| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Editorial |
Head of Pediatric Endocrinology Assistance Publique-Hôpitaux de Paris Hôpital Necker-Enfants Malades Université Paris-Descartes Faculté de Medicine Institut National de la Santé et de la Recherche Médicale U845 Paris, France
Address all correspondence and requests for reprints to: Michel Polak, Head of Pediatric Endocrinology, Assistance Publique-Hôpitaux de Paris, Hôpital Necker-Enfants Malades, 149 rue de Sèvres, Paris, France. E-mail: michel.polak{at}nck.aphp.fr.
Neonatal diabetes mellitus (NDM) is a rare (around 1 in 300,000 newborns) but potentially devastating condition. Two main groups have been recognized on clinical grounds (transient NDM and permanent NDM), which differ in the duration of insulin dependence early in the disease (1).
Transient NDM is a developmental disorder of insulin production that resolves during the postnatal period. Transient NDM contributes 50 to 60% of cases of neonatal diabetes (1, 2). Intrauterine growth retardation is usually present. The high rate of intrauterine growth retardation is in keeping with the crucial role of insulin in fetal growth, especially during the last trimester of pregnancy. Hyperglycaemia, failure to thrive, and, in some cases, dehydration occur after birth. Insulin production is inadequate, requiring exogenous insulin therapy. A defect in cell maturation has been suggested; however, the cellular basis of transient NDM remains unknown. Most patients recover within 1 yr, but a few have persistent glucose intolerance and/or recurrence of diabetes in late childhood or adulthood, and therefore deserve long-term follow-up (3, 4).
Although these recurrences are usually consistent with nonautoimmune type 1 diabetes, whether they are ascribable to insulin deficiency and/or insulin resistance remains unclear (5).
Permanent NDM in our experience is less common than the transient form of the condition. By definition, diabetes develops in the neonatal period, never to go into remission. There are no clinical features that can predict whether a neonate with diabetes but no other dysmorphic features will eventually have permanent or transient disease.
Recently, advances have been made in the understanding of the molecular mechanisms of pancreatic development and ß-cell function that are relevant to permanent and to transient NDM (summarized in Ref. 1).
Diabetes in infancy is nearly always unrelated to classical type 1 autoimmune diabetes (1, 6, 7). However, it may be very rarely due to precocious diffuse autoimmunity such as in the IPEX syndrome (related to mutation in the FOXP3 gene) (1). More than half of transient NDM cases are associated with abnormalities of an imprinted region on chromosome 6q24. A small number of permanent cases are attributed to mutations in the glucokinase (GCK) (leading to a total insensitivity of the ß-cell to glucose). Permanent cases can also be related to pancreatic agenesis/hypoplasia due to mutations in crucially important transcription factors for pancreatic development such insulin promoter factor-1 (IPF-1), and promoter transcription factor 1 (PTF1a, when associated with cerebellar hypoplasia). Mutations in the eukaryotic translation initiation factor-2 kinase3 gene (EIF2AK3) have been implicated in Wolcott-Rallison syndrome, an autosomal recessive disorder characterized by infancy-onset (often within the neonatal period) diabetes associated with a spondyloepiphyseal dysplasia, where both ß-cell development and survival are altered.
Knowing the key role of the KATP channels [heterooctamers assembled from KIR6.2 (KCNJ11) and the high-affinity sulfonylurea receptor, SUR1 (ABCC8)] in insulin secretion, several studies reported sequencing of KCNJ11, in patients with permanent NDM without known etiology and found different mutations for 30 to 50% of the cases (8, 9, 10). Mutations in KCNJ11 were also identified but with less frequency in transient neonatal diabetes cases (11). Most of the mutations described (approximately 80%; Ref. 8) are de novo but cases have been reported with a demonstrated germline mosaicism, suggesting that this possibility should be considered when counseling recurrence risk (12). These mutations resulted in reduced ATP-sensitivity of the KATP channels compared with the wild types, and the level of channel block was responsible for different clinical features: the "mild" form conferring isolated permanent neonatal diabetes whereas the severe form combining diabetes and neurological symptoms such as epilepsy, developmental delay, muscle weakness, and mild dysmorphic features (8, 9, 10, 13). The muscle weakness observed can have both a muscular and neurological basis because Kir6.2 is expressed in both cells. The developmental delay and the epilepsy result from defective KATP channels present in the central nervous system.
We identified activating mutations in ABCC8, which encodes SUR1, as a cause of neonatal diabetes (14). Of interest, some mutations were familial and showed vertical transmission with neonatal and adult-onset diabetes. Diabetes resulted from a novel mechanism whereby the basal Mg-nucleotide-dependent stimulatory action of SUR1 on the KIR pore is elevated, impairing insulin secretion, with no reduced ATP sensitivity (14). Further cases with SUR1 mutations are being described (15, 16, 17). In the French network, most of the mutations in SUR1 were associated with transient NDM, whereas Kir6.2 mutations were with early onset (before 6 months) permanent (17). The phenotypic variability of SUR1 mutations is broader (14). Indeed, mutations in SUR1 can be linked to ketoacidosis in a newborn as well as to bona fide type 2 diabetes in a young adult (Ref. 14 and supplementary information). In the English case series with transient NDM, mutations in SUR1 explained as many cases as mutations in Kir6.2 (16). This discrepancy deserves further exploration.
In the paper by Suzuki et al. (18), the molecular basis of neonatal diabetes in a large cohort of 31 Japanese patients is provided. It represents a large number of patients and an interesting study. It teaches us that the molecular basis of neonatal diabetes is not different in the Japanese population as in the Caucasians so far mostly reported. The genetic basis was elucidated in 23 of these 31 patients and was mostly linked to 6q24 abnormalities and KCNJ11 mutations (18). Neuropsychological features were present in patients with KCNJ11 mutations but also four patients with unknown cause, showing us that neonatal diabetes should indeed be considered as a neuroendocrine disorder (18).
The high affinity of sulfonylurea to the pancreatic KATP channels suggested that these drugs may be used to replace insulin in these patients with KATP channels mutations. Intravenous injection of tolbutamide could stimulate insulin secretion in patients with a KCNJ11 mutation even when they did not respond to iv glucose (8). Subsequently, six patients had been successfully switched from insulin sc injections to oral sulfonylurea therapy. The dose of glibenclamide required has been up to 0.8 mg/kg/d, which is a much higher dose than that used for type 2 diabetes treatment. The delay needed to stop insulin is between 3 d and 8 wk, with a follow-up of nearly 2 yr now. These results show great heterogeneity between the patients, even with the same KCNJ11 mutations.
Our recent European collaborative study reported a total of 49 consecutive patients (from 3 months to 36 yr old) from 40 families having permanent neonatal diabetes due to heterozygous mutation of KCNJ11 (19). Of these 49 patients treated with an adequate dose of sulfonylureas (0.8 mg/kg/d equivalent glibenclamide=glipizide), 44 (90%) were able to stop insulin treatment. The median dose of glibenclamide initially required was 0.45 mg/kg/d (0.05 to 1.5 mg/kg/d). Glycemic control improved in all 38 patients tested, with a mean glycated hemoglobin level falling from 8.1% before sulfonylurea to 6.4% at 12 wk after cessation of insulin, without enhancing the frequency of hypoglycemia. Eighty percent (four of five) of the patients unable to stop insulin had neurological features in contrast with only 14% (six of 44) in the successful group. Five patients had transitory diarrhea; no other side effect was reported. Some studies have shown that closure of KATP channels by sulfonylurea therapy could induce ß-cell apoptosis in human islets and therefore precipitate the decrease in the ß-cell mass in type 2 diabetes patients, but this needs further clarification. However, in consideration of patients with KCNJ11 mutations, even if glibenclamide efficacy is transitory lasting for several years, it means years without daily sc injections and the subsequent improvement of quality of life.
Numerous questions remained to be answered including the following: what is the exact frequency of mutations in the genes encoding the potassium channel subunits, in type 2 diabetes and gestational diabetes? How can we define in fine ways the neuropsychological and neuromotor dysfunctions of the children with potassium channel mutation? What is the long-term efficacy and safety data in those children transferred from insulin injections to oral sulfonylureas? How can we go further in elucidating the etiology of other forms of neonatal diabetes that will inform on normal pancreatic development and on the basis of pathology underlying pancreatic dysfunction?
This recent advance in the field of neonatal diabetes illuminates how the molecular understanding of some monogenic form of diabetes may lead to an unexpected change of the treatment in children. This is a spectacular example by which a pharmacogenomic approach improves in a tremendous way the quality of life of our young patients. In France and some others countries, the transfer of the patients from insulin to sulfonylureas should be made within the legal rules of the country most often in the context of clinical trials, approved by the health authorities because the sulfonylureas are not licensed (even contraindicated in some countries) to be used in children. In other countries, such as the United States, an "off label use" can allow the treatment of the patients. These legal aspects of the treatment should not be underestimated due to potential deleterious side effects of the sulfonylurea. Moreover, realizing how difficult it is to take care of a child of this age with diabetes mellitus should prompt clinicians to transfer these children to specialized centers. Insulin therapy and high caloric intake are the basis of the initial treatment.
Acknowledgments
We thank our collaborators in the field of the study of neonatal diabetes: Pr. P. Czernichow, Dr. K. Busiah, Dr. I. Flechtner, Pr. J. J. Robert, and the nursing team, pediatric endocrinology, and R. Scharfmann, INSERM U845, Hôpital Necker Enfants Malades, Paris, France. We thank the following individuals for their long-standing and fruitful cooperation in the field of genetics: Dr. H. Cavé and Ms. S. Pereira (genetic biochemistry, Hôpital Robert Debré, Paris, France); Dr. M. Vaxillaire, Ms. A. Dechaume, and Pr. P. Froguel [Centre National de la Recherche Scientifique UMR 8090, Institute of Biology and Pasteur Institute (M.V., A.D., P.F.), Lille, France; Imperial College, Hammersmith Hospital (P.F.), London, United Kingdom]. We also thank all the physicians and families involved in the French Network for the Study of Neonatal Diabetes.
Footnotes
Abbreviation: NDM, Neonatal diabetes mellitus.
Received July 31, 2007.
Accepted August 1, 2007.
References
This article has been cited by other articles:
 |
|
 |
|
  R. Ganesh, R. Arvindkumar, and T. Vasanthi Infantile-Onset Diabetes Mellitus: A 1-Year Follow-up Study Clinical Pediatrics, April 1, 2009; 48(3): 271 - 274. [Abstract] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
