Glutamate decarboxylase antibody levels predict rate of β-cell decline in adult-onset diabetes

doi:10.1016/0168-8227(95)01026-A

Diabetes Research and Clinical Practice

Volume 27, Issue 2, February 1995, Pages 133-140

https://doi.org/10.1016/0168-8227(95)01026-A Get rights and content

Abstract

Glutamate decarboxylase autoantibodies (GAD65Ab) and β-cell function were evaluated at and 3 years after diabetes onset in consecutive subjects over 15 years of age. At onset, $2132$ (66%) insulin-treated patients (mean age 43, range 16–79 years) had GAD65Ab; all GAD65Ab persisted 3 years later. At onset, $2082$ (24%) non-insulin-treated patients (mean age 56, range 20–79 years) had GAD65Ab. Of those with persistent GAD65Ab, 8 non-insulin-treated and 11 insulin-treated patients consented to follow-up glucose and glucagon stimulation tests. For non-insulin-treated patients, quantitative GAD65Ab index at onset correlated inversely with 1+3 min C-peptide response to glucose (r = −0.68, P < 0.05) and to glucagon (r = −0.79, P < 0.05) 3 years later. Those with high (> 0.50) initial GAD65Ab index had lower C-peptide (fasting, 1+3 min after glucose and after glucagon) 3 years later, versus those with low (<0.50) initial GAD65Ab index (P < 0.05). In conclusion, not only did GAD65Ab presence predict future insulin dependence, but higher GAD65Ab levels may mark more rapid decline in β-cell function in apparent non-insulin-dependent diabetes.

References (41)

G.F. Bottazzo et al.
ICA in diabetes mellitus with autoimmune polyendocrine deficiencies
Lancet
(1974)
A.C. MacCuish et al.
Antibodies to pancreatic islet-cells in IDDM with coexistent autoimmune disease
Lancet
(1974)
M.A. Atkinson et al.
64,000 Mr autoantibodies as predictors of IDDM
Lancet
(1990)
E. Bonifacio et al.
Quantification of islet cell antibodies and prediction of insulin dependent diabetes
Lancet
(1990)
K.M. Spencer et al.
Fluctuating ICA in unaffected relatives of patients with IDDM
Lancet
(1984)
A. Elamin et al.
ICA and endogenous insulin secretion in Sudanese diabetic children
Diabetes Res. Clin. Pract.
(1992)
W. Gepts
Pathologic anatomy of the pancreas in juvenile diabetes mellitus
Diabetes
(1965)
W.J. Irvine et al.
Pancreatic islet cell antibodies in diabetes mellitus correlated with the duration and type of diabetes, coexistent autoimmune disease and HLA-type
Diabetes
(1977)
J.P. Palmer et al.
Insulin antibodies in insulin-dependent diabetics before insulin treatment
Science
(1983)
S. Baekkeskov et al.
Autoantibodies in newly diagnosed diabetic children immunoprecipitate human pancreatic islet cell proteins
Nature
(1982)

S. Baekkeskov et al.

Identification of the 64K autoantigen in IDDM as the GABA-synthesizing enzyme glutamic acid decarboxylase

Nature

(1990)

A. Karlsen et al.

Recombinant GAD representing the single isoform expressed in human islets detects IDDM-associated 64K autoantibodies

Diabetes

(1992)

H. Bärmeier et al.

Risk for developing IDDM and the presence of islet 64K antibodies

Diabetologia

(1991)

M. Christie et al.

Persistence of antibodies to a 64,000-Mr islet cell protein after onset of type 1 diabetes

Diabetes

(1990)

M.J. Rowley et al.

Antibodies to GAD discriminate major diabetes type

Diabetes

(1992)

M. Landin-Olsson et al.

Islet cell antibodies and fasting C-peptide predict insulin requirement at diagnosis of diabetes mellitus

Diabetologia

(1990)

M. Landin-olsson et al.

Islet cell and thyrogastric antibodies in 633 consecutive 15–34 year old patients in the Diabetes Incidence Study in Sweden (DISS)

Diabetes

(1992)

A. Gottsäter et al.

β-Cell function in relation to ICA during the first 3 years after clinical diagnosis of diabetes in Type II diabetics

Diabetes Care

(1993)

J.P. Palmer

Predicting IDDM: use of humoral immune markers

Diabetes Rev.

(1993)

W.A. Hagopian et al.

Quantitative assay using recombinant human islet glutamic acid decarboxylase (GAD-65) shows 64K ab positivity at onset predicts diabetes type

J. Clin. Invest.

(1993)

Cited by (60)

Prevention of autoimmune disease: The type 1 diabetes paradigm
2019, The Autoimmune Diseases
Prevention of autoimmune disease based on a mechanistic understanding of pathogenesis will be the ultimate triumph of immunological good over evil. Identifying high-risk individuals before or soon after the onset of pathology and accurately predicting subsequent clinical disease is a prerequisite for prevention. For this reason alone, type 1 diabetes (T1D) is arguably the paradigm for autoimmune disease prevention. Autoimmunity against pancreatic islet insulin-producing beta cells in T1D is associated with specific human leukocyte antigen (HLA) haplotypes that denote high genetic risk, a basis for primary prevention. The majority of children and young adults who will develop T1D can be identified by the presence of circulating islet autoantibodies before the age of three, a basis for secondary prevention. The gut microbiome, a bellwether of the environment as well as genotype, is altered before and after the onset of islet autoimmunity. The rising incidence of T1D is promoted by environmental factors on a background of genetic susceptibility, raising the possibility of primary prevention through environment–lifestyle modification. Certain agents are effective in some autoimmune diseases, for example, immune biologics in rheumatoid arthritis, but multiple prevention trials of immune and nonimmune agents in recent-onset T1D have failed to avert loss of residual beta-cell function in the longer term. The problem is that the onset of pathology and the onset of symptoms in T1D are usually many months or years apart. Recognizing that T1D is primarily an autoimmune beta-cell disorder that progresses to a metabolic syndrome only in its end-stage will expand therapeutic options for earlier intervention and improve prospects of prevention.
Prevention of Autoimmune Disease: The Type 1 Diabetes Paradigm
2013, The Autoimmune Diseases: Fifth Edition
Prevention of autoimmune disease based on a mechanistic understanding of pathogenesis would be the ultimate triumph of immunological good over evil. Until a therapeutic intervention is shown to be completely safe, accurate prediction of clinical disease will be a prerequisite for prevention. For this reason alone, type 1 diabetes (T1D) is arguably the best paradigm for prevention. Although other autoimmune diseases are associated with the appearance of autoantibodies before clinical presentation, T1D exemplifies par excellence the application of autoantibody markers to preclinical diagnosis and prediction (in T1D relatives), and identification of high-risk individuals as candidates for prevention. The rising incidence of T1D and other autoimmune diseases is driven by environmental factors on a background of genetic susceptibility, implying that some degree of primary prevention, before the onset of disease, is feasible. Secondary prevention, after initiation of pathogenic autoimmunity but before clinical presentation, places a high premium on therapeutic modalities that are safe, which to date are applied only in T1D in the form of autoantigen-based vaccination to induce immune tolerance. Tertiary “prevention,” after clinical presentation, equates with treatment of established disease. Certain agents are effective in slowing progression of some autoimmune diseases, e.g., biologics in rheumatoid arthritis, but multiple trials of immune and non-immune agents in T1D have failed to prevent loss of residual β cell function over the long term. Systematic study of the natural histories of autoimmune diseases from early life in relation to environmental exposures, together with the development of safer immunotherapeutic approaches and lessons learnt from ongoing clinical trials in T1D, should improve the prospects for autoimmune disease prevention.
The age of onset of diabetes and glutamic acid decarboxylase titer measured long after diagnosis are associated with the clinical stage of slow-onset type 1 diabetes
2013, Diabetes Research and Clinical Practice
Citation Excerpt :
In other words, the time from the onset of disease to each clinical stage did not differ greatly among stages I–III. The follow-up periods of previous prospective studies that examined the relationship between GAD-Ab and β-cell function were 2–6 years [13–18] except for a single longer study (12 years) [11] that did not indicate a cut-off point for GAD-Ab or consider the age of onset. The speed of β-cell destruction is slower in patients with stages I and II than in those with stage III.
Diabetes mellitus is divided into 3 clinical stages: not insulin requiring, insulin requiring for control, and insulin requiring for survival. We investigated the clinical characteristics of patients with slow-onset type 1 diabetes (T1D) to examine which clinical factors influence the clinical stage.
One hundred fifty patients with slow-onset T1D were divided into 3 groups based on disease stage, and clinical features were compared among these groups. The patients were also divided into 4 groups based on the age of onset and the glutamic acid decarboxylase antibody (GAD-Ab) titer, which was measured long after diagnosis (mean, 9.2 years). The frequencies of the 3 stages were compared among these 4 groups.
The age of onset and the log (GAD-Ab) titer differed significantly among the 3 stages. The number of patients not requiring insulin was significantly higher and the number of those requiring insulin for survival was significantly lower in the group in which the age of onset was ≥50 and the log (GAD-Ab) titer <0.6, while the opposite pattern was observed in the group in which the age of onset was <50 and the log(GAD-Ab) titer ≥0.6.
Our results suggest that the combination of the age of onset and GAD-Ab titer measured long after diagnosis might predict the clinical stage of slow-onset T1D.
Clinical and metabolic characteristics of patients with latent autoimmune diabetes in adults (LADA): Absence of rapid beta-cell loss in patients with tight metabolic control
2010, Diabetes and Metabolism
The present study compared the clinical and metabolic characteristics of latent autoimmune diabetes in adults (LADA) with type 2 diabetes, as well as the residual beta-cell function and progression to insulin treatment, over a 2-year follow-up period, of antibody (Ab)-positive and Ab-negative patients who achieved tight glycaemic control (HbA_1c 7.0 ± 0.8% and 6.5 ± 0.9%, respectively, at the time of entry into the study).
Glutamic acid decarboxylase antibodies (GADA) and/or islet cell antibodies (ICA) were detected in 10% of patients presenting with non-insulin-dependent diabetes. Around half of Ab-positive patients required insulin treatment during the follow-up. Ab-positive patients displayed lower stimulated C-peptide levels both at entry and during the follow-up compared with Ab-negative patients, although no significant decline in C-peptide levels was observed in either subgroup over two years. Nevertheless, Ab-positive patients progressed more frequently to insulin treatment, and stimulated C-peptide tended to decrease in LADA patients who subsequently required insulin, whereas it remained stable in those who were non-insulin-dependent. In those who progressed, the trend towards C-peptide decline persisted even after starting insulin treatment.
LADA patients demonstrate lower residual beta-cell function than do type 2 diabetes patients. However, those who achieve tight metabolic control do not present with a rapid decline in beta-cell function. Further studies are needed to determine the optimal treatment strategy in such patients.
Nous avons (1) comparé les caractéristiques cliniques et métaboliques de patients ayant un diabète de type 1 lent et de patients diabétiques de type 2 et (2) étudié prospectivement sur deux ans l’insulinosécrétion résiduelle et l’évolution vers l’insulinorequérance de patients ayant ou non des autoanticorps et ayant un diabète bien contrôlé (HbA_1c initiale 7,0 ± 0,8 % et 6,5 ± 0,9 %, respectivement).
Dix pour cent des diabétiques non insulinodépendants inclus avaient des anticorps (Ac) anti-GAD et/ou des ICA. La moitié des patients Ac positifs ont nécessité une insulinothérapie au cours du suivi. Le peptide C stimulé était plus bas chez les patients Ac positifs que chez les patients Ac négatifs, mais dans aucun des deux sous-groupes, nous n’avons observé de décroissance significative du peptide C au cours du suivi de deux ans. Les patients Ac positifs ont nécessité plus fréquemment une insulinothérapie et le peptide C tendait à décroître chez les patients évoluant vers l’insulinorequérance, tandis qu’il restait stable chez ceux restant non insulinodépendants. Chez les patients insulinorequérants, la tendance à la décroissance du peptide C persistait après initiation de l’insuline.
Les diabétiques de type 1 lent ont une insulinosécrétion résiduelle plus faible que les diabétiques de type 2. Néanmoins, lorsque le diabète est bien équilibré, on n’observe pas de décroissance rapide de l’insulinosécrétion. La stratégie thérapeutique optimale chez ces patients reste à définir.
Multiple factors affect the loss of measurable C-peptide over 6 years in newly diagnosed 15- to 35-year-old diabetic subjects
2007, Journal of Diabetes and its Complications
Citation Excerpt :
The latter patients were two to four times more likely to lose C-peptide over a 6-year period following diagnosis than autoantibody-negative patients, depending on the number of autoantibodies present at the time of diagnosis. These findings are consistent with several previously published studies demonstrating that autoantibody positivity confers an increased risk for progression to insulin dependence (Landin-Olsson et al., 1990; Törn, Landin-Olsson, Ostman et al., 2000) and loss of C-peptide (Gottsäter et al., 1993; Gottsäter et al., 1995; Sabbah et al., 1999; Törn, Landin-Olsson, Lernmark et al., 2000; Törn, Landin-Olsson, Lernmark et al., 2001; Borg et al., 2002; Scholin et al., 2004). Additional analysis of the autoantibody-positive patients identified a number of risk factors associated with an increased likelihood of losing β cell function.
The aim of this study is to identify risk factors for the loss of measurable plasma C-peptide in newly diagnosed 15- to 35-year-old diabetic subjects.
This Swedish study included 778 subjects. C-peptide levels were obtained each year for 6 years after diagnosis. Loss of measurable C-peptide was defined as a level at or below the lower detection limit of the local assay (0.13 nmol/l). In addition to C-peptide, other baseline covariates included gender, age, body mass index, HLA genotype, and autoantibody levels.
Compared with autoantibody-negative subjects, autoantibody-positive subjects had lower median baseline C-peptide (0.27 vs. 0.50, P<.001), their levels declined over the study period, and the risk of losing measurable C-peptide was significantly higher when more than one autoantibody was present [odds ratio (OR), 4.0; 95% confidence interval (CI), 2.13–7.54]. Among autoantibody-positive individuals, the presence of GAD65Ab (OR, 1.8; 95% CI, 1.24–2.51) and islet cell antibodies (OR, 1.6; 95% CI, 1.19–2.18) conferred a higher risk for loss of measurable C-peptide as did female gender (OR, 1.6; 95% CI, 1.17–2.11) and time after diagnosis (OR, 1.5 for each additional year postdiagnosis; 95% CI, 1.41–1.57). Higher baseline C-peptide levels were protective (OR, 0.5 for each additional log_e nanomoles per liter; 95% CI, 0.36–0.58).
This study identified autoantibody status, gender, and baseline C-peptide levels as factors that will be useful for predicting the disease course of 15- to 35-year-old diabetic individuals.
Prevention of Autoimmune Disease: Type 1 Diabetes as a Paradigm
2006, The Autoimmune Diseases, Fourth Edition
It is noted that in the case of type 1 diabetes (T1D) the disease process begins months to years before major loss of beta-cell function heralds symptoms of hyperglycemia. Increasing knowledge of the mechanisms of beta-cell destruction and the ability to identify individuals at risk for T1D, together with proof-of-principle for therapeutic intervention in the non-obese diabetic (NOD) mouse model, are platforms for T1D prevention in humans. Indeed, T1D can be seen as a paradigm for the preclinical diagnosis and prevention of autoimmune disease in general. Preventing T1D applies not only to people at risk but also to those with clinical diabetes, in order to preserve residual beta-cell function, allow possible beta-cell regeneration and prevent recurrent autoimmune disease after therapeutic betacell replacement or regeneration. Furthermore, T1D can be eradicated by primary prevention aimed at avoiding or averting the environmental factors that are thought to precipitate disease in genetically at-risk individuals. Prevention is, however, mostly applicable to subclinical disease, rather than to clinical disease when beta-cell destruction is end-stage.

View all citing articles on Scopus

^☆: Part of this work was presented in abstract form at the 28th Meeting of the Scandinavian Society for the Study of Diabetes (SSSD), Stockholm, 1993.

View full text

Glutamate decarboxylase antibody levels predict rate of β-cell decline in adult-onset diabetes☆

Abstract

Lancet

Lancet

Lancet

Lancet

Lancet

Diabetes Res. Clin. Pract.

Pathologic anatomy of the pancreas in juvenile diabetes mellitus

Diabetes

Pancreatic islet cell antibodies in diabetes mellitus correlated with the duration and type of diabetes, coexistent autoimmune disease and HLA-type

Diabetes

Insulin antibodies in insulin-dependent diabetics before insulin treatment

Science

Autoantibodies in newly diagnosed diabetic children immunoprecipitate human pancreatic islet cell proteins

Nature

Identification of the 64K autoantigen in IDDM as the GABA-synthesizing enzyme glutamic acid decarboxylase

Nature

Recombinant GAD representing the single isoform expressed in human islets detects IDDM-associated 64K autoantibodies

Diabetes

Risk for developing IDDM and the presence of islet 64K antibodies

Diabetologia

Persistence of antibodies to a 64,000-Mr islet cell protein after onset of type 1 diabetes

Diabetes

Antibodies to GAD discriminate major diabetes type

Diabetes

Islet cell antibodies and fasting C-peptide predict insulin requirement at diagnosis of diabetes mellitus

Diabetologia

Islet cell and thyrogastric antibodies in 633 consecutive 15–34 year old patients in the Diabetes Incidence Study in Sweden (DISS)

Diabetes

β-Cell function in relation to ICA during the first 3 years after clinical diagnosis of diabetes in Type II diabetics

Diabetes Care

Predicting IDDM: use of humoral immune markers

Diabetes Rev.

Quantitative assay using recombinant human islet glutamic acid decarboxylase (GAD-65) shows 64K ab positivity at onset predicts diabetes type

J. Clin. Invest.