemental food between meals and before bedtime, and while this approach reduced the degree of hypoglycemia, episodes were not completely prevented. Interestingly, elevated levels of anti-insulin antibodies persisted for almost 1 year after the diagnosis of insulinoma (Table 2).

Discussion

We have presented here a rare case of insulinoma with the concomitant presence of anti-insulin antibodies. Insulinoma is a rare neuroendocrine tumor of the pancreas that produces excess endogenous insulin, resulting in hypoglycemia (1). In the present case, plasma concentrations of glucose (43 mg/dL, 2.4 mmol/L) and serum concentrations of insulin (21.1 µU/mL), proinsulin (16.9 pmol/L), and C-peptide (2.72 ng/mL) during a prolonged fasting test satisfied the criteria for spontaneous hyperinsulinemia proposed by the United States Endocrine Society (6). The calculated insulin/C-peptide molar ratio was 0.14, and a ratio < 1 is considered suggestive of insulinoma (4). Imaging modalities such as CT, magnetic resonance imaging, and endoscopic ultrasonography can help localize insulinomas, and CT in the present case revealed a 14 mm hypervascular tumor in the tail of the pancreas, consistent with insulinoma (2). Based on these findings, insulinoma was diagnosed.

IAS is characterized by recurrent episodes of spontaneous hypoglycemia, the presence of serum anti-insulin autoantibodies, significantly elevated serum levels of insulin, and no history of exposure to exogenous insulin. IAS is more common in Asian populations than in populations from Western countries, and a recent study estimated the prevalence of IAS as 0.017 cases per 100,000 per capita in the Japanese general population (3). Although many drugs are known to cause IAS, including methimazole and alpha lipoic acid (4), the present patient stated that she had not taken any of these drugs in the past. Moreover, despite not receiving any of these drugs during the clinical course in our hospital, serum levels of anti-insulin antibodies remained elevated almost 1 year after the insulinoma was diagnosed. Serum anti-insulin antibodies from patients with IAS have shown low affinity/high binding capacity against insulin by Scatchard analysis (4). We therefore tried to characterize the serum anti-insulin antibodies, but Scatchard analysis failed owing to the low titer of antibodies. Although previous cases of Japanese IAS patients have shown HLA-DRB1*0403, HLA-DRB1*0406, and HLA-DRB1*0407 alleles, HLA-DRB1*0406 has been exclusively associated with increased susceptibility to IAS (7). Moreover, HLA-DRB1*0406 is more frequent in East Asian patients than in non-East Asian patients (8). The present patient, however, did not show the HLA-DRB1*0406 allele.

The present case involved insulinoma in the pancreas and serum anti-insulin antibodies, both of which can cause hypoglycemia. Whereas serum insulin concentration in IAS patients often exceeds 1000 μU/mL (4), the present patient showed a markedly lower insulin concentration (21.1 μU/mL). Moreover, the insulin-to-C-peptide molar ratio during hypoglycemia was 0.14 in the present case, and any ratio less than 1.0 is indicative of insulinoma (4). Only one previous case report written in English appears to have described insulinoma concomitant with serum anti-insulin antibodies (5), although a few similar cases were reported in Japanese at academic Japanese conferences. Similar to the present case, the insulin concentration in that case was 106 μU/mL and the insulin-to-C-peptide molar ratio was 0.20. Radiofrequency ablation of the insulinoma successfully prevented further hypoglycemic episodes in that case, suggesting that the hypoglycemia was primarily caused by the insulinoma rather than by serum anti-insulin antibodies. We assumed that the episodes of hypoglycemia in the present case were likewise principally induced by the insulinoma.

In patients with diabetes mellitus treated using insulin therapy, exogenous insulin is modified and becomes immunogenic, resulting in the production of anti-insulin antibodies. In contrast, anti-insulin antibodies in IAS develop via interactions between a genetic predisposition and environmental triggers such as some types of drugs. These two types of antibodies show differing affinities and binding capacities against insulin. The patient in the present case had no history of exogenous insulin injection or exposure to sulfhydryl-containing drugs. Further, the patient did not have the HLA-DRB1*0406 allele, which is characteristic of IAS in Asian patients. We thus speculate that anti-insulin antibodies in the present case were not produced by the mechanisms seen with these two types of anti-insulin antibodies, although the precise mechanisms leading to the production of antibodies remain unknown in the present case. Hypoglycemia in insulinoma typically develops during fasting, whereas anti-insulin antibodies in IAS cause hypoglycemia after meals. Thus, in cases of insulinoma with plasma anti-insulin antibodies, the involvement of anti-insulin antibodies would be suspected when insulin concentrations are extremely high (> 1000 pmol/L) during hypoglycemia and hypoglycemia is more pronounced postprandially than in the fasting state.

Surgical removal is a fundamental treatment for insulinoma, and surgery should be performed where possible once the diagnosis is confirmed (2). However, when the patient is not a surgical candidate or chooses not to undergo surgery, alternative pharmacotherapies such as diazoxide, somatostatin analogs (e.g. octreotide), and mTOR inhibitors (e.g. everolimus) should prevent severe hypoglycemia (2). Diazoxide was used in the present case because the advanced age and physical frailty of our patient were considered to contraindicate surgery. Diazoxide inhibits insulin secretion from pancreatic beta cells by opening ATP-sensitive K channels and reducing the opening of calcium channels, resulting in a potent hyperglycemic effect (2). Some reports have already described patients with insulinoma for whom FGM has proven helpful for corroborating hypoglycemia induced by insulinoma and monitoring treatment outcomes after diazoxide initiation (9, 10). In the present case, the FSL-Pro clearly demonstrated that diazoxide suppressed hypoglycemia in both fasting and postprandial states (Fig. 2). However, diazoxide administration induced weight gain and exacerbation of peripheral edema and was discontinued after 2 months. The patient subsequently tried to prevent severe hypoglycemia by eating supplemental food between meals and before bedtime.

In conclusion, we have presented the details of a rare case of insulinoma concomitant with serum anti-insulin antibodies. Serum insulin concentrations in the patient during episodes of hypoglycemia were much lower than reported from typical IAS patients. Serum levels of anti-insulin antibodies remained elevated during the clinical course. An FGM system demonstrated the preventive effects of diazoxide on hypoglycemia. These findings may prove helpful for physicians encountering similar cases.

Declaration of interest

The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the study reported.

Funding statement

This work did not receive any specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Patient consent

Written informed consent was obtained from the patient for publication of the case report.

Author contribution statement

RN, DS, HI, JI, KI, MF, and HY were involved in the clinical care of the patient. RN and HY wrote the manuscript.