Researchers have found a non-invasive way to identify the exact position of very small insulinomas – life-threatening tumours of the pancreas – thereby enabling surgeons to operate successfully to remove growths that can be less than one centimetre in diameter.

Presenting the findings at the joint International Congress of Endocrinology/European Congress of Endocrinology, Professor Emanuel Christ, a clinical researcher in the Department of Endocrinology at the University Hospital of Bern, Switzerland, said that at present surgical removal of insulinomas was hampered by difficulties in localising them using conventional imaging procedures.

However, he and his colleagues had found that the insulinomas had high densities of a particular type of receptor on their cell surfaces that could be targeted with a radioactively labelled drug.

When patients were then scanned by a 2-D/3-D imaging technique called Single Photon Emission Computed Tomography (SPECT) combined with conventional computerised tomography (CT) scan, this technique allowed the researchers to detect where the drug had bound to insulinoma cells, indicating the precise location of the tumours, even when they were as small as about one centimetre in diameter.

“These data suggest that it’s possible to detect very small, life-threatening insulinomas within the pancreas, based on the characteristic receptors on the surface of these tumours,” said Prof Christ.

“This technique avoids the more invasive tests to localise the insulinomas and facilitates the surgical approach. Surgery is still the only method of curing this particular disease.”

Insulinomas are very rare, with an incidence of between two and four cases per million people a year. They are known as secreting neuroendocrine tumours (NET) as they are pancreatic tumours that secrete insulin. Insulinomas are derived from beta cells, which, under normal circumstances, secrete insulin in response to increases in blood glucose in order to reduce the blood glucose to normal levels. However, insulinomas secrete insulin in an unregulated way, causing blood glucose to fall lower than normal.

“In the majority of cases these tumours are benign, but they are life-threatening because they can lead to severe hypoglycaemia – low blood glucose,” explained Prof Christ.

“In contrast to other NETs, insulinomas do not exhibit receptors for a hormone called somatostatin in sufficient number and density to enable them to be detected by the well-established Octreoscan – a method of scanning that targets somatostatin receptors in NETs. However, they do have high numbers and density of another receptor for a hormone called glucagon-like peptide-1 (GLP-1). In collaboration with researchers in basic science and radiochemistry, we have developed a compound that successfully targets the GLP-1 receptors in humans – a radioactively labelled drug called 111In-exendin-4 – and which can be detected using SPECT/CT.”

The researchers injected the drug into 30 patients (18 men, 12 women) and then scanned them after 30 minutes, four, 23 and 96 hours, and up to 168 hours after the injection.

The most important time points were four and 23 hours after injection. Whereas conventional imaging (magnetic resonance imaging (MRI), computerised tomography (CT) and endosonography) detected the tumours in 17 patients, SPECT imaging with 111In-exendin-4 correctly detected 23 benign insulinomas, one malignant insulinoma, two islets hyperplasia (groups of proliferating cells) and two uncharacterised lesions.

There were no false negative results. In addition, the technique also detected another malignant insulinoma and an islets hyperplasia, which – although classified as true negative results because they did not meet the strict specification for a positive result – were important clinically as they also indicated over-production of insulin. The positive predictive value of the test (the proportion of correctly identified positive results) was 85% and the negative predictive value (proportion of patients who were correctly identified as not having the disease) was 100%.

After the scans, the tumours were removed surgically and the diagnosis confirmed when pathologists examined the tissues.

Prof Christ said: “This suggests that when clinical and biochemical tests have proved that unregulated over-production of insulin is occurring, but there is a lack of evidence for insulinomas using MRI, CT or endosonography, then GLP-1 receptor imaging could replace the more invasive approach in this situation.”

Current invasive approaches to detect small insulinomas include inserting catheters into the arteries and veins around the pancreas, injecting calcium gluconate into the arteries supplying the pancreas and taking samples from veins to test for insulin – a process known as arterial stimulation with venous sampling (ASVS).

Prof Christ and his colleagues now intend to see if positron emission tomography (PET), another 2-D/3-D nuclear medicine imaging technique, could improve the quality of tumour localisation even further.

Source: European Society of Endocrinology