Congenital hyperinsulinism (CHI) has two main histological types: diffuse and focal. Heterozygous paternally inherited ABCC8/KCNJ11 mutations (depending upon whether recessive or dominant acting and occurrence of somatic maternal allele loss) can give rise to either phenotype. However, the relative proportion of these two phenotypes in a large cohort of CHI patients due to paternally inherited heterozygous ABCC8/KCNJ11 mutations has not been reported.
The purpose of this study is to highlight the variable clinical phenotype and to characterise the distribution of diffuse and focal disease in a large cohort of CHI patients due to paternally inherited heterozygous ABCC8/KCNJ11 mutations.
A retrospective chart review of the CHI patients due to heterozygous paternally inherited ABCC8/KCNJ11 mutations from 2000 to 2013 was conducted.
Paternally inherited heterozygous ABCC8/KCNJ11 mutations were identified in 53 CHI patients. Of these, 18 (34%) either responded to diazoxide or resolved spontaneously. Fluorine-18 l-3, 4-dihydroxyphenylalanine positron emission tomography computerised tomography (18F DOPA–PET CT) scanning in 3/18 children showed diffuse disease. The remaining 35 (66%) diazoxide-unresponsive children either had pancreatic venous sampling (n=8) or 18F DOPA–PET CT (n=27). Diffuse, indeterminate and focal disease was identified in 13, 1 and 21 patients respectively. Two patients with suspected diffuse disease were identified to have focal disease on histology.
Paternally inherited heterozygous ABCC8/KCNJ11 mutations can manifest as a wide spectrum of CHI with variable 18F DOPA–PET CT/histological findings and clinical outcomes. Focal disease was histologically confirmed in 24/53 (45%) of CHI patients with paternally inherited heterozygous ABCC8/KCNJ11 mutations.
Congenital hyperinsulinism (CHI) is one of the main causes of hypoglycaemia and is characterised by inappropriate insulin secretion (1). Mutations in nine different genes (ABCC8, KCNJ11, GLUD1, GCK, HADH, SLC16A1, UCP2, HNF4A and HNF1A) have been reported so far as the genetic causes of CHI (1, 2). The pancreatic β-cell ATP-sensitive potassium channel (KATP channel) regulates glucose-mediated insulin release and is composed of two subunits: Kir6.2 encoded by KCNJ11 and SUR1 encoded by ABCC8 gene (2). Both genes are localised in the 11p15.1 region and mutations in these accounts for the majority of CHI patients (3, 4, 5, 6, 7, 8, 9, 10, 11).
There are two main forms of CHI (focal and diffuse) that are clinically identical but differ in histology, underlying genetic mechanism and management. In diffuse CHI, histologically there is an increase in the β-cell nuclear size throughout the pancreas. Diffuse CHI is genetically heterogeneous and most commonly due to mutations in ABCC8 or KCNJ11 genes (3, 12, 13, 14, 15). Diazoxide, an agonist that targets the SUR1 subunit, is usually ineffective in patients with autosomal recessive form of ABCC8 or KCNJ11 mutations. The management of the medically unresponsive diffuse CHI patients involves near-total pancreatectomy.
The focal lesions are characterised by nodular hyperplasia of islet-like cell clusters with ductuloinsular complexes and scattered giant β-cell nuclei with normal surrounding tissue (16). The focal CHI results from a paternally inherited heterozygous ABCC8/KCNJ11 mutation together with a somatic loss of the maternal chromosome in the 11p15 region (most likely caused by paternal isodisomy) (16, 17). The resulting loss of heterozygosity (LOH) renders the β-cells biallelic for the abnormal foci, altering the KATP channel and resulting in dysregulated insulin secretion within the focal lesion (16, 17, 18). The consequent imbalance in the expression of adjacent imprinted genes implicated in cell proliferation (such as CDKN1C and H19 normally expressed from the maternal allele and IGF2 paternally expressed) within the 11p15.5 region leads to focal islet cell adenomatous hyperplasia (19, 20, 21, 22). Focal CHI is usually medically unresponsive, although diazoxide-responsive focal CHI has been recently described (23). Targeted surgical removal of the lesion will cure the patient.
Fluorine-18 l-3, 4-dihydroxyphenylalanine positron emission tomography computerised tomography (18F DOPA–PET CT) scan can help to differentiate the focal from diffuse forms of CHI and aids in the clinical management of these patients (24). Before the advent of 18F DOPA–PET CT, pancreatic venous sampling (PVS) and selective pancreatic arterial calcium stimulation with hepatic venous sampling (ASVS) were used to differentiate between the two subtypes (25, 26). A recent systematic review and meta-analysis has found 18F DOPA–PET CT to be far superior to PVS and ASVS in diagnosing and localising focal CHI (27).
Patients with CHI associated with heterozygous paternally inherited ABCC8/KCNJ11 mutations may or may not have the second hit of somatic maternal allele loss in the pancreatic β-cell. Accordingly, these patients can either have focal or diffuse disease. Furthermore, diffuse disease can either be diazoxide-responsive or unresponsive depending on the underlying molecular basis of the mutation and other unclear genetic or environmental influences (28). There are isolated case reports of heterozygous paternally inherited ABCC8/KCNJ11 mutations leading to diffuse CHI in the literature (3, 10, 29, 30). The proportion of patients with paternally inherited heterozygous ABCC8/KCNJ11 mutations who develop focal CHI has not been reported in any study so far. Herein, we describe the heterogeneous clinical presentation and the histological basis of a relatively large cohort of patients with heterozygous paternally inherited ABCC8/KCNJ11 mutations. We also report the proportion of focal and diffuse disease in a large cohort of CHI patients with paternally inherited heterozygous ABCC8/KCNJ11 mutations.
Subjects and methods
The study was approved by the Ethics Committee of Great Ormond Street Children's Hospital and the Institute of Child Health. Informed written consent was obtained from the parents of children enrolled for molecular genetic testing for CHI (ABCC8/KCNJ11 sequencing).
This is a descriptive study of the clinical characteristics of children with CHI due to paternally inherited heterozygous ABCC8/KCNJ11 mutations. Clinical presentation, disease course and outcome for consecutive patients with CHI associated with paternally inherited heterozygous ABCC8/KCNJ11 mutations between the years 2000 and 2013 were retrospectively reviewed.
CHI was diagnosed by a controlled fasting test that demonstrated an inappropriately detectable serum insulin and/or C-peptide concentration and/or inappropriately low β-hydroxybutyrate and non-esterified fatty acids concentrations in the presence of hypoglycaemia (plasma glucose concentration <3.0 mmol/l). The patients with confirmed diagnosis of CHI underwent molecular genetic analysis after informed consent (see below). Diazoxide (5–15 mg/kg per day in three divided doses) was commenced as the first-line medical treatment. The patients were defined as diazoxide unresponsive if age appropriate fasting tolerance could not be achieved by treatment with 15 mg/kg per day diazoxide for a minimum of 5 days. Diazoxide-unresponsive patients associated with paternally inherited heterozygous ABCC8/KCNJ11 mutation underwent further investigations (PVS or 18F DOPA–PET CT) to differentiate between focal and diffuse disease. PVS was performed and interpreted as described previously by de Lonlay-Debeney et al. (26). The detailed protocol used for 18F DOPA–PET CT and analysis of the images has already been described (31).
Diazoxide-unresponsive diffuse disease was managed by octreotide injections (5–30 μg/kg per day in three to four divided doses)±near-total pancreatectomy along with carbohydrate-rich feeds. The focal disease was managed with the resection of the focal lesion or partial pancreatectomy. Post-surgery, glycaemic control was periodically assessed by 24-h blood glucose profile and controlled fast to ensure cure of the focal CHI and appropriate control of diffuse CHI.
Genomic DNA was extracted from peripheral leukocytes using standard procedures. The single-coding exon of the KCNJ11 gene and the 39 exons of the ABCC8 gene were amplified using the PCR. Unidirectional sequencing was performed using universal M13 primers and a Big Dye Terminator Cycle Sequencing Kit v3.1 (Applied Biosystems) according to the manufacturer's instructions. The reactions were analysed on an ABI 3730 Capillary Sequencer (Applied Biosystems) and the sequences were compared with the reference sequences (NM_000525 and NM_000352.2) using Mutation Surveyor v3.24 (SoftGenetics, PA, USA). Dosage analyses by multiplex ligation-dependent probe amplification were done in patients with a heterozygous ABCC8/KCNJ11 mutation and diffuse disease on 18F DOPA–PET scan/PVS (32).
LOH was investigated by microsatellite analysis of DNA extracted from paraffin-embedded pancreatic tissue and peripheral leukocytes, mainly when there was a discrepancy between the histology and 18F DOPA–PET CT/PVS result. Six markers (D11S2071, D11S1964, D11S419, D11S1397, D11S1888 and D11S4138) spanning chromosome 11p15.1–11p15.5 were amplified by PCR and allele peak heights were compared using GeneMarker v1.85 (SoftGenetics).
Over a period of 14 years (2000–2013), more than 300 children with confirmed biochemical and genetic diagnosis of CHI were managed in our centre. Of these, 53 children had paternally inherited heterozygous ABCC8/KCNJ11 mutation (ABCC8, 42 and KCNJ11, 11). Of the 35 different ABCC8 mutations seen in 42 patients, 19 affected a single amino acid (missense or in-frame deletion/insertion). Of the ten different KCNJ11 mutations seen in 11 patients, eight affected a single amino acid (missense or in-frame deletion/insertion), one was a frame shift mutation and one was a non-stop codon mutation resulting in extension of the Kir6.2 protein sequence by three amino acids (Fig. 1). As shown in Fig. 1, mutations occurred throughout SUR1 and Kir6.2.
Clinical and biochemical characteristics
The mean (±2 s.d.) gestational age and birth weight of our cohort was 38.9 (±1.8) weeks and 3980 (±760) g respectively. Of these, 19 infants (36%) were macrosomic (birth weight >2 s.d.) at birth. The gender distribution was 37 males and 16 females. The majority of patients in our cohort were of Caucasian descent. Only one proband was a product of the consanguineous marriage. There were two pairs of two siblings and the rest of the cohort was unrelated. The median age of presentation with hypoglycaemia was <1 week (range <1–130 weeks). More than 85% (46/53) patients presented with hypoglycaemia within first week of life. The mean (±s.d.) serum insulin during hypoglycaemia was 21.1 (±34.2) mU/l. In three infants in whom serum insulin was undetectable, C-peptide level was inappropriately elevated for the blood glucose measurement. The serum ammonia was within normal range and β-hydroxybutyrate/non-esterified fatty acids were inappropriately low in all children (Table 1).
Clinical and biochemical characteristics, mutational analysis and clinical outcome of 53 patients with paternally inherited ABCC8/KCNJ11 mutations.
|Patient ID||GA (weeks), birth weight (g)||Gender||Age at presentation (weeks)||Blood glucose (mmol/l)||Serum insulin (mU/l)||Mutation protein description (DNA description)||LOH||Dzx Resp||PET CT/PVS||Outcome|
|1||40, 3150||Male||52||2.4||4.3||L1431F/N (c.4291C>T/N)||Yes||–||On Dzx at 6.4 years|
|2||40, 4000||Male||2||2.4||1.9||p.?/N (c.2697+4A>T/N)||Yes||–||Off Dzx at 2 years|
|3||40, 5010||Female||<1||2.6||8.6||E1507K/N (c.4519G>A/N)||Yes||–||On Dzx at 2.3 years|
|4||40, 5600||Male||<1||2.0||7.5||A1508P/N (c.4522G>C/N)||Yes||–||On Dzx at 12 years|
|5||37, 4820||Male||<1||2.0||9.0||A1153T/N (c.3457G>A/N)||Yes||–||On Dzx at 4 years|
|6||38, 3630||Female||72||3.1||<2||A1153T/N (c.3457G>A/N)||Yes||–||On Dzx at 2 years|
|8||35, 2820||Male||<1||1.2||12.9||A1185V/N (c.3554C>T/N)||Yes||–||Off Dzx at 8 months|
|9||36, 4450||Male||<1||2||28.5||p.?/N (c.3992-9G>A/N)||Yes||–||Off Dzx after 4.5 months|
|10||41, 2780||Female||<1||1||9.3||D1472N/N (c.4414G>A/N)||Yes||–||Off Dzx after 10 months|
|11||38, 3750||Male||<1||2.6||5.90||V601I/N (c.1801G>A/N)||Yes||–||Off Dzx after 14 months|
|13||40, 4160||Female||<1||2.4||14.8||V185fs/N (c.554delT/N)||No||Diffuse||Off octreotide at 5 years|
|14||37, 3090||Male||<1||0.6||12.4||p.?/N (c.3992-9G>A/N)||No||Focal||On octreotide at 9.5 years|
|15||40, 3600||Male||<1||2.7||6.7||H627fs/N (c.1879delC/N)||No||Diffuse||Off octreotide at 18 months|
|16||40, 4700||Female||<1||2.0||<2||E1507K/N (c.4519G>A/N)||NA||–||No treatment required|
|17||40, 4200||Male||<1||1.0||<2||D1031N/N (c.3091G>A/N)||NA||–||No treatment required|
|18||41, 4850||Male||<1||1.2||10.1||M1V/N (c.1A>G/N)||No||Diffuse (PVS)||Near-total pancreatectomy (95%)|
|19||38, 2400||Male||<1||2.1||16.3||D1194V; R1437Q/N (c.3581A>T; c.4310G>A/N)||Yes||No||Focal||Hypoglycaemia resolved after removal of focal lesion|
|20||40, 4580||Female||<1||1.1||103||A1493T/N (c.4477G>A/N)||Yesa||No||Diffuse||Near-total pancreatectomy (95%)|
|21||40, 4600||Male||<1||1.2||22.5||K890fs/N (c.2669_2675del/N)||Yes||No||Focal||Hypoglycaemia resolved after removal of focal lesion|
|22||40, 4335||Male||<1||2.6||3.4||p.?/N (c.3992-9G>A/N)||Yes||No||Focal||Partial pancreatectomy|
|23||40, 3030||Male||<1||1.8||15.6||p.?/N (c.3992-9G>A/N)||Yes||No||Focal||Partial pancreatectomy|
|24||40, 2770||Male||<1||2.2||3.4||p.?/N (c.580-1G>C/N)||No||Indeterminate (PVS)||Partial pancreatectomy – focal lesion on histology|
|25||41, 4290||Male||2||2.3||4.32||E128K/N (c.382G>A/N)||No||Focal||Hypoglycaemia resolved after removal of focal lesion|
|27||40, 5095||Male||<1||2.0||10.5||L1171X/N (c.3512delT/N)||Yes||No||Focal||Hypoglycaemia resolved after removal of focal lesion|
|29||37, 3560||Male||<1||1.6||21.8||p.?/N (c.1629-2A>C/N)||No||Focal||Hypoglycaemia resolved after removal of focal lesion|
|30||40, 2750||Male||<1||1.5||16.4||G111R/N (c.331G>A/N)||Yes||No||Focal||Hypoglycaemia resolved after removal of focal lesion|
|31||37, 3340||Male||<1||2.1||15||H627fs/N (c.1879delC/N)||Yes||No||Focal||Hypoglycaemia resolved after removal of focal lesion|
|32||41, 4950||Male||<1||1.2||11.95||R934X/N (c.2800C>T/N)||No||Focal||Hypoglycaemia resolved after removal of focal lesion|
|33||40, 3080||Female||12||1.5||4.6||S12X/N (c.35C>A/N)||Yes||No||Focal||Hypoglycaemia resolved after removal of focal lesion|
|34||39, 3600||Male||<1||2.5||8||R1494W/N (c.4480C>T/N)||No||Diffuse (PVS)||Near-total pancreatectomy (95%)|
|37||36, 3410||Male||<1||0.9||17.42||R1494W/N (c.4480C>T/N)||No||Focal||Partial pancreatectomy|
|38||39, 4900||Female||<1||1.4||23.61||A113V/N (c.338C>T/N)||No||Diffuse (PVS)||Near-total pancreatectomy (95%)|
|39||36, 3210||Male||<1||0.6||114||Mosaic Q54X (c.160C>T)||No||Diffuse||Near-total pancreatectomy (95%)|
|40||41, 4300||Female||<1||2.1||17||Q954X/N (c.2860C>T/N)||No||Diffuse||Near-total pancreatectomy (95%)|
|41||36, 2730||Male||<1||?||?||p.?/N (c.1333-1013A>G/N)||Yes||No||Focal||Hypoglycaemia resolved after removal of focal lesion|
|42||37, 4480||Female||<1||1.7||5.6||A1263T/N (c.3787G>A/N)||No||Diffuse (PVS)||Near-total pancreatectomy (95%)|
|45||39, 3960||Male||<1||2.4||23.6||p.?/N (c.1-?_c.1176+1?/N)||No||Focal||Hypoglycaemia resolved after removal of focal lesion|
|46||34, 3800||Male||<1||0.8||201||D855E/N (c.2565C>A/N)||No||Focal||Hypoglycaemia resolved after removal of focal lesion|
|47||41, 4780||Male||<1||2.4||15.9||M429X/N (c.1254_1284dup/N)||No||Focal||Hypoglycaemia resolved after removal of focal lesion|
|48||40, 3800||Female||<1||2.4||5.9||M429X/N (c.1254_1284dup/N)||No||Diffuse||On octreotide at 7 months|
|49||40, 4100||Male||<1||0.8||9.1||R74W/N (c.220C>T/N)||No||Focal||Hypoglycaemia resolved after removal of focal lesion|
|51||38, 3730||Female||<1||1.9||21.6||E995X (c.2983G>T)||No||Focal||Hypoglycaemia resolved after removal of focal lesion|
|7||38, 3780||Female||<1||1.7||29.6||M209I/N (c.627G>A/N)||Yes||Diffuse||On Dzx at 2.8 years|
|12||40, 3450||Male||<1||2||6.7||R54H/N (c.161G>A/N)||Yes||–||Off Dzx after 4 months|
|26||36, 3880||Male||<1||<1.0||6.1||E292K/N (c.874G>A/N)||No||Diffuse (PVS)||Near-total pancreatectomy (95%)|
|28||40, 4600||Male||<1||1.3||38.87||R136fs/N (c.405dupG/N)||No||Diffuse (PVS)||Near-total pancreatectomy (95%)|
|35||38, 4120||Female||<1||2.2||12.0||T294M/N (c.881C>T/N)||No||Diffuse (PVS)||Near-total pancreatectomy (95%)|
|36||37, 3960||Male||<1||1.9||7.0||G312C/N (c.934G>T/N)||No||Focal||Hypoglycaemia resolved after removal of focal lesion|
|43||40, 4100||Female||<1||1.0||35||I284del/N (c.850_852delATC/N)||Yes||Diffuse||Off Dzx at 4 years of age|
|44||40, 4580||Female||32||2.2||7||p.*391Rext*94/N (c.1171T>C/N)||No||Focal||Partial pancreatectomy, currently on Dzx|
|50||37, 3980||Male||<1||0.5||2.9||T62M/N (c.185C>T/N)||Yes||–||Off Dzx at 1 month|
|52||40, 5190||Male||<1||2.5||8.4||R177W/N (c.529A>T/N)||Yes||Diffuse||On Dzx at 3 months of age|
|53||40, 3920||Male||<1||2.1||9.2||E292K/N (c.874G>A/N)||Yes||Diffuse||Off Dzx at 5 months of age|
Dzx, diazoxide; PET, positron emission tomography; PVS, pancreatic venous sampling; GA, gestational age; Wt, weight; LOH, loss of heterozygosity; Resp, responsiveness.
This patient had giant focal lesion and virtually occupied the whole of pancreas.
Clinical course, 18F DOPA–PET CT and PVS findings and final outcome
In two (4%) patients (Table 1: patients 16 and 17), CHI resolved spontaneously within few weeks before investigations were repeated in our centre (Fig. 2). Previous functional work has established milder disease phenotype with a mutation seen in one of these two patients (ABCC8 (E1507K)) (33). The mutation identified in the second patient (ABCC8 (D1031N)) is not reported in dbSNP and predicted to be disease causing (Table 2).
In silico pathogenicity prediction of diazoxide-responsive patients' mutations.
|Mutation||Prediction||Mutation previously reported|
|Found in dbSNP||Provean||SIFT||MutationTaster|
|L1431F (c.4291C>T)||No||Deleterious||Damaging||Disease causing||(38)a|
|p.? (c.2697+4A>T)||No||Not available||Not available||Disease causing||Novel|
|E1507K (c.4519G>A)||Yes||Deleterious||Damaging||Disease causing||(32)|
|A1508P (c.4522G>C)||No||Deleterious||Damaging||Disease causing||(38)a|
|A1153T (c.3457G>A)||No||Deleterious||Damaging||Disease causing||Novel|
|A1185V (c.3554C>T)||No||Neutral||Damaging||Disease causing||Novel|
|p.? (c.3992-9G>A)||No||Not available||Not available||Disease causing||(5)a|
|D1472N (c.4414G>A)||Yes||Deleterious||Damaging||Disease causing||(24)a|
|V601I (c.1801G>A)||No||Neutral||Tolerated||Disease causing||Novela|
|V185fs (c.554delT)||Yes||Not available||Not available||Disease causing||Novel|
|H627fs (c.1879delC)||No||Not available||Not available||Disease causing||(9)a|
|D1031N (c.3091G>A)||No||Neutral||Tolerated||Disease causing||Novel|
|M1V (c.1A>G)||No||Neutral||Damaging||Disease causing||(6)a|
|A1493T (c.4477G>A)||Yes||Deleterious||Damaging||Disease causing||(14)a|
|R1494W (c.4480C>T)||Yes||Deleterious||Damaging||Disease causing||(25)a|
|M209I (c.627G>A)||No||Neutral||Damaging||Disease causing||Novel|
|R54H (c.161G>A)||No||Deleterious||Damaging||Disease causing||Novela|
|E292K (c.874G>A)||No||Deleterious||Damaging||Disease causing||Novel|
|T62M (c.185C>T)||No||Deleterious||Damaging||Disease causing||(4)|
|R177W (c.529A>T)||No||Deleterious||Damaging||Disease causing||Novel|
These patients were included in reference (3).
Sixteen patients (30%) with presumably diffuse CHI responded to diazoxide treatment (Fig. 2). In three patients from diazoxide-responsive group who underwent 18F DOPA–PET CT, diffuse uptake throughout the pancreas was noticed. Nine patients (17%) successfully managed to come off diazoxide treatment after a variable period of time. The remaining seven (13%) patients were on diazoxide for a median duration of 2.8 years (range 3 months–12 years) at the time of writing.
The remaining 35 (66%) patients were unresponsive to maximum doses of diazoxide and were investigated either by using PVS (n=8) or 18F DOPA–PET CT scan (n=27).
PVS was suggestive of diffuse disease in seven patients and no differentiation between focal and diffuse was possible in one patient. The suspected diffuse disease patients were managed with near-total pancreatectomy. Although histology confirmed diffuse disease in six patients, there was one patient who had focal adenomatous hyperplasia which was presumably missed on PVS. The patient with indeterminate PVS underwent partial pancreatectomy and focal lesion was seen on the histology of the resected pancreas.
Of the 27 patients who were investigated with 18F DOPA–PET scan, focal uptake was seen in 21 and diffuse uptake in six patients (Fig. 3). The focal lesion was successfully managed either with lesionectomy (n=16) or partial pancreatectomy (n=4). One patient was managed successfully with octreotide injections.
The patients with diffuse tracer pancreatic uptake were either managed with near-total pancreatectomy (n=3) or octreotide injections (n=3). One patient from this subgroup had atypical histology with pancreatic β-cell enlargement in parts of body and head of the pancreas due to mosaic interstitial paternal uniparental isodisomy for chromosome 11p15.1 (36).
This study reports the proportion of focal disease in a large cohort of patients with CHI associated with paternally inherited heterozygous ABCC8/KCNJ11 mutations. In addition, it highlights the clinical and histological heterogeneity of CHI associated with paternally inherited heterozygous ABCC8/KCNJ11 mutation.
Mutations in the ABCC8/KCNJ11, which account for the majority of genetically confirmed CHI cases, can either be biallelic or monoallelic (3, 4, 37). Biallelic ABCC8/KCNJ11 mutations result in diffuse CHI, whereas monoallelic mutations can either be asymptomatic (recessive acting), or lead to focal CHI (if paternally inherited and associated with loss of somatic maternal 11p allele) or diffuse CHI (dominant acting). Diazoxide, a KATP channel opener, which is the first-line management for CHI, is usually ineffective in biallelic disease and monoallelic focal disease (3, 4). However, recently a case of diazoxide-responsive focal CHI was described (23). It may be possible that as more diazoxide-responsive CHI patients are investigated with 18F DOPA–PET CT scan, more diazoxide-responsive focal lesions are identified. However, currently diazoxide responsiveness in our institution is generally considered equivalent to excluding monoallelic focal disease and this group of patients are not usually considered as suitable candidates for further investigations with 18F DOPA–PET CT. On the other hand, diazoxide-unresponsive subgroup, particularly associated with paternally inherited monoallelic ABCC8/KCNJ11 mutations or no identified ABCC8/KCNJ11 mutations, is investigated with 18F DOPA–PET CT to identify focal or diffuse subtype before proceeding with definitive treatment as the treatment for focal (limited pancreatectomy) and diffuse CHI (near-total pancreatectomy) is drastically different (38).
In our cohort of 53 CHI patients associated with paternally inherited monoallelic ABCC8/KCNJ11 mutations, 18 (33%) patients either responded to diazoxide or resolved spontaneously and are likely to have dominant acting CHI. Mutations in these 18 patients are predicted to be pathogenic by different mutation prediction programmes (Table 2). Functional work has already established some of these mutants to be dominant acting and exerting their effect by a dominant-negative mechanism (33, 39, 40). Although there was no history of hypoglycaemia in the fathers of our cohort, formal evaluation with controlled provocation fasting studies has not been done. Other studies also have reported parents to be asymptomatic carriers of dominant acting mutations (41). In view of severe diazoxide side effects and parental request, three patients from this subgroup underwent 18F DOPA–PET CT imaging to avoid long-term treatment with diazoxide. This was in the hope that these patients could be managed with removal of focal lesion if found to be the underlying cause for CHI. However, as suspected from the phenotype of diazoxide-responsiveness, 18F DOPA–PET CT identified diffuse tracer uptake throughout the pancreas.
Evaluation with 18F DOPA–PET CT in 27 diazoxide-unresponsive patients with paternally inherited heterozygous ABCC8/KCNJ11 mutations identified focal disease in 21 (78%) patients. Majority of these patients presented within the first week of life. There was no difference in the age of presentation between those who were diagnosed with diffuse or focal CHI. Apart from one patient who was managed with octreotide therapy, these patients were either managed with resection of the focal lesion or partial pancreatectomy. Follow-up assessment highlighted age-appropriate fasting tolerance and resolution of hypoglycaemia in all except one (19/20; 95%) after the removal of the focal lesion. Similar to our results, Lord et al. (42) recently reported that 94% of their cohort of focal CHI after pancreatectomy remained euglycaemic and required no treatment for blood glucose abnormalities. In the cohort reported by Beltrand et al. (43), 91.4% (43/47) patients with focal CHI had complete resolution of hypoglycaemia post-surgery. One patient in whom the evidence of CHI persisted despite the removal of a histologically confirmed focal lesion was managed with diazoxide and regular feeds. A repeat 18F DOPA–PET CT revealed diffuse tracer uptake in the remaining pancreas. Molecular basis for CHI in this particular patient is unclear and further studies are in progress.
In six (21%) diazoxide-unresponsive patients investigated with 18F DOPA–PET CT, diffuse tracer uptake throughout the pancreas was noticed (Fig. 3). Three of these patients were managed with near-total pancreatectomy and in one of these patients, histology and microsatellite analysis was suggestive of giant focal lesion (44). Diffuse uptake in this particular patient was due to the size of the focal lesion, which occupied nearly the whole pancreas. Diffuse CHI has been described in the literature with paternally inherited ABCC8/KCNJ11 mutations. Fernandez-Marmiesse et al. (10) reported five patients with paternally inherited KATP mutations and whose post pancreatectomy pathology result was not consistent with that of focal lesion. Similar findings have been reported by other studies (1, 21, 27, 30).
There are a number of possible mechanisms as to how a heterozygous ABCC8/KCNJ11 mutation can lead to diffuse disease. For example there could be a post-zygotic second hit within the pancreas or a maternally inherited mutation may reside within a non-coding regulatory/intronic region of the ABCC8/KCNJ11 gene. Recently, in a patient with focal CHI but no exonic or splice site ABCC8/KCNJ11 mutation, next-generation sequencing identified a deep intronic ABCC8 mutation inherited from the unaffected father, which created a cryptic spice site and an out of frame pseudoexon (32). A third possibility is allelic drop-out due to a rare polymorphism within a primer binding site. In this scenario, a maternally inherited coding mutation could escape detection by Sanger sequencing. Besides, some of the missense mutations may be dominantly acting (dominant negative). As compared with recessive acting mutations where there is defective biogenesis or trafficking of KATP channel to the β-cell membrane, the dominant acting mutants traffic normally to the plasma membrane but show impaired responsiveness to the channel agonists, MgADP and diazoxide (45). Diazoxide-unresponsive mutations produce a more severe impairment in the expressed channel response to activation by diazoxide and MgADP as compared with diazoxide-responsive mutations (28).
A novel genetic mechanism of diffuse disease due to mosaic paternal uniparental isodisomy in a patient with heterozygous ABCC8/KCNJ11 mutations was recently described (36). Lastly, it could also very rarely be due to failure of Sanger sequencing to detect the second ABCC8/KCNJ11 mutation or a missed histological diagnosis of focal CHI in these patients. Although ABCC8 and KCNJ11 are not known to be imprinted genes, they are next to the 11p15.5 imprinted region, so another hypothesis could be the involvement of epigenetic mechanisms.
In summary, of the 53 patients associated with paternally inherited ABCC8/KCNJ11 mutations, 29 (55%) patients had diffuse CHI (18 (34%) patients – diazoxide-responsive; 11 (21%) patients – diazoxide-unresponsive) and 24 (45%) patients had confirmed focal CHI either based on histology or LOH studies. One patient had giant focal lesion based on histology and LOH studies, confusing the interpretation of 18F DOPA–PET CT. From our results, it is clear that paternally inherited heterozygous ABCC8/KCNJ11 mutations can manifest as a wide spectrum of CHI with focal disease in 45% patients. This heterogeneous clinical picture could be due to modifying genes or other unknown factors, e.g. environmental, that influence the expression of the phenotype in CHI patients.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
DarendelilerFFournetJCBasFJunienCGrossMSBundakRSakaNGunozH. ABCC8 (SUR1) and KCNJ11 (KIR6.2) mutations in persistent hyperinsulinemic hypoglycemia of infancy and evaluation of different therapeutic measures. Journal of Pediatric Endocrinology & Metabolism200215993–1000.
TornovskySCraneACosgroveKEHussainKLavieJHeymanMNesherYKuchinskiNBen-ShushanEShatzO. Hyperinsulinism of infancy: novel ABCC8 and KCNJ11 mutations and evidence for additional locus heterogeneity. Journal of Clinical Endocrinology and Metabolism2004896224–6234. (doi:10.1210/jc.2004-1233).
DamajLle LorchMVerkarreVWerlCHubertLNihoul-FeketeCAigrainYde KeyzerYRomanaSPBellanne-ChantelotC. Chromosome 11p15 paternal isodisomy in focal forms of neonatal hyperinsulinism. Journal of Clinical Endocrinology and Metabolism2008934941–4947. (doi:10.1210/jc.2008-0673).
VerkarreVFournetJCde LonlayPGross-MorandMSDevillersMRahierJBrunelleFRobertJJNihoul-FeketeCSaudubrayJM. Paternal mutation of the sulfonylurea receptor (SUR1) gene and maternal loss of 11p15 imprinted genes lead to persistent hyperinsulinism in focal adenomatous hyperplasia. Journal of Clinical Investigation19981021286–1291. (doi:10.1172/JCI4495).
CaltonEATempleIKMackayDJLeverMEllardSFlanaganSEDaviesJHHussainKGrayJC. Hepatoblastoma in a child with a paternally-inherited ABCC8 mutation and mosaic paternal uniparental disomy 11p causing focal congenital hyperinsulinism. European Journal of Medical Genetics201356114–117. (doi:10.1016/j.ejmg.2012.12.001).
ShumanCSmithACSteeleLRayPNClericuzioCZackaiEParisiMAMeadowsATKellyTTichauerD. Constitutional UPD for chromosome 11p15 in individuals with isolated hemihyperplasia is associated with high tumor risk and occurs following assisted reproductive technologies. American Journal of Medical Genetics. Part A20061401497–1503. (doi:10.1002/ajmg.a.31323).
de LonlayPFournetJCRahierJGross-MorandMSPoggi-TravertFFoussierVBonnefontJPBrussetMCBrunelleFRobertJJ. Somatic deletion of the imprinted 11p15 region in sporadic persistent hyperinsulinemic hypoglycemia of infancy is specific of focal adenomatous hyperplasia and endorses partial pancreatectomy. Journal of Clinical Investigation1997100802–807. (doi:10.1172/JCI119594).
MaioranaABarbettiFBoianiARufiniVPizzoferroMFrancalanciPFaletraFNicholsCGGrimaldiCde Ville de GoyetJ. Focal congenital hyperinsulinism managed by medical treatment: a diagnostic algorithm based on molecular genetic screening. Clinical Endocrinology2014In pressdoi:10.1111/cen.12400).
HardyOTHernandez-PampaloniMSafferJRScheuermannJSErnstLMFreifelderRZhuangHMacMullenCBeckerSAdzickNS. Accuracy of [18F]fluorodopa positron emission tomography for diagnosing and localizing focal congenital hyperinsulinism. Journal of Clinical Endocrinology and Metabolism2007924706–4711. (doi:10.1210/jc.2007-1637).
StanleyCAThorntonPSGangulyAMacMullenCUnderwoodPBhatiaPSteinkraussLWannerLKayeRRuchelliE. Preoperative evaluation of infants with focal or diffuse congenital hyperinsulinism by intravenous acute insulin response tests and selective pancreatic arterial calcium stimulation. Journal of Clinical Endocrinology and Metabolism200489288–296. (doi:10.1210/jc.2003-030965).
BlombergBAMoghbelMCSabouryBStanleyCAAlaviA. The value of radiologic interventions and (18)F-DOPA PET in diagnosing and localizing focal congenital hyperinsulinism: systematic review and meta-analysis. Molecular Imaging and Biology20131597–105. (doi:10.1007/s11307-012-0572-0).
Bellanne-ChantelotCSaint-MartinCRibeiroMJVauryCVerkarreVArnouxJBValayannopoulosVGobrechtSSempouxCRahierJ. ABCC8 and KCNJ11 molecular spectrum of 109 patients with diazoxide-unresponsive congenital hyperinsulinism. Journal of Medical Genetics201047752–759. (doi:10.1136/jmg.2009.075416).
MohnikeKWielandIBarthlenWVogelgesangSEmptingSMohnikeWMeissnerTZenkerM. Clinical and genetic evaluation of patients with K channel mutations from the German registry for congenital hyperinsulinism. Hormone Research in Paediatrics201481156–168. (doi:10.1159/000356905).
MeintjesMEndozoRDicksonJErlandssonKHussainKTownsendCMenezesLBomanjiJ. 18F-DOPA PET enhanced CT imaging for congenital hyperinsulinism: initial UK experience from a technologist's perspective. Nuclear Medicine Communications201334601–608. (doi:10.1097/MNM.0b013e32836069d0).
FlanaganSEXieWCaswellRDamhuisAVianey-SabanCAkcayTDarendelilerFBasFGuvenASiklarZ. Next-generation sequencing reveals deep intronic cryptic ABCC8 and HADH splicing founder mutations causing hyperinsulinism by pseudoexon activation. American Journal of Human Genetics201392131–136. (doi:10.1016/j.ajhg.2012.11.017).
BanerjeeISkaeMFlanaganSERigbyLPatelLDidiMBlairJEhtishamSEllardSCosgroveKE. The contribution of rapid KATP channel gene mutation analysis to the clinical management of children with congenital hyperinsulinism. European Journal of Endocrinology2011164733–740. (doi:10.1530/EJE-10-1136).
ThorntonPSMacMullenCGangulyARuchelliESteinkraussLCraneAAguilar-BryanLStanleyCA. Clinical and molecular characterization of a dominant form of congenital hyperinsulinism caused by a mutation in the high-affinity sulfonylurea receptor. Diabetes2003522403–2410. (doi:10.2337/diabetes.52.9.2403).
PinneySEMacMullenCBeckerSLinYWHannaCThorntonPGangulyAShyngSLStanleyCA. Clinical characteristics and biochemical mechanisms of congenital hyperinsulinism associated with dominant KATP channel mutations. Journal of Clinical Investigation20081182877–2886. (doi:10.1172/JCI35414).
(V B Arya and M Guemes contributed equally to this work)