Today ADA announced the 8 top abstracts. They are selected from 3100 abstracts submitted this year. They represent the best of the best cutting-edge research the ADA Meeting has to offer, covering a range of basic and clinical science related to important aspects of diabetes care.
Two abstracts about reserach in children, including the enviromental determinanats of diabetes in the Young (TEDDY) with prof Lernmark, Sweden. There is an association between cumulative negative stress from live events and islet autoimmunity in HLA DR3/4 risk-category children 6 years and younger.
Read also new feutures from Diabetes Prevention Program DPP Outcomes Study, exploring thre relationship between physical actitvity and the development of T2DM. After now more than 10 years of follow up, DPP investigators report findings of a significant association between physical activity and cunulative diabetes T2DM incidence.
www red DiabetologNytt
1. Cumulative Life Stress Is Associated with Islet Autoimmunity in HLA DR3/4 ChildrenView session detail
Author Block: SUZANNE B. JOHNSON, KRISTIAN LYNCH, WILLIAM HAGOPIAN, JEFFREY P. KRISCHER, AKE LERNMARK, MARIAN REWERS, JIN-XIONG SHE, JORMA TOPPARI, ANETTE ZIEGLER, ROSWITH L. ROTH, TEDDY STUDY GROUP, Tallahassee, FL, Tampa, FL, Seattle, WA, Malmö, Sweden, Aurora, CO, Augusta, GA, Turku, Finland, Neuherberg, Germany
Stress is considered a possible precursor to islet autoimmunity (IA) and type 1 diabetes (T1D). The Environmental Determinants of Diabetes in the Young (TEDDY) study is following 8676 genetically at risk children from 3 months until 15 years for development of IA and T1D. Life stress data are collected by parent report every 3 months until age 4, biannually thereafter. Using Cox regression, we examined the link between negative life events (NLE) and IA in TEDDY children ≤ 6 years old. We distinguished between life events negatively impacting the parent from events negatively impacting the child. We further separated NLEs into three categories: Early NLEs occurring during the first year after enrollment; Recent NLEs occurring in the year preceding IA; and Cumulative NLEs which include all NLEs preceding IA. Approximately 25% of children experience a NLE in any given year. Approximately 10% of children experience consistently high rates of NLEs year after year. After accounting for factors associated with IA (country of origin, family history of T1D, gender, maternal smoking), Cumulative Child NLEs were independently associated with increased risk for IA among HLA DR3/4 children who have the highest T1D risk (p < .003) but not among children from other HLA risk groups (interaction p<0.009, see Figure). As parent NLEs were not associated with IA, we conclude that child cumulative life stress is a risk factor for IA in HLA-DR3/4 children.
Authors: SUZANNE B. JOHNSON, KRISTIAN LYNCH, WILLIAM HAGOPIAN, JEFFREY P. KRISCHER, AKE LERNMARK, MARIAN REWERS, JIN-XIONG SHE, JORMA TOPPARI, ANETTE ZIEGLER, ROSWITH L. ROTH, TEDDY STUDY GROUP, Tallahassee, FL, Tampa, FL, Seattle, WA, Malmö, Sweden, Aurora, CO, Augusta, GA, Turku, Finland, Neuherberg, Germany
2. T Cell Mitochondrial Dysfunction in Human Type 1 DiabetesView session detail
Author Block: JING CHEN, ANNA V. CHERNATYNSKAYA, MATTHEW R. KIMBRELL, JIAN-WEI LI, BRITTNEY N. NEWBY, ADRIAN C. JOHNSTON, CLAYTON E. MATHEWS, Gainesville, FL
Proper mitochondrial (mt) functions are essential for T cell survive, activation and proliferation. Mitochondrial dysfunction in T cells has been associated with multiple human autoimmune diseases including SLE and rheumatoid arthritis. Animal studies also suggested that T cell mt dysfunction participates in type 1 diabetes (T1D). We detected T cell mt innermembrane hyperpolarization (MHP) in patients with T1D (P=0.027; n=29) compared to healthy controls (n=38). T cell MHP was not correlated with age, disease duration or HbA1c levels. Using a second cohort (39 T1D and 28 controls), T cell MHP in T1D was confirmed (P=0.0042).
Functional studies revealed that T cell MHP is associated with altered activation-induced IFN production (P=0.022), which is further linked with activation-induced mtROS generation (R2=0.097, P=0.023). T cells from T1D patients exhibit lower cellular ATP content at resting status (64.79 ± 8.706 vs 97.64 ± 12.33 pmol/mg protein, P=0.046). In vitro activation was able to increase ATP level in T cells from T1D. T cells from individuals positive for multiple autoantibodies are resistant to activation-induced apoptosis (AICD). CD8+ T cells from an individual with T cell MHP exhibited higher antigen specific cytotoxicity when compared to those from normal mt membrane potential. Expression of genes related with mt function, including those regulating membrane potential, fission and fussion, apoptosis, translocation and transport, were failed to upregulate in T cells from T1D patient after in vitro activation when compared to healthy control. Genetic analysis detected associations between T1D-risk alleles and T cell mt dysfunction, including MHP, mtROS, and AICD.
These data demonstrate that T cell mt dysfunction in T1D patients are intrinsic to T cells and result in abnormal T cell regulation. In conclusion, T cell mt dysfunction participates in human T1D pathogenesis.
Authors: JING CHEN, ANNA V. CHERNATYNSKAYA, MATTHEW R. KIMBRELL, JIAN-WEI LI, BRITTNEY N. NEWBY, ADRIAN C. JOHNSTON, CLAYTON E. MATHEWS, Gainesville, FL
3. Brain Gray Matter Volume Changes in Youth with Type 2 DiabetesView session detail
Author Block: JACOB M. REDEL, JENNIFER VANNEST, MEKIBIB ALTAYE, LAWRENCE M. DOLAN, GREGORY LEE, HERMINE BRUNNER, MARK DIFRANCESCO, SCOTT HOLLAND, CASSANDRA BRADY, AMY S. SHAH, Cincinnati, OH, Nashville, TN
Preliminary work suggests that youth with type 2 diabetes mellitus (T2DM) have changes in brain structure and poorer cognitive function scores compared to their peers. However, total and regional brain volume has not been comprehensively assessed. We examined differences in total and regional brain gray matter volumes in 20 youth with T2DM (mean duration 2.8± 2.1 years and mean hemoglobin A1c 7.9±2.1%) and 20 race, sex and age matched controls, all lacking neuropsychological disease or prior abnormal MRI. Comparisons were made using high-resolution T1-weighted structural MRI scans and voxel-based morphometry analysis. Statistical significance for each voxel was established at p <0.05, after controlling for multiple testing using a family-wise error rate procedure. Only clusters with ≥100 contiguous voxels are reported. Compared to controls, T2DM youth had decreased total gray matter volume (717±74 vs. 656 ±71 p= 0.012). In addition, T2DM youth had ten regions (within the temporal lobe, occipital lobe, parietal lobe, cingulate gyrus, and basal ganglia) with significantly less gray matter volume than controls and five regions (within the frontal lobe and basal ganglia) with significantly greater gray matter volume than controls. Youth with short duration of T2DM show significant differences in gray matter volume. Whether these findings explain poorer cognitive scores observed previously remains to be determined.
Authors: JACOB M. REDEL, JENNIFER VANNEST, MEKIBIB ALTAYE, LAWRENCE M. DOLAN, GREGORY LEE, HERMINE BRUNNER, MARK DIFRANCESCO, SCOTT HOLLAND, CASSANDRA BRADY, AMY S. S
4. Glucagon Stimulates Energy Expenditure and Reduces Obesity via Hepatic Farnesoid X Receptor in MiceView session detail
Author Block: TEAYOUN KIM, CHRISTINE M. LOYD, CASSIE L. HOLLEMAN, MARTIN E. YOUNG, RICHARD D. DIMARCHI, DIEGO PEREZ-TILVE, KIRK M. HABEGGER, Birmingham, AL, Bloomington, IN, Cincinnati, OH
Glucagon, an essential regulator of glucose homeostasis, also modulates lipid metabolism and promotes weight loss. Thus, novel single molecule coagonist drugs that include glucagon receptor (GcgR) agonist have emerged as promising therapeutic candidates for obesity and diabetes. We have previously shown that chronically enhanced GcgR agonism in mice induced hyperglycemia, yet concomitantly reduced body fat and plasma cholesterol. Notably, this treatment restored circulating bile acids while also stimulating expression of hepatic farnesoid X receptor (FXR). This nuclear bile acid receptor plays an important role in energy and lipid metabolism. We hypothesized that GcgR agonism regulates energy metabolism, at least in part, through FXR signaling. To test this hypothesis, we utilized whole body and liver-specific FXR knockout (FXR∆liver) mouse models. Chronic administration of a GcgR agonist (IUB288) predictably induced hyperglycemia and glucose intolerance regardless of genotype. However, this treatment prevented diet-induced obesity (DIO) in wild type (WT), but not FXR∆liver mice. IUB288-reduced body weight was greater in WT than FXR∆liver DIO mice. The body weight loss in WT mice was primarily due to fat mass loss (p<0.05) and was associated with a reduction in adipocyte size in WT mice, whereas this was not observed in FXR∆liver mice. Indirect calorimetry (CLAMS) analysis in this model revealed that IUB288 substantially increased energy expenditure (p<0.01) and respiration (p<0.05) in DIO WT mice, however these effects were lost in FXR∆liver mice. IUB288 reduced hepatosteatosis in WT but not FXR∆liver mice. Our data clearly demonstrate that GcgR agonism, via hepatic FXR, enhances energy expenditure and fatty acid oxidation in vivo. This heretofore-unappreciated aspect of glucagon biology has implications for the use of GcgR agonism in therapeutic strategies for multiple components of the metabolic syndrome.
Authors: TEAYOUN KIM, CHRISTINE M. LOYD, CASSIE L. HOLLEMAN, MARTIN E. YOUNG, RICHARD D. DIMARCHI, DIEGO PEREZ-TILVE, KIRK M. HABEGGER, Birmingham, AL, Bloomington, IN, Cincinnati, OH
5. Closed-Loop Glucagon Administration for the Automated Prevention and Treatment of Hypoglycemia in Type 1 DiabetesView session detail
Author Block: COURTNEY BALLIRO, LAYA EKHLASPOUR, FIRAS ELKHATIB, DEBBIE MONDESIR, MANASI SINHA, KENDRA MAGYAR, MALLORY HILLARD, RAJENDRANATH SELAGAMSETTY, LISA DAO, ARYAN ESMAEILI, STEVEN J. RUSSELL, Boston, MA, Riverside, CA
Tight glycemic regulation is difficult to achieve without hypoglycemia. The effectiveness of suspending insulin delivery based on data from a continuous glucose monitor (CGM) is limited by the continued absorption of already-delivered insulin. An alternative approach is automated delivery of micro-doses of glucagon.
We conducted a double-blinded, randomized, placebo-controlled crossover study involving 22 adult subjects with type 1 diabetes who used an insulin pump or multiple daily injections and who have reduced hypoglycemia awareness. Participants administered their own insulin as usual while receiving either glucagon or placebo (randomized daily) from an automated bionic pancreas. The primary outcome was area over the curve and <60 mg/dl (AOC<60), a measure of total hypoglycemia exposure.
On glucagon vs. placebo days AOC<60 mg/dl was reduced by 75% (851±748 vs. 3,414±2,242 mg/dl·min, p<0.0001) and there was a 91% reduction in AOC<60 at night (117±204 vs. 1,309±1,476 mg/dl·min, p<0.0001). There were half as many symptomatic hypoglycemia episodes on glucagon vs. placebo days (0.6±0.4 versus 1.2±0.8 incidents per day; p<0.0001), but no difference in mean CGM glucose (153±28 vs. 152±27 mg/dl, p=0.60). Self-reported nausea was not significantly different on glucagon vs. placebo days (1.1±0.6 vs. 0.4±0.7 cm on a 10 cm visual analog scale, p=0.053) and subjects correctly guessed their daily assignment to glucagon or placebo in a daily survey on only 42% of days. AOC measurements exhibited a significant decreasing trend over time in both groups (p=0.030) but there was no carry over effect from day to day (p=0.37). There was no unexpected or severe adverse events on either glucagon or placebo days. Automated glucagon administration effectively reduced hypoglycemia in patients with type 1 diabetes, and was well tolerated.
Authors: COURTNEY BALLIRO, LAYA EKHLASPOUR, FIRAS ELKHATIB, DEBBIE MONDESIR, MANASI SINHA, KENDRA MAGYAR, MALLORY HILLARD, RAJENDRANATH SELAGAMSETTY, LISA DAO, ARYAN ESMAEILI, STEVEN J. R
6. Physical Activity and Diabetes Development: The Diabetes Prevention Program (DPP) Outcomes StudyView session detail
Author Block: ANDREA M. KRISKA, BONNY ROCKETTE-WAGNER, SHARON L. EDELSTEIN, ELIZABETH M. VENDITTI, GEORGE A. BRAY, LINDA M. DELAHANTY, EDWARD S. HORTON, MARY A. HOSKIN, WILLIAM C. KNOWLER, DPP RESEARCH GROUP, Pittsburgh, PA, Rockville, MD, Baton Rouge, LA, Boston, MA, Phoenix, AZ
After more than ten years of DPP follow-up, cumulative diabetes incidence remained lower in the lifestyle compared to placebo and metformin groups and this could not be completely explained by weight change. Since physical activity (PA) was ascertained annually by the past-year Modifiable Activity Questionnaire and diabetes was determined annually by OGTT and semi-annually by fasting plasma glucose, we were able to examine whether PA contributed to the lower cumulative diabetes incidence observed in the lifestyle group.
All analyses were carried out in a subgroup of 1,796 participants who were also part of a DPP ancillary study. Cumulative PA was significantly higher in the lifestyle than in the metformin and placebo groups (P=0.003, in a mixed model of PA over the entire study).
The association between time-dependent PA and diabetes incidence was examined using Cox proportional hazards models controlled for age, sex, baseline PA, and baseline and time-dependent weight. Regardless of treatment group, over an average 11.9 years, there was a 2% (HR 0.98 [0.97, 0.99]; P<0.0001) decrease in diabetes incidence for each 6 MET-hrs/week increase in PA (equivalent to 1.5 more hours/week of a brisk walk). The protective effect of increasing PA was greater in participants reporting <150 minutes/week at baseline (5% decrease, n=728; HR 0.95 [0.93, 0.97]; P<0.0001) relative to those who already met that goal at baseline or if the analysis was limited to the lifestyle group alone.
In summary, although the DPP study was not designed to examine the separate effects of PA (or weight) on diabetes incidence, post-hoc analyses indicate that PA levels over the entire DPP and its Outcomes Study were inversely related to incident diabetes when controlled for weight and other important covariates. These results suggest that the lower cumulative diabetes incidence found in the lifestyle group across the entire study was partially explained by PA differences.
Authors: ANDREA M. KRISKA, BONNY ROCKETTE-WAGNER, SHARON L. EDELSTEIN, ELIZABETH M. VENDITTI, GEORGE A. BRAY, LINDA M. DELAHANTY, EDWARD S. HORTON, MARY A. HOSKIN, WILLIAM C. KNOWLER, DPP RESEARCH GROUP, Pittsburgh, PA, Rockville, MD, Baton Rouge, LA, Boston, MA, Phoenix, AZ
Author Block: SILVIO E. INZUCCHI, CATHERINE M. VISCOLI, LAWRENCE H. YOUNG, KAREN L. FURIE, MARK GORMAN, ANNE LOVEJOY, GREGORY G. SCHWARTZ, WALTER N. KERNAN, New Haven, CT, Providence, RI, Burlington, VT, Denver, CO
The Insulin Resistance Intervention after Stroke (IRIS) trial recently found that the thiazolidinedione drug, pioglitazone, reduced the risk of the primary composite outcome of fatal and non-fatal stroke or myocardial infarction, as well as the secondary outcome of progression to diabetes, in insulin resistant patients with cerebrovascular disease. This report analyzes pioglitazone’s effect on key glycemic parameters and on diabetes prevention within clinically relevant subgroups. 3876 non-diabetic patients (no prior history and fasting plasma glucose [FPG] <126 mg/dl) with recent ischemic stroke or transient ischemic attack and insulin resistance by the Homeostasis Model Assessment of Insulin Resistance (HOMA-IR >3.0) were randomly assigned to pioglitazone 45 mg QD vs. placebo. Surveillance for diabetes onset was by annual interview and FPG testing. At baseline, mean FPG, HbA1c, insulin, and HOMA-IR were 98.2 mg/dL, 5.8%, 22.4 uIU/ml, and 5.4, respectively. After 1 year, mean HOMA-IR decreased to 4.1 in the pioglitazone group and increased to 5.7 in the placebo group (p<0.0001); the corresponding 1-year FPG values were 95.1 and 99.7 mg/dl, respectively (p<0.0001). During 4.8 years, diabetes developed in 73 (3.8%) patients in the pioglitazone group vs. 149 (7.7%) patients in the placebo group (HR 0.48; 95% CI 0.33, 0.69; p<0.0001). The absolute risk reduction (ARR) was greatest for patients with FPG ≥100 mg/dl vs. <100 mg/dL (ARR 8.5% vs. 0.8%), HbA1c ≥5.7% vs. <5.7% (ARR 5.6% vs. 1.0%), and HOMA-IR ≥4.6 vs <4.6 (ARR 6.3% vs. 1.4%). Among insulin resistant but non-diabetic patients with cerebrovascular disease, pioglitazone decreases the risk of diabetes, especially in those with prediabetes (i.e., impaired fasting glucose or increased HbA1c) and worse insulin resistance. Pioglitazone is therefore the first glucose-lowering drug to prevent both progression to diabetes and major cardiovascular events as prespecified outcomes in a single trial.
Authors: SILVIO E. INZUCCHI, CATHERINE M. VISCOLI, LAWRENCE H. YOUNG, KAREN L. FURIE, MARK GORMAN, ANNE LOVEJOY, GREGORY G. SCHWARTZ, WALTER N. KERNAN, New Haven, CT, Providence, RI, Burlington, VT, Denver, CO
8. Novel Genetic Determinants of Diabetic Kidney DiseaseView session detail
Author Block: JENNIFER N. TODD, RANY SALEM, NIINA SANDHOLM, ERKKA A. VALO, LINDA T. HIRAKI, CHEN DI LIAO, MARCUS G. PEZZOLESI, ADAM SMILES, SUNA ONENGUT-GUMUSCU, WEI MIN CHEN, STUART MCGURNAGHAN, PAUL MCKEIGUE, AMY J. MCKNIGHT, ALEXANDER P. MAXWELL, HELEN M. COLHOUN, ANDRZEJ S. KROLEWSKI, ANDREW D. PATERSON, STEPHEN S. RICH, JOEL N. HIRSCHHORN, JOSE C. FLOREZ, Boston, MA, Cambridge, MA, Helsinki, Finland, Toronto, ON, Canada, Charlottesville, VA, Edinburgh, United Kingdom, Belfast, United Kingdom
Diabetic kidney disease (DKD) is a devastating diabetes complication, with known heritability not fully revealed by previous genetics studies.
We performed the largest genome-wide association study of type 1 DKD to date, in a 13-cohort consortium of 15,590 individuals of European ancestry genotyped on the Illumina HumanCoreExome Beadchip, which allows exploration of coding variation in addition to genomic markers. As prior work has shown that different characterizations of the DKD phenotype highlight distinct genetic associations, we investigated a spectrum of DKD definitions based on proteinuria and renal function criteria. Controls were DKD-free after a minimum of 15 years diabetes duration; cases had diabetes for at least 10 years prior to DKD diagnosis.
Our top finding was a missense mutation in COL4A3, rs55703767 (Asp326Tyr); the minor allele is common in Europeans (20%) and East Asians (13%) but not Africans (2%). This SNP had a genome-wide significant association with traditionally defined DKD (macroalbuminuria or end-stage renal disease [ESRD], (OR= 0.79, P=1.9×10-9), and a suggestive association with macroalbuminuria (OR= 0.79, P=1.6×10-6) and ESRD (OR= 0.79, P=4.5×10-5) individually. The association was consistent across cohorts (Phet=0.84). Though its PolyPhen score is 0.3 (benign), this SNP has been implicated as a splice site disruptor. The COL4A3 gene encodes the alpha 3 subunit of Type IV collagen, the major structural component of basement membranes. Pathogenic mutations in COL4A3 have been identified in thin basement membrane nephropathy, familial focal segmental glomerulosclerosis, and Alport syndrome. A proxy (r2=0.6) for rs55703767 had no significant associations in the CKDGen consortium, suggesting its pathogenicity occurs solely in the setting of hyperglycemia.
By significantly increasing sample size we have discovered a novel locus underlying DKD risk, paving the way for better understanding of pathology, prevention, and treatment.
Authors: JENNIFER N. TODD, RANY SALEM, NIINA SANDHOLM, ERKKA A. VALO, LINDA T. HIRAKI, CHEN DI LIAO, MARCUS G. PEZZOLESI, ADAM SMILES, SUNA ONENGUT-GUMUSCU, WEI MIN CHEN, STUART MCGURNAGHAN, PAUL MCKEIGUE, AMY J. MCKNIGHT, ALEXANDER P. MAXWELL, HELEN M. COLHOUN, ANDRZEJ S. KROLEWSKI, ANDREW D. PATERSON, STEPHEN S. RICH, JOEL N. HIRSCHHORN, JOSE C. FLOREZ, Boston, MA, Cambridge, MA, Helsinki, Finland, Toronto, ON, Canada, Charlottesville, VA, Edinburgh, United Kingdom, Belfast, United Kingdom