Treating young adults with type 2 diabetes or monogenic diabetes

Best Practice & Research Clinical Endocrinology & Metabolism 30 (2016) 455e467

Contents lists available at ScienceDirect

Best Practice & Research Clinical Endocrinology & Metabolism journal homepage: www.elsevier.com/locate/beem

10

Treating young adults with type 2 diabetes or monogenic diabetes Katharine R. Owen, BSc, MD, FRCP, Associate Professor of Diabetes * University of Oxford, UK

a r t i c l e i n f o Article history: Available online 27 May 2016 Keywords: monogenic diabetes young type 2 diabetes personalised medicine

It is increasingly recognised that diabetes in young adults has a wide differential diagnosis. There are many monogenic causes, including monogenic beta-cell dysfunction, mitochondrial diabetes and severe insulin resistance. Type 2 diabetes in the young is becoming more prevalent, particularly after adolescence. It's important to understand the clinical features and diagnostic tools available to classify the different forms of young adult diabetes. Classic type 1 diabetes is characterised by positive b-cell antibodies and absence of endogenous insulin secretion. Young type 2 diabetes is accompanied by metabolic syndrome with obesity, hypertension and dyslipidaemia. Monogenic b-cell dysfunction is characterised by non-autoimmune, C-peptide positive diabetes with a strong family history, while mitochondrial diabetes features deafness and other neurological involvement. Severe insulin resistance involves a young-onset metabolic syndrome often with a disproportionately low BMI. A suspected diagnosis of monogenic diabetes is confirmed with genetic testing, which is widely available in specialist centres across the world. Treatment of young adult diabetes is similarly diverse. Mutations in the transcription factors HNF1A and HNF4A and in the b-cell potassium ATP channel components cause diabetes which responds to low dose and high dose sulfonylurea agents, respectively, while

Abbreviations: CRP, C-reactive protein; HNF1A, hepatocyte nuclear factor 1 alpha; HNF4A, hepatocyte nuclear factor 4 alpha; HNF1B, hepatocyte nuclear factor 1 beta; GCK, glucokinase; IPSAD, International Society for Paediatric and Adolescent Diabetes; LADA, Latent Autoimmune Diabetes in Adults; MELAS, Mitochondrial Encephalopathy, Lactic Acidosis and Stroke-like episodes; MIDD, Maternally Inherited Diabetes and Deafness; MODY, maturity-onset diabetes of the young; ND, neonatal diabetes; PNDM, Permanent Neonatal Diabetes; SGLT-2, sodium glucose co-transporter-2; T1DM, type 1 diabetes mellitus; T2DM, type 2 diabetes mellitus; TNDM, Transient Neonatal Diabetes; US, United States. * Oxford Centre for Diabetes, Endocrinology and Metabolism, Churchill Hospital, Oxford OX3 7LJ, UK. Tel.: þ44 01865 857226; Fax: þ44 01865 857299. E-mail address: [email protected]. http://dx.doi.org/10.1016/j.beem.2016.05.002 1521-690X/© 2016 Elsevier Ltd. All rights reserved.

456

K.R. Owen / Best Practice & Research Clinical Endocrinology & Metabolism 30 (2016) 455e467

glucokinase mutations require no treatment. Monogenic insulin resistance and young-onset type 2 diabetes are both challenging to treat, but first line management involves insulin sensitisers and aggressive management of cardiovascular risk. Outcomes are poor in young-onset type 2 diabetes compared to both older onset type 2 and type 1 diabetes diagnosed at a similar age. The evidence base for treatments in monogenic and young-onset type 2 diabetes relies on studies of moderate quality at best and largely on extrapolation from work conducted in older type 2 diabetes subjects. Better quality, larger studies, particularly of newer agents would improve treatment prospects for young adults with diabetes. © 2016 Elsevier Ltd. All rights reserved.

Introduction Classically, diabetes presenting in the young is assumed to be Type 1 diabetes mellitus (T1DM). While T1DM still represents the majority of cases of diabetes in young children in most countries, the increased awareness of monogenic forms of diabetes, plus the increasing prevalence of obesity leading to Type 2 diabetes mellitus (T2DM) presenting at a younger age, mean that those working with young adults should be familiar with the differential diagnosis and management of all forms of diabetes. This article reviews the range of presentation of different types of diabetes in young adults and discusses evidence-based recommendations for management strategies. Background: epidemiology and clinical features Monogenic diabetes Monogenic diabetes comprises maturity-onset diabetes of the young (MODY), neonatal diabetes (ND), mitochondrial diabetes and the inherited causes of severe insulin resistance. MODY, ND and mitochondrial mutations predominantly lead to b-cell defects, whilst monogenic severe insulin resistance causes young onset metabolic syndrome and abnormalities of body fat distribution (often without obesity). Over recent years there has been a huge expansion in the number of genes known to cause diabetes, particularly those linked to ND [1], however most are rare. Good prevalence data is not available, but monogenic diabetes probably represents 3e5% of diabetes in those diagnosed <45 years [2,3], has a minimum population prevalence of around 1 in 10,000 [4e6], and arises in the first 6 months of life in approximately 1 in every 100,000 births [1]. MODY is characterised by autosomal dominant inheritance, age of onset in the 2nde4th decade, absence of b-cell autoimmunity, lack of metabolic syndrome features, and sustained insulin secretion (C-peptide positivity) [7]. The commonest types of MODY are due to mutations in hepatic nuclear factor 1 alpha (HNF1A), glucokinase (GCK), hepatic nuclear factor 4 alpha (HNF4A) and hepatic nuclear factor 1 beta (HNF1B), representing 52%, 32%, 10% and 6%, respectively of known MODY cases in the UK [5].

Glucokinase MODY This is the commonest form of monogenic diabetes in children. Glucokinase, known as the pancreatic b-cell glucose sensor, regulates the initiation of glucose-stimulated insulin secretion. Heterozygous inactivating GCK mutations cause a lifelong, non-progressive, fasting hyperglycaemia (fasting glucose 5.5e8.0 mmol/l, HbA1c 40e60 mmol/mol) [8]. Unlike other forms of hyperglycaemia, insulin secretion remains intact in GCK-MODY and regulated, although the ambient blood glucose is shifted 2e3 mmol/l higher. This results in low post-prandial glucose excursions compared to other

K.R. Owen / Best Practice & Research Clinical Endocrinology & Metabolism 30 (2016) 455e467

457

forms of diabetes. A large observational study has demonstrated that GCK-MODY patients do not develop microvascular complications, nor appear to have a higher risk of macrovascular complications than age-matched non-diabetic individuals [9]. Thus intensive follow-up is not recommended other than an annual HbA1c. In pregnancy, insulin secretion in the foetus, and thus birth weight depends on whether the fetus is affected or not, so foetal growth should be monitored. Diagnosis of GCK-MODY is often an incidental finding when glucose is measured for another reason. It is a common cause of asymptomatic fasting hyperglycaemia in the 1ste2nd decade of life [10], but in older adults represents a much lower proportion of cases [11]. HNF1A-MODY This is the commonest type of MODY, caused by heterozygous mutations in the transcription factor HNF1A. Affected individuals are normoglycaemic in childhood, but develop progressive b-cell dysfunction that results in diabetes in the 2nde4th decade of life. Similar to T1DM and T2DM, microvascular and macrovascular complications are common when glycaemic targets are not met; therefore regular HbA1c monitoring is required. Extra pancreatic features of HNF1A-MODY include a low renal threshold for glucose [12] and low levels of C-reactive protein (CRP) [3]. CRP has been suggested as a screening biomarker for HNF1A-MODY. HNF4A-MODY Mutations in HNF4A have a similar clinical presentation to HNF1A-MODY. One difference is that (for reasons not well understood) HNF4A mutations cause hyperinsulinaemia in utero and neonatal life, leading to macrosomia and neonatal hypoglycaemia [13]. This is generally transient, progressing to normoglycaemia in childhood and then onset of diabetes in adolescent or young adult life. The renal glucose threshold and CRP levels are normal. HNF1B MODY Mutations in HNF1B cause the “Renal Cysts and Diabetes Syndrome”, which is a syndrome of developmental abnormalities in the pancreas, kidneys and genito-urinary system. Renal abnormalities may be detected before birth on foetal anomaly scans and are frequently the first presentation. Nondiabetic renal failure can occur. Pancreatic atrophy leads to both diabetes and pancreatic exocrine insufficiency in adult life [14]. Rare types of MODY Mutations in KCNJ11, ABCC8 and INS cause up to 1% of MODY cases, but are far better known for their role in neonatal diabetes (see below). Other genes that have been implicated in MODY families include IPF1, NEUROD1, BLK, KLF11 and CEL1. However, they have been reported in only a small number of cases and the genetic evidence is frequently not compelling. In some strong MODY pedigrees, no causative gene has been identified, despite the use recently of next generation sequencing approaches in these families. Mitochondrial diabetes Mutations in the mitochondrial genome causes diabetes with a classic syndrome of Maternally Inherited Diabetes and Deafness (MIDD). Closer assessment often reveals that these individuals have multisystem involvement including other neurological symptoms, myopathy, and cardiomyopathy [15]. Clinical features are commonly similar to MODY, although diabetes onset may be later and is often preceded by deafness. Specialist monitoring for the manifestations of mitochondrial disease is advised. Genetic counselling should be offered to female carriers as their children invariably inherit the mutation with unpredictable penetrance and clinical consequences.

458

K.R. Owen / Best Practice & Research Clinical Endocrinology & Metabolism 30 (2016) 455e467

The commonest mutation is the m.3243A > G point substitution. Testing for this mutation is widely available, while sequencing of the whole mitochondrial genome is offered in more specialist centres. Neonatal diabetes mellitus (ND) Diabetes presenting in the first 6 months of life is unlikely to be T1DM, based on negative pancreatic antibodies and presence of protective HLA variants [16]. Molecular investigation reveals a monogenic aetiology in a high proportion of cases [1]. Two distinct subgroups exist: Transient Neonatal Diabetes (TNDM), which usually remits by 12 weeks but may relapse after some years, and Permanent Neonatal Diabetes (PNDM). The commonest type of PNDM (40% of cases) is caused by activating mutations in the KCNJ11 and ABCC8 genes encoding subunits of the pancreatic b-cell KATP channel, making the channel unresponsive to ATP. This results in markedly decreased or absent insulin secretion [17]. Insulin gene (INS) mutations are also a common cause of PNDM (15% of cases) [18]. Homozygous or compound heterozygous inactivating mutations of GCK are a very rare cause of PNDM [19]. In TNDM, the commonest underlying mechanism is a methylation defect at chromosome 6q24 (ZAC and HYMAI genes) [20]. Mutations in KCNJ11, ABCC8 and INS can also present as TNDM [21]. A large number of other genes are involved in syndromes which include neonatal diabetes [1]. Inherited severe insulin resistance Monogenic severe insulin resistance can be divided broadly into defects of adipose tissue (the lipodystrophies) and disorders of insulin signalling [22]. Like MODY, the diagnosis is often missed and the true prevalence is unknown. The lipodystrophies Lipodystrophies are characterised by lack of adipose tissue which can be partial or generalised, with genetic or acquired aetiology [23]. There is limited capacity to store fat, and so triglycerides accumulate in other organs (liver, pancreas, pericardium). Affected individuals present with metabolic syndrome (dyslipidaemia, hypertension, dysglycaemia, fatty liver, hyperandrogenism in women) accompanied by abnormal fat distribution and acanthosis nigricans. Young-onset metabolic syndrome (not always with diabetes), which seems out of proportion to the degree of obesity should trigger suspicion of this diagnosis. Partial lipodystrophy due LMNA or PPARG mutations is most commonly encountered leading to subcutaneous fat loss, characteristically in the limb and gluteal depots starting in the 2nde3rd decade. Generalised lipodystrophy presents with absent adipose tissue from birth. It is very rare and caused by mutations in AGPAT2, BSCL2, CAV1 and PTRF. Disorders of insulin signalling Inherited disorders of insulin signalling are mostly accounted for by insulin receptor (INSR) mutations, with a few rarer causes [22]. Individuals present with features of insulin resistance without lipodystrophy, or, more rarely, with severe childhood forms (Donohue or RabsoneMendenhall Syndromes). The pathognomonic feature of INSR mutations is a normal or high adiponectin level whereas in obesity and other forms of insulin resistance adiponectin levels are usually low. Lipid profiles are also usually normal. Young-onset type 2 diabetes Over the past 2 decades, T2DM in children and adolescents has gone from being a peculiarity to commonplace, although prevalence in different countries varies widely. Evidence for the increasing rates of T2DM in the young was initially noted from paediatric epidemiological studies in the USA. The SEARCH study reported that the prevalence of T2DM in youths aged 10e19 years in US was 0.46/1000 in 2009, an increase of 35% over the prevalence of 0.34/1000 in 2001

K.R. Owen / Best Practice & Research Clinical Endocrinology & Metabolism 30 (2016) 455e467

459

[24], suggesting that there were at least 20,000 cases of type 2 diabetes in this age group in the US. The prevalence was particularly high in youths from ethnic minority backgrounds. T2DM in childhood and adolescence is much less common in Europe than in the US, although there is evidence that cases have increased over the last 15 years. In the UK, surveys and latterly national audit data have shown a rise in reported cases of T2DM from just 25 children in 2000 (minimum prevalence 0.21/100,000, higher in South Asian children) [25] to 328 cases in 2009 (3/100,000) and 450 in 2012 [26], equating to 1.1% of boys and 3.6% of girls with diabetes diagnosed under 19 years. The relative proportion was higher in children with diabetes from ethnic minority backgrounds (2.1% of Asian children, 8.7% of Black children). Most cases arise in adolescents over the age of 15. In contrast, a recent survey of Danish children with diabetes showed only 7 cases of T2DM on the Danish national register [27]. T2DM in young adults is considerably more common, with UK data suggesting that T2DM now accounts for 25% of the diabetes in the 20e29 year age group [28], although only 0.5% of all cases of T2DM. Therefore, in the UK at least, there is a large jump in the numbers affected with T2DM during the transition from childhood to young adulthood. Published data from population registries support these observations [29,30]. In all reports, the increase in young-onset T2DM disproportionately affects those from minority ethnic backgrounds. Young-onset T2DM is associated with poor outcomes Young-onset T2DM is characterised by the presence of obesity and other adverse cardiovascular risk factors [31]. In patients with T2DM diagnosed before 35 years of age who were attending a specialist clinic [32], 63% had suboptimal glycaemic control (HbA1c > 7% [53 mmol/mol]) and 65% had at least 3 cardiovascular risk factors. This typical phenotype of obesity, poor glycaemic control and other features of the metabolic syndrome has also been noted in several other UK cohorts of young-onset T2DM [9,33,34] as well as in the large East Asian JADE cohort [35]. A high risk of diabetes-related complications was noted in a observational study of patients with young-onset T2DM (diagnosed <45 years) after approximately 10 years disease duration: 36% patients had clinically-significant microvascular disease, i.e. more than background retinopathy or persistent microalbuminuria/proteinuria, and 30% had clinically-significant macrovascular disease [9]. Data also suggest a worse overall prognosis of young-onset T2DM compared with those diagnosed at an older age. In US patients with early-onset (diagnosed aged 18e45 years) versus usual-onset (diagnosed aged > 45 years) newly-diagnosed type 2 diabetes [36], those with early-onset T2DM had double the overall hazard of developing any macrovascular complication compared with lateronset T2DM over a 4-year follow-up. The rate of development of microvascular complications was similar between the two groups in this study. Studies in UK and Chinese cohorts have reported similar excess risks of macrovascular complications presenting in middle age, in part determined by prolonged disease duration [34,37]. Studies have also shown that the prognosis of young-onset T2DM is worse compared with individuals with T1DM diagnosed at a similar age. Examination of an Australian hospital diabetes clinical database (with >20 years of follow-up data) [38], revealed that patients with young-onset T2DM had higher BMIs and other adverse cardiovascular risk factors such as high triglycerides, low HDLcholesterol, hypertension, and higher prescription of statins and anti-hypertensive medications. Despite shorter duration of diabetes and comparable levels of glycaemia, mortality rates were higher in young-onset T2DM and there was a marked excess of macrovascular disease, including ischaemic heart disease and stroke, in young-onset T2DM compared with T1DM. Similar observations were made in American and Chinese cohort studies [39,40]. These studies suggest the increased cardiovascular risk is driven by features of metabolic syndrome, rather than by dysglycaemia alone. Diagnostic differentiation of young adult diabetes Both monogenic diabetes and T2DM are differentiated from T1DM by the absence of b-cell antibodies, preservation of some insulin secretion (C-peptide positive) and an HLA profile which is low risk or protective for T1DM. In practice the diagnosis is frequently not straightforward, as patients do not

460

K.R. Owen / Best Practice & Research Clinical Endocrinology & Metabolism 30 (2016) 455e467

always fit into defined categories. 10e20% of T1DM are negative for commonly tested antibodies, a small proportion maintain some long-term insulin secretion and HLA typing is not commonly used in clinical practice. Clinical features are often similar between T1DM and MODY, while young T2DM tend to have a prominent phenotype of obesity and insulin resistance. However mixed phenotypes are common with T1DM and MODY who are obese, and young T2DM presenting with ketosis or with an antibody positive Latent Autoimmune Diabetes in Adults (LADA) type picture. As more than one form of diabetes is seen not infrequently within a family [41], genetic testing should be performed to confirm a suspected diagnosis of monogenic diabetes before treatment decisions are made. Molecular genetic diagnosis is available in most Western countries via direct sequencing, or, increasingly next-generation sequencing panels [42]. The implications of molecular diagnosis are important not only for the proband, by allowing personalised management, but also for their relatives by prompting cascade screening and definitive diagnoses. It is clear that for young adults affected with diabetes, defining the aetiology determines the treatment (Table 1) and this is one of the best examples of stratified medicine in chronic disease.

Treatment of monogenic diabetes HNF1A- and HNF4A-MODY One of the most important features of HNF1A-MODY (and HNF4A-MODY) is the sensitivity observed to sulfonylurea drugs. This was initially noted anecdotally [43], and then confirmed in a small but elegant randomised controlled trial [44]. Gliclazide and metformin were compared in a crossover study of HNF1A-MODY and BMI-matched type 2 diabetes; the HNF1A-MODY patients showed a fourfold greater drop in fasting plasma glucose on gliclazide compared with those who had T2DM, while the response to metformin was similar in both groups. The authors also showed that the effects are mediated via increased insulin secretion (rather than differences in metabolism of sulfonylurea drugs), probably by bypassing the ATP-generating defects in the b-cell caused by HNF1A mutations, although the mechanism is not completely understood. Another study showed that the response was similar with nateglinide, a non-sulfonylurea prandial insulin secretagogue [45]. Thus a low dose of a sulfonylurea represents the first-line treatment in HNF1A- and HNF4A-MODY, with a suggested starting dose in treatment-naïve patients, or those with diabetes for <10 years, of 20e40 mg gliclazide or equivalent. In patients already commenced on other treatments, including insulin, a sulfonylurea can be substituted safely with as good or better control achieved [33,46,47]. In our experience (also reported by Bacon et al. [46]), a long-term HbA1c of ~ 40 mmol/mol (around 6%) can be achieved without significant hypoglycaemia in patients with HNF1A-or HNF4A-MODY. Another randomised controlled trial examined head-to-head treatment with glimepiride versus liraglutide in HNF1A-MODY [48]. This showed superior glucose lowering in the group treated with glimepiride, although there was an increased prevalence of mild hypoglycaemia. Given the discrepancy in cost between the two classes of drugs, there is insufficient evidence to recommend the use of GLP-1 receptor agonists as first line therapy in HNF1A-MODY. Secondary sulfonylurea failure usually develops over time as b-cell dysfunction progresses. There is little evidence to recommend a specific second line treatment after maximising sulfonylurea therapy in HNF1A- and HNF4A-MODY, but metformin and DPP-4 inhibitors are commonly used, followed by

Table 1 Treatment according to aetiology in young adult diabetes. Type of diabetes

First line drug treatment

Type 1 Type 2 HNF1A/HNF4A mutation Glucokinase mutation Kþ channel mutation Severe insulin resistance Lipodystrophy

Insulin Metformin Low dose sulfonylurea (40 mg gliclazide/day equivalent) None indicated High dose sulfonylurea (0.5 mg/kg glibenclamide/day equivalent) Insulin sensitiser (metformin or pioglitazone) Metreleptin (in generalised LD or where leptin is very low)

K.R. Owen / Best Practice & Research Clinical Endocrinology & Metabolism 30 (2016) 455e467

461

addition of a basal insulin. As HNF1A regulates SGLT2 expression in the kidney, and a low renal threshold is already observed, it is likely that the sodium glucose co-transporter-2 (SGLT-2) inhibitors would not have great efficacy in HNF1A-MODY. Treatment of GCK-MODY No change in HbA1c was seen in a small observational study of 16 GCK-MODY subjects after pharmacological therapy was discontinued following genetic diagnosis [49], suggesting that treatment with oral hypoglycaemic agents or insulin does not affect diabetes control (probably because insulin secretion remains regulated by glucose leading to a small post-prandial glucose increment). This, along with the lack of vascular complications seen in GCK-MODY as discussed above [9], have resulted in the recommendation that glucose-lowering therapies can be stopped in most individuals with GCK-MODY (although insulin is often used in pregnancy). Annual monitoring of HbA1c is recommended to identify any individuals who develop worsening of hyperglycaemia because of concomitant obesity/insulin resistance. In this situation a rising HbA1c should be managed with metformin, and then as per T2DM recommendations. Secondary care follow-up of patients is probably not warranted, so long as clear information is given to both the patient and their primary care provider. Other types of MODY Patients with HNF1B mutations are not sensitive to sulfonylureas, and while they can be managed on oral agents initially, generally progress to insulin fairly rapidly [50]. Metformin may be contraindicated in the presence of renal impairment. There is little evidence for any particular treatment in the rarer forms of MODY, although for the small proportion of those with potassium channel mutations (in KCNJ11 and ABCC8) who present after the neonatal period, response to sulfonylureas has been observed [51]. Treatment of mitochondrial diabetes The evidence base for any particular treatment is again sparse. Whilst oral agents can be initially successful, as the diabetes is caused by b-cell dysfunction, progression to insulin treatment usually occurs relatively quickly. There is a theoretical argument for avoiding metformin in mitochondrial diabetes because of a potential risk of lactic acidosis. A slightly raised lactate is sometimes seen in mitochondrial disease, and lactic acidosis is a feature of the rare mitochondrial syndrome of MELAS (Mitochondrial Encephalopathy, Lactic Acidosis and Stroke-like episodes). Neonatal diabetes Molecular diagnosis of neonatal diabetes is important because those shown to have mutations in the KATP channel components can be treated with sulfonylureas. ABCC8 encodes the sulfonylurea receptor, and binding of a sulfonylurea restores the activity of the channel and thus insulin secretion very successfully, even many years after diagnosis. This results in a lower HbA1c in most cases, and less glucose variability than with insulin therapy [52]. Unlike HNF1A-MODY and HNF4A-MODY, high doses of oral sulfonylureas, e.g. 0.5 mg/kg (range 0.13e1.2 mg/kg) equivalent of glibenclamide, are required. The developmental delay observed in some individuals with KATP channel mutations may also be improved by sulfonylureas [53]. Monogenic insulin resistance Treatment of severe insulin resistance can be challenging. Logically, insulin sensitisers (metformin and thiazolidinediones) should be the first line oral treatment, but high doses of insulin are often required. The U500 insulin formulation (5 times as concentrated as usual insulin preparations) can be useful in reducing the volume of formulation required. New more concentrated basal insulins such as Degludec (U200) and the U300 glargine (Toujeo) aimed at people with T2DM on high insulin doses may

462

K.R. Owen / Best Practice & Research Clinical Endocrinology & Metabolism 30 (2016) 455e467

also be useful. In generalised lipodystrophy, recombinant leptin (Metreleptin) treatment can be effective in reversing the metabolic abnormalities seen [54], but is not widely available. Hypertriglyceridaemia can be difficult to treat and aggressive cardiovascular risk management is recommended [55]. Treatment of young-onset T2DM It is accepted that the evidence-base to guide optimal management of patients with young-onset T2DM is limited. Guidelines published by the International Society for Paediatric and Adolescent Diabetes (IPSAD) on the management of type 2 diabetes in children and adolescents are largely based on consensus expert opinion in the absence of high-quality comparative-effectiveness research studies [56]. The guidelines emphasise management strategies such as weight loss, achieving glycaemic targets and treatment of other metabolic syndrome features such as hypertension and dyslipidaemia. A joint EASD/ADA position statement has emphasised the need for a patient-centred approach in the management of young-onset T2DM and suggested that HbA1c targets should take into account the poor outcomes seen in younger patients, as outlined earlier in this article [57]. However, in neither of these documents is there specific guidance as to how these goals should be achieved and management decisions are therefore largely extrapolated from the evidenced-based protocols that guide current treatment of T2DM presenting in older adults. These may not always be suitable or applicable to a younger age group. As only insulin and metformin are licensed for use in paediatric populations, there is likely to be a variation in treatment offered depending on whether care is provided by paediatricians, adult physicians or general practitioners. Metformin was studied as a monotherapy versus placebo in a small randomised controlled trial of treatment-naïve patients with young-onset T2DM aged 10e16 [58]. This showed that metformin was effective in lowering blood glucose in this age group, with similar adverse effects to those seen in adults. During the 16-week study, 65% of those randomised to placebo required additional medication before study end because of inadequate glycaemic control, versus only 10% in the metformin treated group. Metformin and glimepiride were compared in a randomised trial of paediatric subjects age 8e17 years who had young-onset T2DM inadequately controlled with lifestyle or oral monotherapy [59]. The two agents showed similar efficacy in HbA1c reduction over 24 weeks, with no difference in hypoglycaemia or other adverse effects. The most important to date in young-onset T2DM is the Treatment Option for Type 2 diabetes in Adolescents and Youth (TODAY) study. This multi-ethnic, multi-centre, randomised controlled trial compared metformin monotherapy, metformin plus rosiglitazone, and metformin plus intensive lifestyle intervention in 699 obese youths aged 10e17 years with newly-diagnosed T2DM [60]. Over a mean ~4-year follow-up, there was no difference between the responses seen in the metformin alone and metformin plus lifestyle intervention groups, with around 50% of patients failing to achieve a durable response (defined as HbA1c < 8% (64 mmol/mol) for at least 6 months). The failure rate in the metformin plus rosiglitazone group was lower at 40%. In this study, most patients required escalation to multiple oral hypoglycaemic agents and/or insulin to achieve good glycaemic control, suggesting that response to oral agents in young-onset T2DM is less long-lived than that reported in older-onset T2DM patients [61]. Furthermore, there was a high incidence of microalbuminuria, hypertension, and dyslipidaemia over the course of the TODAY study, consistent with other data suggesting that the development of serious diabetes complications occurs prematurely in young-onset T2DM [62]. A recent study compared liraglutide in comparison with placebo in 10e17 year old T2DM participants with HbA1c above target on lifestyle or metformin monotherapy [63]. This study showed a superior HbA1c drop in the group randomised to liraglutide, although only 21 patients were recruited (19 completed the trial). GI side effects were most commonly seen during dose escalation, with no serious adverse effects were reported. A larger multicentre liraglutide study is currently recruiting. GLP-1 receptor agonists could be promising treatments for young-onset T2DM, given their positive effects on weight loss and liver steatosis as well as glycaemia. Current US and European regulatory requirements for development of new T2DM therapies now stipulate that they should be evaluated in children over 10 years old. These studies are challenging to perform due to low prevalence of disease in any given centre and the difficulty of recruiting children.

K.R. Owen / Best Practice & Research Clinical Endocrinology & Metabolism 30 (2016) 455e467

463

The European Medicines Agency advises on applications for paediatric investigation plans for drug trials in children, which should lead to more studies in this age range. However most studies currently in progress are to define safety and pharmacokinetic parameters rather than to provide definitive information on treatment response. Table 2 lists the 4 published randomised trials discussed in the text above. Although we assume that in most cases the response to hypoglycaemic agents will mirror that seen in older onset type 2 diabetes, age-related differences in underlying physiology could affect this. For example, renal glucose threshold is lower in younger individuals and a meta-analysis of the response to SGLT-2 inhibitors suggested that the HbA1c reduction was greater in younger patients [64]. To investigate this observation, the effect of a single dose of dapagliflozin was compared in younger (40 years) and older adults (>40 years) with T2DM [65]. This confirmed a greater increase in urinary glucose and a

Table 2 Summary of published randomised controlled trials of treatment in children with T2DM. Drug studied

Inclusion criteria (those with positive b-cell antibodies and recent DKA were excluded)

Outcome

Metformin (MF) vs placebo [58]

Treatment naïve or no recent treatment Age 8e16 years FPG 7e13.3 mmol/l, HbA1c  7.0%, Stimulated Cpeptide  0.5 nmol/l

Metformin (MF) vs glimepiride (GL) [59]

Treatment failure on lifestyle or monotherapy No recent prolonged insulin therapy Age 8e17 years, HbA1c 7.1 e12.0% Stimulated Cpeptide  0.5 nmol/l

TODAY study: Metformin (MF) vs Metformin þ Rosiglitazone (MF þ R) vs Metformin þ lifestyle (MF þ Life) [60]

T2DM < 2 years, BMI > 85th percentile for age Age 10e17 years HbA1c < 8% on Metformin alone Fasting C-peptide > 0.2 nmol/l

Liraglutide (LIR) vs placebo [63]

Age 10e17, BMI > 85th percentile for age Treatment with lifestyle or metformin HbA1c 6.5e11% or FPG 6.1

82 participants (17% of those screened), 16 week study Mean FPG decreased by 2.4 mmol/l in MF vs 1.2 mmol/l rise in placebo Mean HbA1c MF 7.5% vs placebo 8.6% 10% of MF and 65% of placebo needed additional treatment before the end of 16 weeks 285 participants, 24 weeks of treatment At 24 weeks: HbA1c decrease equivalent between groups (0.7% GL vs 0.85% MF). % of patients reaching target < 7% e equivalent between groups BMI change: þ0.26 kg/m2 GL vs 0.33 kg/m2 MF, p ¼ 0.003 Hypoglycaemia rates equivalent between groups 699 participants, FU minimum 2 years, mean 3.8 years, max 6.5 years Primary outcome: time to treatment failure of HbA1c  8% over a period of 6 months or metabolic decompensation requiring insulin treatment Mean time (months) to treatment failure did not differ between groups: MF 10.3 vs MF þ R 12.0 vs MF þ Life 11.8 months, p ¼ 0.63 Overall failure rates: MF 51.7%, MF þ R 38.6%, MF þ Life 46.6%. MF vs MF þ R, p ¼ 0.006 BMI not a determinant of treatment failure Female patients had better response to Rosiglitazone. None-white ethnic groups had higher failure rate 21 participants (19 completed trial): 14 LIR, 7 placebo. Study duration 6 weeks Primary endpoint was safety and tolerability Secondary endpoints were (continued on next page)

464

K.R. Owen / Best Practice & Research Clinical Endocrinology & Metabolism 30 (2016) 455e467

Table 2 (continued ) Drug studied

Inclusion criteria (those with positive b-cell antibodies and recent DKA were excluded)

Outcome

e13.3 mmol/l BMI > 85th centile for age/ gender

pharmacokinetic profile plus changes in HbA1c, FPG and body weight over 5 weeks of treatment. HbA1c change: 0.86% LIR vs 0.04% placebo, p ¼ 0.0007 No differences in body weight observed GI side effects observed in LIR group during titration e similar to adults.

larger decrease in plasma glucose excursions on continuous glucose monitoring after exposure to dapagliflozin in the younger adults. There is a clear need for high-quality research studies to determine the optimal therapeutic strategy in patients with young-onset T2DM. We would expect that early glycaemic optimisation would have the same legacy effect seen in other trials of intensive treatment [66], which is essential to improve outcomes in these high risk patients. Other outstanding areas of uncertainty include the role of bariatric surgery, at what age lipid lowering treatment should be commenced, whether these patients should be managed in primary or secondary care and the use of teratogenic medications in women of child-bearing age.

Summary There is increasing recognition that diabetes seen in childhood and young adult life has a wide range of aetiologies, including the many monogenic forms of diabetes and increasing numbers of young-onset T2DM as well as classical and more slowly progressive T1DM, and LADA. Differentiating these different subtypes from each other and from T1DM can be challenging, but is important for correct management and for stratifying risk of future outcomes for both patients and their families. A common feature of the monogenic forms of diabetes and young-onset T2DM is that the evidence base for optimum management is based largely on anecdotal reports, small randomised controlled trials and extrapolation from studies in classic older-onset T2DM. Therefore there is a need for better, larger studies, particularly in young-onset T2DM, and for increased collaboration between centres to collate data on responses to treatment and outcomes in the rarer forms of diabetes.

Practice points  Clinicians should be aware of the clinical features of the different forms of young adult diabetes so that the correct aetiology can be assigned  Aetiology influences management in young adult diabetes  3e5% of those diagnosed before age 45 have monogenic diabetes for whom molecular genetic testing can confirm the diagnosis  Increasing numbers of children and young adults have T2DM, although prevalence varies widely between different countries  Young-onset T2DM is associated with suboptimal management of glycaemia and cardiovascular risk factors and with poor outcomes, compared with both older-onset T2DM and T1DM  High quality evidence for treatment responses and outcomes is lacking in both monogenic and young-onset T2DM

K.R. Owen / Best Practice & Research Clinical Endocrinology & Metabolism 30 (2016) 455e467

465

Research agenda  Understanding treatment responses to current and new treatments in monogenic forms of diabetes  Targeting treatment for monogenic forms of diabetes based on an understanding of the pathophysiology  Gene discovery in monogenic diabetes using next-generation sequencing techniques  Using a genetic diagnosis to predict future prognosis, e.g. glucokinase mutation, syndromic forms of neonatal diabetes where the genetic aetiology is known before all features of the syndrome present  Larger studies of new and existing treatments in young-onset T2DM

References *[1] De Franco E, Flanagan SE, Houghton JA, et al. The effect of early, comprehensive genomic testing on clinical care in neonatal diabetes: an international cohort study. Lancet 2015;386(9997):957e63. [2] Pihoker C, Gilliam LK, Ellard S, et al., S.f.D.i.Y.S. Group. Prevalence, characteristics and clinical diagnosis of maturity onset diabetes of the young due to mutations in HNF1A, HNF4A, and glucokinase: results from the SEARCH for diabetes in youth. J Clin Endocrinol Metab 2013;98(10):4055e62. [3] Thanabalasingham G, Shah N, Vaxillaire M, et al. A large multi-centre European study validates high-sensitivity C-reactive protein (hsCRP) as a clinical biomarker for the diagnosis of diabetes subtypes. Diabetologia 2011;54(11):2801e10. [4] Kropff J, Selwood MP, McCarthy MI, et al. Prevalence of monogenic diabetes in young adults: a community-based, crosssectional study in Oxfordshire, UK. Diabetologia 2011;54(5):1261e3. *[5] Shields BM, Hicks S, Shepherd MH, et al. Maturity-onset diabetes of the young (MODY): how many cases are we missing? Diabetologia 2010;53(12):2504e8. [6] Eide SA, Raeder H, Johansson S, et al. Prevalence of HNF1A (MODY3) mutations in a Norwegian population (the HUNT2 study). Diabet Med 2008;25(7):775e81. *[7] Thanabalasingham G, Owen KR. Diagnosis and management of maturity onset diabetes of the young (MODY). BMJ 2011; 343:d6044. [8] Steele AM, Wensley KJ, Ellard S, et al. Use of HbA1c in the identification of patients with hyperglycaemia caused by a glucokinase mutation: observational case control studies. PLoS One 2013;8(6):e65326. *[9] Steele AM, Shields BM, Wensley KJ, et al. Prevalence of vascular complications among patients with glucokinase mutations and prolonged, mild hyperglycemia. JAMA 2014;311(3):279e86. *[10] Feigerlova E, Pruhova S, Dittertova L, et al. Aetiological heterogeneity of asymptomatic hyperglycaemia in children and adolescents. Eur J Pediatr 2006;165(7):446e52. [11] Gloyn AL, van de Bunt M, Stratton IM, et al. Prevalence of GCK mutations in individuals screened for fasting hyperglycaemia. Diabetologia 2009;52(1):172e4. [12] Stride A, Ellard S, Clark P, et al. Beta-cell dysfunction, insulin sensitivity, and glycosuria precede diabetes in hepatocyte nuclear factor-1alpha mutation carriers. Diabetes Care 2005;28(7):1751e6. [13] Pearson ER, Boj SF, Steele AM, et al. Macrosomia and hyperinsulinaemic hypoglycaemia in patients with heterozygous mutations in the HNF4A gene. PLoS Med 2007;4(4):e118. [14] Bellanne-Chantelot C, Chauveau D, Gautier JF, et al. Clinical spectrum associated with hepatocyte nuclear factor-1beta mutations. Ann Intern Med 2004;140(7):510e7. [15] Murphy R, Turnbull DM, Walker M, et al. Clinical features, diagnosis and management of maternally inherited diabetes and deafness (MIDD) associated with the 3243A > G mitochondrial point mutation. Diabet Med 2008;25(4):383e99. [16] Iafusco D, Stazi MA, Cotichini R, et al. Early Onset Diabetes Study Group of the Italian Society of Paediatric, and Diabetology, permanent diabetes mellitus in the first year of life. Diabetologia 2002;45(6):798e804. [17] Gloyn AL, Pearson ER, Antcliff JF, et al. Activating mutations in the gene encoding the ATP-sensitive potassium-channel subunit Kir6.2 and permanent neonatal diabetes. N Engl J Med 2004;350(18):1838e49. [18] Stoy J, Edghill EL, Flanagan SE, et al. Insulin gene mutations as a cause of permanent neonatal diabetes. Proc Natl Acad Sci U S A 2007;104(38):15040e4. [19] Njolstad PR, Sovik O, Cuesta-Munoz A, et al. Neonatal diabetes mellitus due to complete glucokinase deficiency. N Engl J Med 2001;344(21):1588e92. [20] Temple IK, Shield JP. 6q24 transient neonatal diabetes. Rev Endocr Metab Disord 2010;11(3):199e204. [21] Flanagan SE, Patch AM, Mackay DJ, et al. Mutations in ATP-sensitive Kþ channel genes cause transient neonatal diabetes and permanent diabetes in childhood or adulthood. Diabetes 2007;56(7):1930e7. *[22] Parker VE, Savage DB, O'Rahilly S, et al. Mechanistic insights into insulin resistance in the genetic era. Diabet Med 2011; 28(12):1476e86. [23] Agarwal AK, Garg A. Genetic disorders of adipose tissue development, differentiation, and death. Annu Rev Genomics Hum Genet 2006;7:175e99. *[24] Dabelea D, Mayer-Davis EJ, Saydah S, et al. Prevalence of type 1 and type 2 diabetes among children and adolescents from 2001 to 2009. JAMA 2014;311(17):1778e86.

466

K.R. Owen / Best Practice & Research Clinical Endocrinology & Metabolism 30 (2016) 455e467

[25] Ehtisham S, Hattersley AT, Dunger DB, et al. First UK survey of paediatric type 2 diabetes and MODY. Arch Dis Child 2004; 89(6):526e9. [26] RCPCH. National paediatric diabetes audit report 2011e12. 2013. Available from: http://www.rcpch.ac.uk/system/files/ protected/page/NPDA%20report%202011-12%20high%20res%20FINAL.pdf. [27] Oester IM, Kloppenborg JT, Olsen BS, et al. Type 2 diabetes mellitus in Danish children and adolescents in 2014. Pediatr Diabetes 2015. http://dx.doi.org/10.1111/pedi.12291. [28] National Diabetes Audit. Available from: http://www.hscic.gov.uk/nda; 2012. [29] Harron KL, Feltbower RG, McKinney PA, et al. Rising rates of all types of diabetes in south Asian and non-south Asian children and young people aged 0e29 years in West Yorkshire, U.K., 1991e2006. Diabetes Care 2011;34(3):652e4. [30] Holden SH, Barnett AH, Peters JR, et al. The incidence of type 2 diabetes in the United Kingdom from 1991 to 2010. Diabetes Obes Metab 2013;15(9):844e52. [31] Wilmot EG, Leggate M, Khan JN, et al. Type 2 diabetes mellitus and obesity in young adults: the extreme phenotype with early cardiovascular dysfunction. Diabet Med 2014;31(7):794e8. [32] Benhalima K, Song SH, Wilmot EG, et al. Characteristics, complications and management of a large multiethnic cohort of younger adults with type 2 diabetes. Prim Care Diabetes 2011;5(4):245e50. [33] Thanabalasingham G, Pal A, Selwood MP, et al. Systematic assessment of etiology in adults with a clinical diagnosis of young-onset type 2 diabetes is a successful strategy for identifying maturity-onset diabetes of the young. Diabetes Care 2012;35(6):1206e12. [34] Song SH, Hardisty CA. Early onset type 2 diabetes mellitus: a harbinger for complications in later years e clinical observation from a secondary care cohort. QJM 2009;102(11):799e806. *[35] Yeung RO, Zhang Y, Luk A, et al. Metabolic profiles and treatment gaps in young-onset type 2 diabetes in Asia (the JADE programme): a cross-sectional study of a prospective cohort. Lancet Diabetes Endocrinol 2014;2(12):935e43. [36] Hillier TA, Pedula KL. Complications in young adults with early-onset type 2 diabetes: losing the relative protection of youth. Diabetes Care 2003;26(11):2999e3005. [37] Chan JC, Lau ES, Luk AO, et al. Premature mortality and co-morbidities in young-onset diabetes e a 7 year prospective analysis. Am J Med 2014;127(7):616e24. *[38] Constantino MI, Molyneaux L, Limacher-Gisler F, et al. Long-term complications and mortality in young-onset diabetes: type 2 diabetes is more hazardous and lethal than type 1 diabetes. Diabetes Care 2013;36(12):3863e9. [39] Dart AB, Martens PJ, Rigatto C, et al. Earlier onset of complications in youth with type 2 diabetes. Diabetes Care 2014; 37(2):436e43. [40] Luk AO, Lau ES, So WY, et al. Prospective study on the incidences of cardiovascular-renal complications in chinese patients with young-onset type 1 and type 2 diabetes. Diabetes Care 2014;37(1):149e57. [41] Kavvoura F, Raimondo A, Thanabalasingham G, et al. Reclassification of diabetes etiology in a family with multiple diabetes phenotypes. J Clin Endocrinol Metab 2014:jc20133641. [42] Ellard S, Lango Allen H, De Franco E, et al. Improved genetic testing for monogenic diabetes using targeted nextgeneration sequencing. Diabetologia 2013;56(9):1958e63. [43] Pearson ER, Liddell WG, Shepherd M, et al. Sensitivity to sulphonylureas in patients with hepatocyte nuclear factor 1 alpha gene mutations: evidence for pharmacogenetics in diabetes. Diabet Med 2000;17:543e5. [44] Pearson ER, Starkey BJ, Powell RJ, et al. Genetic cause of hyperglycaemia and response to treatment in diabetes. Lancet 2003;362(9392):1275e81. [45] Tuomi T, Honkanen EH, Isomaa B, et al. Improved prandial glucose control with lower risk of hypoglycemia with nateglinide than with glibenclamide in patients with maturity-onset diabetes of the young type 3. Diabetes Care 2006;29(2): 189e94. [46] Bacon S, Kyithar MP, Rizvi SR, et al. Successful maintenance on sulfonylurea therapy and low diabetes complication rates in a HNF1A-MODY cohort. Diabet Med 2015. http://dx.doi.org/10.1111/dme.12992. [47] Shepherd M, Pearson ER, Houghton J, et al. No deterioration in glycemic control in HNF-1alpha maturity-onset diabetes of the young following transfer from long-term insulin to sulphonylureas. Diabetes Care 2003;26(11):3191e2. [48] Ostoft SH, Bagger JI, Hansen T, et al. Glucose-lowering effects and low risk of hypoglycemia in patients with maturityonset diabetes of the young when treated with a GLP-1 receptor agonist: a double-blind, randomized, crossover trial. Diabetes Care 2014;37(7):1797e805. [49] Stride A, Shields B, Gill-Carey O, et al. Cross-sectional and longitudinal studies suggest pharmacological treatment used in patients with glucokinase mutations does not alter glycaemia. Diabetologia 2014;57(1):54e6. [50] Pearson ER, Badman MK, Lockwood CR, et al. Contrasting diabetes phenotypes associated with hepatocyte nuclear factor1alpha and -1beta mutations. Diabetes Care 2004;27(5):1102e7. [51] Bowman P, Flanagan SE, Edghill EL, et al. Heterozygous ABCC8 mutations are a cause of MODY. Diabetologia 2012;55(1): 123e7. [52] Pearson ER, Flechtner I, Njolstad PR, et al., G. Neonatal Diabetes International Collaborative. Switching from insulin to oral sulfonylureas in patients with diabetes due to Kir6.2 mutations. N Engl J Med 2006;355(5):467e77. [53] Beltrand J, Elie C, Busiah K, et al. Sulfonylurea therapy benefits neurological and psychomotor functions in patients with neonatal diabetes owing to potassium channel mutations. Diabetes Care 2015;38(11):2033e41. [54] Rodriguez AJ, Mastronardi CA, Paz-Filho GJ. New advances in the treatment of generalized lipodystrophy: role of metreleptin. Ther Clin Risk Manag 2015;11:1391e400. [55] Hegele RA. Premature atherosclerosis associated with monogenic insulin resistance. Circulation 2001;103(18):2225e9. [56] Rosenbloom AL, Silverstein JH, Amemiya S, et al. Type 2 diabetes in children and adolescents. Pediatr Diabetes 2009; 10(Suppl. 12):17e32. [57] Inzucchi SE, Bergenstal RM, Buse JB, et al. Management of hyperglycaemia in type 2 diabetes: a patient-centered approach. Position statement of the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD). Diabetologia 2012;55(6):1577e96. [58] Jones KL, Arslanian S, Peterokova VA, et al. Effect of metformin in pediatric patients with type 2 diabetes: a randomized controlled trial. Diabetes Care 2002;25(1):89e94.

K.R. Owen / Best Practice & Research Clinical Endocrinology & Metabolism 30 (2016) 455e467

467

[59] Gottschalk M, Danne T, Vlajnic A, et al. Glimepiride versus metformin as monotherapy in pediatric patients with type 2 diabetes: a randomized, single-blind comparative study. Diabetes Care 2007;30(4):790e4. *[60] Group TS, Zeitler P, Hirst K, et al. A clinical trial to maintain glycemic control in youth with type 2 diabetes. N Engl J Med 2012;366(24):2247e56. [61] Turner RC, Cull CA, Frighi V, et al. Glycemic control with diet, sulfonylurea, metformin, or insulin in patients with type 2 diabetes mellitus: progressive requirement for multiple therapies (UKPDS 49). UK Prospective Diabetes Study (UKPDS) Group. JAMA 1999;281(21):2005e12. [62] Narasimhan S, Weinstock RS. Youth-onset type 2 diabetes mellitus: lessons learned from the TODAY study. Mayo Clin Proc 2014;89(6):806e16. [63] Klein DJ, Battelino T, Chatterjee DJ, et al. Liraglutide's safety, tolerability, pharmacokinetics, and pharmacodynamics in pediatric type 2 diabetes: a randomized, double-blind, placebo-controlled trial. Diabetes Technol Ther 2014;16(10): 679e87. [64] Monami M, Nardini C, Mannucci E. Efficacy and safety of sodium glucose co-transport-2 inhibitors in type 2 diabetes: a meta-analysis of randomized clinical trials. Diabetes Obes Metab 2014;16(5):457e66. [65] Nakamura Y, Nagai Y, Terashima Y, et al. Better response to the SGLT2 inhibitor dapagliflozin in young adults with type 2 diabetes. Expert Opin Pharmacother 2015:1e7. [66] Murray P, Chune GW, Raghavan VA. Legacy effects from DCCT and UKPDS: what they mean and implications for future diabetes trials. Curr Atheroscler Rep 2010;12(6):432e9.