NYHETER Läkartidningen 180302
Katrin Trysell www.lakartidningen.se skriver
Svenska forskare vill se ny klassiﬁcering av diabetes hos vuxna
Dela in diabetes som debuterar i vuxen ålder i fem olika subtyper för att bättre kunna skräddarsy behandling. Det föreslår svenska och finska forskare bakom en ny studie i Lancet Diabetes and Endocrinology
Fyra distinkta subtyper av typ 2-diabetes och en autoimmun form av diabetes med olika sjukdomsutveckling och risk för diabeteskomplikationer – det identifierade forskarna genom att analysera data för närmare 15 000 patienter över 18 år med nydebuterad diabetes i Sverige och Finland.
En av dem som står bakom studien i Lancet Diabetes & Endocriology är professor Leif Groop, bland annat verksam vid Lunds universitet
– Det finns evidens för att tidig behandling vid diabetes är avgörande för att förhindra komplikationer som förkortar livslängden. Att diagnostisera diabetes mer exakt kan ge oss värdefulla insikter om hur den kommer att utvecklas med tiden och förutse och behandla komplikationer innan de utvecklas, säger han i ett pressmeddelande.
I analysen tittade forskarna på ålder, BMI, långtidsblodsockret HbA , betacellsfunktion, insulinresistens och förekomst av antikroppar som förknippas med autoimmun diabetes. De gjorde också genetiska analyser och jämförde progressionen av sjukdomen, behandling och utveckling av diabeteskomplikationer.
Ur detta utkristalliserades tre allvarliga och två lindriga former av diabetes.
En allvarlig form av diabetes – typ 3 eller allvarlig insulinresistent diabetes – innebar allvarlig insulinresistens och högre risk för njurkomplikationer.
En annan – typ 2 eller diabetes med allvarlig insulinbrist – omfattade relativt unga individer med insulinbrist och dålig metabol kontroll utan autoantikroppar.
Den tredje svåra typen – typ 1 eller allvarlig autoimmun diabetes – omfattade patienter med insulinbrist och antikroppar som förknippas med autoimmun diabetes.
Den vanligaste typen av diabetes var en av de mer lindriga formerna – typ 5 eller lindrig åldersrelaterad diabetes – som sågs hos 39 till 47 procent av patienterna.
Den andra lindriga formen – typ 4 eller lindrig fetmarelaterad diabetes – sågs främst hos feta individer.
Alla fem subtyperna av diabetes var också genetiskt distinkta.
Forskarna såg också att många patienter inte fick en lämplig behandling utifrån sin risk för komplikationer med mera. Utifrån det drar de slutsatsen att dagens klassificering av diabetes inte förmår att skräddarsy behandling för patienter med sjukdomen på ett bra sätt.
De betonar dock att det behövs ytterligare studier för att testa och förena de fem typerna av diabetes genom att inkludera även information om till exempel blodlipider, blodtryck och genetiskt riskscore.
Adult-onset diabetes consists of five types of disease that have different physiological and genetic profiles, rather than the traditional type 1 and 2 classification, say Scandinavian researchers, findings that could bring the promise of personalized medicine a step closer.
Gathering data on almost 15 000 patients from across five cohorts in Sweden and Finland, they found that using six standard measurements identified five clusters of patients with diabetes.
These divided into three severe and two mild forms of disease:
• one corresponding to type 1 diabetes
• the remaining four representing subtypes of type 2 diabetes.
The clusters included one of very insulin-resistant individuals at significantly higher risk of diabetic nephropathy, another of relatively young insulin deficient individuals with poor metabolic control (high HbA1c), and a large group of elderly patients with the most benign disease course.
Crucially, treatment often did not correspond to the type of diabetes.
The research, published online March 1 in the Lancet Diabetes & Endocrinology, could have important implications not only for the diagnosis and management of diabetes but for future therapeutic guidance.
"Existing treatment guidelines are limited by the fact they respond to poor metabolic control when it has developed, but do not have the means to predict which patients will need intensified treatment," lead author Leif Groop, MD, PhD, Lund University Diabetes Center, Malmö, Sweden, and Folkhalsan Research Centre, Helsinki, Finland, said in a press release by the journal.
"This study moves us towards a more clinically useful diagnosis, and represents an important step towards precision medicine in diabetes."
In an accompanying editorial, Rob Sladek, MD, McGill University and Genome Quebec Innovation Centre, Montreal, Canada, points out that future studies will have to take into account the effect of age on patient outcomes, and that other factors not included in the current analysis may also have an impact.
"Nevertheless, the finding that simple parameters assessed at the time of diagnosis could reliably stratify patients with diabetes according to prognosis is compelling and poses the challenge of development of methods to predict outcomes of patients with type 2 diabetes that are more generalizable and comprehensive," he writes.
"Additionally, the physiological basis of the features characterizing each cluster provides a strong rationale to investigate the genetic architecture and molecular mechanisms that lead to heterogeneity in the presentation and progression of diabetes in adults."
Sladek told Medscape Medical News that he "was not completely surprised" that there were as many as five clusters of diabetes.
"We already know that there is a group of adult-onset patients that are severely insulin deficient. In addition, we think of diabetes as being a balance between insulin needs or insulin resistance, say from obesity, and insulin production," he said.
"So I might have expected that a couple [of the] groups would identify patients with insulin resistance."
Clusters 1 and 2 Had Highest HbA1c Levels
Diabetes is currently classified as type 1 diabetes, type 2 diabetes, and a number of less common diseases such as latent autoimmune diabetes in adults (LADA), maturity-onset diabetes in the young (MODY), and secondary diabetes.
The classification of diabetes into type 1 and type 2 relies predominantly on the presence or absence, respectively, of autoantibodies against pancreatic beta-cell antigens and younger age. On this basis, 75% to 85% of patients are identified as having type 2 diabetes.
Recent research on glutamate acid decarboxylase antibodies (GADA) and gene sequencing has demonstrated that type 2 diabetes in particular is highly heterogeneous.
Furthermore, Groop noted, "evidence suggests that early treatment for diabetes is crucial to prevent life-shortening complications."
"More accurately diagnosing diabetes could give us valuable insights into how it will develop over time, allowing us to predict and treat complications before they develop."
The researchers therefore set out to establish a more refined diabetes classification that could allow individualized treatment and identify patients at diagnosis who are most at risk of complications.
They gathered data from five cohorts: Swedish All New Diabetics in Scania (ANDIS), Scania Diabetes Registry (SDR), All New Diabetics in Uppsala (ANDIU), Diabetes Registry Vaasa (DIREVA), and Malmö Diet and Cancer Cardiovascular Arm (MDC-CVA).
The team used six variables to conduct a data-driven cluster analysis of 8980 patients from ANDIS, all of whom were newly diagnosed with diabetes between 2008 and 2016.
Variables included the presence of GADA; age at diagnosis; body mass index (BMI); HbA1c; and homeostatic model assessment 2 (HOMA2) estimates of beta-cell function (HOMA2-B) and insulin resistance (HOMA2-IR), based on C-peptide concentrations (which performs better than insulin in patients with diabetes) calculated using the HOMA calculator.
The analysis revealed the presence of five clusters of diabetes in men and women, with similar distributions between the two, as shown in the table.
Five Clusters of Diabetes
Cluster N (%) Characteristics Name
Cluster (1,2,3,4,5) N (%) Characterics Name
1. 577 (6.4) Early disease onset (at a young age), essentially corresponds with type 1 diabetes and LADA, relatively low BMI, poor metabolic control, insulin deficiency (impaired insulin production), GADA+
Severe autoimmune diabetes (SAID)
2. 1575 (17.5) Similar to cluster 1 but GADA–, high HbA1c, highest incidence of retinopathy Severe insulin-deficient diabetes (SIDD)
3. 1373 (15.3) Insulin resistance, high BMI, highest incidence of nephropathy Severe-insulin resistant diabetes (SIRD)
4, 1942 (21.6) Obesity, younger age, not insulin resistant Mild obesity-related diabetes (MOD)
5 .3513 (39.1) Older age, modest metabolic alterations Mild age-related diabetes (MARD)
ANDIS: Swedish All New Diabetics in Scania; BMI: body mass index; GADA: glutamic acid decarboxylase antibodies; LADA: latent autoimmune diabetes in adults. N in ANDIS cohort.
Researchers then tested the clusters in 1466 patients from SDR, 844 patients from ANDIU and 3485 patients from DIREVA, and identified similar patient distributions and cluster characteristics.
Looking at disease progression and treatment, the team found that clusters 1 and 2 had substantially higher HbA1c levels than the other clusters, which persisted throughout follow-up.
Patients in clusters 1 and 2 were also more likely to have ketoacidosis at diagnosis (31% and 25%) compared with other clusters (< 5%), of which HbA1cwas the strongest predictor (odds ratio [OR] per SD change, 2.73; P < .0001).
Insulin was prescribed to 42% of cluster 1 patients and 29% of cluster 2 patients, but to less than 4% of patients in other clusters. The time to sustained insulin use was also shortest in these two clusters.
A First Step Towards Precision Medicine in Diabetes
Metformin use was highest in cluster 2 and lowest in clusters 1 and 3. Kidney function and adverse reactions did not have a major effect on metformin use.
Cluster 3 was at highest risk of developing chronic disease, at a mean follow-up of 3.9 years. This cluster also had a higher risk of diabetic nephropathy and macroalbuminuria than other patients (hazard ratio [HR], 2.18; P = .0026).
Patients in cluster 3 also had a substantially higher risk of end-stage renal disease (HR vs cluster 5, 4.89; P < .0001).
Diabetic retinopathy was more common in cluster 2, at an OR of 1.6 vs cluster 5 (in ANDIS).
The team also reports that there was no one genetic variant associated with all five clusters and that each cluster had a genetic profile distinct from that of type 2 diabetes as a whole.
While acknowledging that their study has several limitations and needs confirmation in other, less homogenous populations, researchers say that the combined information provided by the variables in their analysis is "superior to measurement of only one metabolite, glucose."
"Through combining this information from diagnosis with information in the health care system, this study provides a first step towards a more precise, clinically useful stratification," they continue.
"This new substratification could change the way we think about type 2 diabetes and help to tailor and target early treatment to patients who would benefit most, thereby representing a first step towards precision medicine in diabetes," they add.
The team also believe the clusters they identified "can easily be applied to both existing diabetes cohorts (for example, from drug trials) and patients in diabetes clinics."
"A web-based tool to assign patients to specific clusters, provided the appropriate variables have been measured, is under development," they note.
How Do These Data Relate to Type 3c Diabetes?
The current results follow those of a study published in late 2017, which showed that patients with diabetes resulting from pancreatic dysfunction — type 3c diabetes — are often misdiagnosed as having type 2 diabetes.
As reported by Medscape Medical News, that analysis of more than 30 000 incident diabetes cases showed that type 3c diabetes, also known as diabetes of the exocrine pancreas, is almost as common as type 1 diabetes and misdiagnosed in over 87% of patients.
The misdiagnosis results in an increased risk of poor glycemic control compared with patients with type 2 diabetes and a far greater reliance on insulin.
Sladek said the current study is not directly related to type 3c diabetes, as this latter, rarer form of the disease is considered a 'secondary diabetes.' "In other words, it occurs as a result of a well-recognized disease process that does not alter insulin secretion directly and independently," he said.
Noting that the patients with type 3c diabetes had some form of exocrine pancreatic disease that affected endocrine function, Sladek said "this could occur for many reasons that share the common feature that the exocrine pancreas damage precedes the development of diabetes." "In contrast, patients with type 1 or 2 diabetes have normal exocrine pancreatic function."
This study was funded by the Swedish Research Council, European Research Council, Vinnova, Academy of Finland, Novo Nordisk Foundation, Vasa Hospital District, Scania University Hospital, Sigrid Juselius Foundation, European Union, Swedish Foundation for Strategic Research, Jakobstadsnejden Heart Foundation, Folkhalsan Research Foundation, and Ollqvist Foundation. It was conducted by researchers from Lund University, Skåne University Hospital, Vaasa Central Hospital, Vaasa Health Care Center, Uppsala University, Lund University Hospital, Folkhalsan Research Center, Helsinki University Central Hospital, University of Helsinki, and University of Gothenburg.
Lancet Diabetes Endocrinol. Published online March 1, 2018.
Novel subgroups of adult-onset diabetes and their association with outcomes: a data-driven cluster analysis of six variables
Diabetes is presently classified into two main forms, type 1 and type 2 diabetes, but type 2 diabetes in particular is highly heterogeneous. A refined classification could provide a powerful tool to individualise treatment regimens and identify individuals with increased risk of complications at diagnosis.
We did data-driven cluster analysis (k-means and hierarchical clustering) in patients with newly diagnosed diabetes (n=8980) from the Swedish All New Diabetics in Scania cohort. Clusters were based on six variables (glutamate decarboxylase antibodies, age at diagnosis, BMI, HbA1c, and homoeostatic model assessment 2 estimates of β-cell function and insulin resistance), and were related to prospective data from patient records on development of complications and prescription of medication. Replication was done in three independent cohorts: the Scania Diabetes Registry (n=1466), All New Diabetics in Uppsala (n=844), and Diabetes Registry Vaasa (n=3485). Cox regression and logistic regression were used to compare time to medication, time to reaching the treatment goal, and risk of diabetic complications and genetic associations.
We identified five replicable clusters of patients with diabetes, which had significantly different patient characteristics and risk of diabetic complications. In particular, individuals in cluster 3 (most resistant to insulin) had significantly higher risk of diabetic kidney disease than individuals in clusters 4 and 5, but had been prescribed similar diabetes treatment. Cluster 2 (insulin deficient) had the highest risk of retinopathy. In support of the clustering, genetic associations in the clusters differed from those seen in traditional type 2 diabetes.
We stratified patients into five subgroups with differing disease progression and risk of diabetic complications. This new substratification might eventually help to tailor and target early treatment to patients who would benefit most, thereby representing a first step towards precision medicine in diabetes.
From the article
Diabetes is the fastest increasing disease worldwide and a substantial threat to human health.1 Existing treatment strategies have been unable to stop the progressive course of the disease and prevent development of chronic diabetic complications. One explanation for these shortcomings is that diagnosis of diabetes is based on measurement of only one metabolite, glucose, but the disease is heterogeneous with regard to clinical presentation and progression.
Diabetes classication into type 1 and type 2 diabetes relies primarily on the presence (type 1 diabetes) or absence (type 2 diabetes) of autoantibodies against pancreatic islet β-cell antigens and age at diagnosis (younger for type 1 diabetes). With this approach, 75–85% of patients are classi ed as having type 2 diabetes. A third subgroup, latent autoimmune diabetes in adults (LADA; a ecting <10% of people with diabetes), de ned by the presence of glutamic acid decarboxylase antibodies (GADA), is phenotypically indistinguishable from type 2 diabetes at diagnosis, but becomes increasingly similar to type 1 diabetes over time.2 With the introduction of gene sequencing in clinical diagnostics, several rare monogenic forms of diabetes were described, including maturity- onset diabetes of the young and neonatal diabetes.3,4
Existing treatment guidelines are limited by the fact they respond to poor metabolic control when it has developed, but do not have means to predict which patients will need intensi ed treatment. Evidence suggests that early treatment is crucial for prevention of life-shortening complications because target tissues seem to remember poor metabolic control decades later (so-called metabolic memory).5,6
A re ned classi cation could provide a powerful tool to identify at diagnosis those at greatest risk of complications and enable individualised treatment regimens in the same way as genetic diagnosis of monogenic diabetes guides clinicians to optimal treatment.7 With this aim, we present a novel diabetes classi cation based on unsupervised, data-driven cluster analysis of six com- monly measured variables and compare it metabolically, genetically, and clinically to the current classi cation in four separate populations from Sweden and Finland.
Taken together, the results of our study suggest that this new clustering of patients with adult-onset diabetes is superior to the classic diabetes classi cation because it identi es patients at high risk of diabetic complications at diagnosis and provides information about underlying disease mechanisms, thereby guiding choice of therapy. By contrast with previous attempts to dissect the hetero- geneity of diabetes,23 we used variables re ective of key aspects of diabetes that are monitored in patients. Thus, this clustering can easily be applied to both existing diabetes cohorts (eg, from drug trials) and patients in diabetes clinics. A web-based tool to assign patients to speci c clusters, provided the appropriate variables have been measured, is under development.
Whereas SAID overlapped with type 1 diabetes and LADA, SIDD and SIRD represent two new, severe forms of diabetes previously masked within type 2 diabetes. It would be reasonable to target individuals in these clusters with intensi ed treatment to prevent diabetic complications. The risk of kidney complications was substantially increased in patients with SIRD, reinforcing the association between insulin resistance and kidney disease.24 Insulin resistance has been associated with increasedsaltsensitivity,glomerularhypertension,hyper- ltration, and reduced renal function, all hallmarks of diabetic kidney disease.25 The increased incidence of diabetic kidney disease in this study was in spite of reasonably low HbA1c, suggesting that glucose-lowering therapy is not the optimum way of preventing this complication. In support of this hypothesis, mice with podocyte-speci c knockout of the insulin receptor, mimicking the reduced insulin signaling seen in patients who are insulin resistant, developed diabetic kidney disease, even during normoglycaemic conditions.26 Although di erences in retinopathy were not as pro- nounced as for diabetic kidney disease, insulin de ciency or hyperglycaemia appeared to be important triggers of retinopathy, with the highest prevalence observed in cluster 2 (SIDD).
The fact that clustering led to similar results in newly diagnosed patients and patients with longer-term diabetes, and that C-peptide remained relatively stable over time (appendix), suggests that the clusters are stable and at least partially mechanistically distinct rather than representing di erent stages of the same disease. The di erences in genetic associations also support this view. In particular, the absence of associations between the genetic risk scores for type 2 diabetes and insulin secretion and SIRD indicate that this group might have a di erent aetiology to the other clusters. Hepatic insulin resistance seems to be a feature of non-alcoholic fatty liver disease, because the SNP in the TM6SF2 gene usually associated with non-alcoholic fatty liver disease was associated with SIRD in this study, but not with MOD.
We cannot at this stage claim that the new clusters represent different aetiologies of diabetes, nor that this clustering is the optimal classi cation of diabetes subtypes. Additionally, whether patients (particularly from the periphery of clusters) can move between clusters needs to be shown in future prospective studies, and the exact overlap of weaker association signals will need to be investigated in larger cohorts. It might be possible to re ne the strati cation further through inclusion of additional cluster variables, such as biomarkers, genotypes, or genetic risk scores. Future genome-wide association studies might also be able to better describe the genetic architecture of the di erent clusters and establish the inherited proportion of each cluster with heritability partitioning models.27 This classi cation was derived primarily with patients from northern Europe, with limited non-Scandinavian representation, and the applicability of this strategy to patients of other ethnicities needs to be assessed. Only two types of autoantibodies were measured, and the e ects of other antibodies on clustering performance are unknown. Moreover, we did not have data on some known risk factors for diabetic complications, such as blood pressure and blood lipids, and could therefore not include these in the analysis.
In conclusion, our data suggest that the combined information from a few variables central to the development of diabetes is superior to measurement of only one metabolite, glucose. Through combining this information from diagnosis with information in the health-care system, this study provides a rst step towards a more precise, clinically useful strati cation, representing an important step towards precision medicine in diabetes. This clustering also paves the way for randomised trials targeting insulin secretion in SIDD and insulin resistance in SIRD.
Kommentar från UK www
”However, this study alone is not sufficient to lead to changes in diabetes treatment guidelines, as it was only based on groups of diabetes patients in Scandinavia.
The clusters and associated complications will need to be verified in other populations, including other ethnicities that may have a different risk of diabetes, such as Asian populations. Analysis by Bazian. Edited by NHS Choices”
www red DiabetologNytt