Latent autoimmune diabetes in adults (LADA) is classified as type 1 diabetes because of autoantibody positivity. In contrast, individuals classified as having type 2 diabetes are by definition autoantibody negative. However, autoantibodies may fluctuate (1). We therefore examined whether some individuals have transient evidence of autoimmunity preceding a diagnosis of type 2 diabetes and whether their clinical characteristics differ from those of other individuals classified as having type 2 diabetes.

The study was performed among participants of the second (HUNT2, 1995–1997) and third (HUNT3, 2006–2008) surveys of the Nord-Trøndelag Health Study (HUNT) in Norway. Details about HUNT are available elsewhere (2,3). Participants with self-reported diabetes in HUNT3 (n = 2,264) were given a detailed diabetes-related questionnaire and invited to a follow-up clinical investigation that included measurements of hemoglobin A1c (HbA1c), fasting glucose, C-peptide, and the autoantibodies GADA and islet antigen-2 (IA-2A).

We investigated participants with incident type 2 diabetes between HUNT2 and HUNT3 (n = 965), defined as diabetes without GADA and IA-2A in HUNT3 and no insulin treatment within 1 year after diagnosis. Among them, 171 participants were excluded due to lack of serum samples or possible unknown diabetes at HUNT2, as indicated by nonfasting serum glucose >11.0 mmol/L in HUNT2. Serum samples that were collected during the HUNT2 baseline examination were used for measurements of four autoantibodies: GADA, IA-2A, zinc transporter 8 (ZnT8A), and insulin (IAA). Participants who tested positive for one or more autoantibodies at HUNT2 were also tested in serum samples collected in HUNT3 for ZnT8A and IAA positivity.

All four autoantibodies were assayed at the Hormone Laboratory, Department of Medical Biochemistry, Oslo University Hospital, Oslo, Norway, using in-house and mainly manually immunoprecipitation methods. Autoantibody positivity was defined as GADA >0.05 autoantibody index (ai), IA-2A >0.07 ai, ZnT8A >0.05 ai, and IAA >0.04 ai. Based on the laboratory’s participation in the Islet Autoantibody Standardization Program in 2016, cutoff for autoantibody positivity was set for a specificity for autoimmune diabetes of >97.5%, which entailed a sensitivity of 62%, 70%, 50%, and 16%, respectively.

Among 794 participants with incident type 2 diabetes, 26 participants (3.3%) were positive for autoantibodies before diagnosis (at HUNT2). All were single positive; 2 were positive for GADA, 2 for IA-2A, 7 for ZnT8A, and 15 for IAA. Only one prediagnostic autoantibody–positive participant was positive (for ZnT8A) both before and after diagnosis of diabetes; excluding this participant did not change the results. Characteristics of our study population are given in Table 1. Prediagnostic autoantibody–positive participants were markedly younger at the time of diagnosis than autoantibody-negative participants (median 53 vs. 61 years of age, P < 0.001 for difference). The prediagnostic autoantibody–positive participants also displayed directionwise characteristics compatible with reduced β-cell function (fasting C-peptide median 861 vs. 1,006 pmol/L, HOMA2-%B median 60.8 vs. 78.1), worse metabolic control (HbA1c median 7.1% vs. 6.7%), and more symptoms at diagnosis. However, differences were not statistically significant for these differences and borderline only for symptoms (P = 0.054).

Table 1

Clinical characteristics of prediagnostic autoantibody (Ab)-negative versus Ab-positive participants with incident type 2 diabetes at HUNT3

Ab-negative HUNT2Ab-positive HUNT2P value
NN
Age at participation (years) 768 65 (58–73) 26 58 (53–66) 0.001 
Sex (male) 768 380 (49.5) 26 16 (61.5) 0.226 
Waist/hip ratio 762 0.97 (0.92–1.02) 26 0.98 (0.94–1.03) 0.266 
Waist circumference (cm) 762 103 (95–111) 26 104 (97–109) 0.849 
BMI (kg/cm2763 29.9 (27.1–33.3) 26 30 (27.3–31.9) 0.775 
Diastolic blood pressure (mmHg) 767 75 (68–82) 26 75 (65–80) 0.700 
Systolic blood pressure (mmHg) 767 136 (126–150) 26 131 (123–147) 0.342 
Nonfasting glucose (mmol/L) 746 7.5 (6.0–9.8) 25 8.3 (6.5–11.8) 0.295 
Fasting glucose (mmol/L) 368 7.3 (6.5–8.7) 7.7 (6.6–9.7) 0.684 
Total cholesterol (mmol/L) 746 5 (4.2–5.7) 25 4.9 (4.4–5.7) 0.814 
HDL cholesterol (mmol/L) 746 1.1 (1.0–1.3) 25 1.1 (0.9–1.4) 0.299 
Triglycerides (mmol/L) 763 2.0 (1.4–2.7) 26 2.1 (1.5–2.9) 0.407 
Fasting C-peptide (pmol/L) 500 1,006 (710–1,254) 12 861 (614–1,112) 0.254 
HbA1c (mmol/mol) 517 50 (43–61) 21 54 (44–65) 0.250 
HbA1c (%) 517 6.7 (6.1–7.7) 21 7.1 (6.2–8.1) 0.255 
Diabetes duration (years) 768 3.8 (1.9–6.6) 26 4.2 (3.1–7.3) 0.195 
Age at onset (years) 768 61 (53–69) 26 53 (47–58) <0.001 
HOMA2-%B (β-cell function) 350 78.1 (54.6–105.1) 60.8 (45.6–88.0) 0.288 
HOMA2-%S (insulin sensitivity) 350 40.6 (31.4–57.6) 42.2 (32.2–70.2) 0.619 
HOMA2-IR (insulin resistance) 350 2.5 (1.7–3.2) 2.4 (1.5–3.2) 0.631 
Current daily smoker 743 99 (13.3) 26 6 (23.1) 0.155 
First degree family history of diabetes 767 467 (60.9) 26 17 (65.4) 0.521 
Second degree family history of diabetes 763 427 (56.0) 26 11 (42.3) 0.168 
Symptoms when diagnosed 701 236 (33.7) 25 13 (52.0) 0.058 
GADA (ai) 768 <0.01 (0.04) 26 <0.01 (0.09) <0.001 
IA-2A (ai) 768 <0.04 (0.07) 26 <0.04 (0.11) <0.001 
ZnT8A (ai) 768 <0.01 (0.05) 26 <0.01 (0.32) <0.001 
IAA (ai) 768 <0.05 (<0.05) 26 0.05 (0.08) <0.001 
Ab-negative HUNT2Ab-positive HUNT2P value
NN
Age at participation (years) 768 65 (58–73) 26 58 (53–66) 0.001 
Sex (male) 768 380 (49.5) 26 16 (61.5) 0.226 
Waist/hip ratio 762 0.97 (0.92–1.02) 26 0.98 (0.94–1.03) 0.266 
Waist circumference (cm) 762 103 (95–111) 26 104 (97–109) 0.849 
BMI (kg/cm2763 29.9 (27.1–33.3) 26 30 (27.3–31.9) 0.775 
Diastolic blood pressure (mmHg) 767 75 (68–82) 26 75 (65–80) 0.700 
Systolic blood pressure (mmHg) 767 136 (126–150) 26 131 (123–147) 0.342 
Nonfasting glucose (mmol/L) 746 7.5 (6.0–9.8) 25 8.3 (6.5–11.8) 0.295 
Fasting glucose (mmol/L) 368 7.3 (6.5–8.7) 7.7 (6.6–9.7) 0.684 
Total cholesterol (mmol/L) 746 5 (4.2–5.7) 25 4.9 (4.4–5.7) 0.814 
HDL cholesterol (mmol/L) 746 1.1 (1.0–1.3) 25 1.1 (0.9–1.4) 0.299 
Triglycerides (mmol/L) 763 2.0 (1.4–2.7) 26 2.1 (1.5–2.9) 0.407 
Fasting C-peptide (pmol/L) 500 1,006 (710–1,254) 12 861 (614–1,112) 0.254 
HbA1c (mmol/mol) 517 50 (43–61) 21 54 (44–65) 0.250 
HbA1c (%) 517 6.7 (6.1–7.7) 21 7.1 (6.2–8.1) 0.255 
Diabetes duration (years) 768 3.8 (1.9–6.6) 26 4.2 (3.1–7.3) 0.195 
Age at onset (years) 768 61 (53–69) 26 53 (47–58) <0.001 
HOMA2-%B (β-cell function) 350 78.1 (54.6–105.1) 60.8 (45.6–88.0) 0.288 
HOMA2-%S (insulin sensitivity) 350 40.6 (31.4–57.6) 42.2 (32.2–70.2) 0.619 
HOMA2-IR (insulin resistance) 350 2.5 (1.7–3.2) 2.4 (1.5–3.2) 0.631 
Current daily smoker 743 99 (13.3) 26 6 (23.1) 0.155 
First degree family history of diabetes 767 467 (60.9) 26 17 (65.4) 0.521 
Second degree family history of diabetes 763 427 (56.0) 26 11 (42.3) 0.168 
Symptoms when diagnosed 701 236 (33.7) 25 13 (52.0) 0.058 
GADA (ai) 768 <0.01 (0.04) 26 <0.01 (0.09) <0.001 
IA-2A (ai) 768 <0.04 (0.07) 26 <0.04 (0.11) <0.001 
ZnT8A (ai) 768 <0.01 (0.05) 26 <0.01 (0.32) <0.001 
IAA (ai) 768 <0.05 (<0.05) 26 0.05 (0.08) <0.001 

Data are median (interquartile range), median (range) for autoantibodies, or n (%) unless otherwise indicated. N = number of participants included in the analysis. P values were calculated by Mann-Whitney U test for continuous variables and by χ2 test or Fisher exact test for categorical variables.

We report evidence of transient autoimmune activity prior to diagnosis of type 2 diabetes as classified by current criteria. Furthermore, transient autoantibody positivity associates with earlier onset of type 2 diabetes. To our knowledge, evidence of transient autoimmune activity prior to type 2 diabetes diagnosis has not previously been demonstrated. Notably, the nomenclature used in some previous publications could indicate otherwise; however, in those studies the term type 2 diabetes also included LADA (4,5). In addition, a previous study highlights the impact of fluctuating autoimmunity since some adolescents and young adults can be autoantibody negative at diabetes diagnosis but positive later with a resulting negative effect on C-peptide (6).

Our results indicate that ∼3% of individuals classified as type 2 diabetes may have signs of transient autoimmunity, a prevalence that is similar to maturity-onset diabetes of the young (MODY). The prevalence is sufficiently large to substantially contribute to the recognized phenotypic heterogeneity in type 2 diabetes.

One might ask whether the patients characterized here with “hidden autoimmunity” fit into any of the five clusters of adult-onset diabetes recently described (7). However, no obvious resemblance with any of the clusters seems apparent. In particular, a GADA-negative and β-cell–deficient cluster was, in contrast to our patients, characterized by comparatively low BMI.

A limitation of our study is that we only measured the autoantibodies in a single sample collection and we only have a 10-year interval for sample collection. Hence, we could not detect at which time point the participant turned autoantibody positive or when they later became negative. In addition, the relatively low response rate regarding fasting blood sampling among the participants with diabetes left us with a low number of samples to analyze differences in C-peptide.

In conclusion, we demonstrate signs of transient, prediagnostic evidence of autoimmune activity in patients later diagnosed and classified as having type 2 diabetes. Compared with other participants with type 2 diabetes, this subgroup had earlier onset of diabetes and may have lower β-cell function with possible clinical implications.

Acknowledgments. The HUNT study is a collaboration between HUNT Research Centre (Faculty of Medicine and Health Sciences, Norwegian University of Science and Technology), Nord-Trøndelag County Council, Central Norway Regional Health Authority, and the Norwegian Institute of Public Health.

Funding. The Liaison Committee for Education, Research and Innovation in Central Norway and The Norwegian Diabetes Association funded this study.

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Author Contributions. E.P.S. and V.G. designed the study. E.P.S. did the statistical analysis with support from V.G. and drafted the manuscript. B.O.Å., P.M.T., and V.G. contributed to the interpretation and discussion of the results and reviewed and edited the manuscript. All authors gave their final approval of the version to be published. E.P.S. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Prior Presentation. This study was presented in abstract form at the 54th Annual Meeting of the European Association for the Study of Diabetes, Berlin, Germany, 1–5 October 2018.

1.
Sørgjerd
EP
,
Skorpen
F
,
Kvaløy
K
,
Midthjell
K
,
Grill
V
.
Time dynamics of autoantibodies are coupled to phenotypes and add to the heterogeneity of autoimmune diabetes in adults: the HUNT study, Norway
.
Diabetologia
2012
;
55
:
1310
1318
2.
Holmen
J
,
Midthjell
K
,
Krüger
Ø
, et al
.
The Nord-Trøndelag Health Study 1995–97 (HUNT2): objectives, contents, methods and participation
.
Nor Epidemiol
2003
;
13
:
19
32
3.
Krokstad
S
,
Langhammer
A
,
Hveem
K
, et al
.
Cohort profile: the HUNT study, Norway
.
Int J Epidemiol
2013
;
42
:
968
977
4.
Hampe
CS
,
Hall
TR
,
Agren
A
,
Rolandsson
O
.
Longitudinal changes in epitope recognition of autoantibodies against glutamate decarboxylase 65 (GAD65Ab) in prediabetic adults developing diabetes
.
Clin Exp Immunol
2007
;
148
:
72
78
5.
Lundgren
VM
,
Isomaa
B
,
Lyssenko
V
, et al.;
Botnia Study Group
.
GAD antibody positivity predicts type 2 diabetes in an adult population
.
Diabetes
2010
;
59
:
416
422
6.
Landin-Olsson
M
,
Arnqvist
HJ
,
Blohmé
G
, et al
.
Appearance of islet cell autoantibodies after clinical diagnosis of diabetes mellitus
.
Autoimmunity
1999
;
29
:
57
63
7.
Ahlqvist
E
,
Storm
P
,
Käräjämäki
A
, et al
.
Novel subgroups of adult-onset diabetes and their association with outcomes: a data-driven cluster analysis of six variables
.
Lancet Diabetes Endocrinol
2018
;
6
:
361
369
Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.