Introduction
Latent autoimmune diabetes in adults (LADA) is a subtype of immune-mediated type 1 diabetes [
1-
4]. There are about 97 million patients with type 2 diabetes in China at present [
5]. According to the results of LADA China Study, glutamic acid decarboxylase autoantibody (GADA) positivity is 5.9% among phenotypic type 2 diabetes in patients more than 30 years old in China. Because of its latent and slow progressiveness, the diagnosis criteria and differential diagnosis of LADA are still very confusing [
3,
4,
6,
7,
8], leading to a high rate of misdiagnosis. LADA patients share a similar risk of complications and mortality with GADA negative, phenotypic type 2 diabetic patients [
9], so misdiagnosis of LADA will not only seriously deteriorate patients’ health condition, but also bring heavy economic burden to the public. Under this circumstance, we will review and discuss the present diagnostic criteria of LADA.
The diagnostic criteria was first proposed by Tuomi
et al. in 1992 [
10,
11], and a lot of researchers have discussed it since then [
12,
13]. Pan
et al. [
14] proposed the key points in the diagnosis of LADA for the first time in China. Zhou
et al. [
15,
16], and Chi
et al. [
17] have also proposed the diagnostic criteria of LADA at the early stage thereafter. The most accepted diagnostic criteria currently was drawn up by Immunology of Diabetes Society (IDS) in 2004, which is:(1) the presence of circulating islet autoantibodies including insulin cell antibody (ICA), GADA, protein tyrosine phosphatase antibody ( IA-2A); (2) age at onset of diabetes beyond 30 years; (3)insulin independence for at least 6 months after diagnosis [
4]. So here, we will discuss the above three aspects which are the major points of LADA diagnosis criteria.
Major points of LADA diagnosis criteria
Circulating islet autoantibodies
The diagnosis of LADA mainly relies on seropositivity of antibodies. There are five serum autoantibodies reflecting humoral immunity of LADA, and they are ICA, GADA, IAA, IA-2A, and ZnT8A, respectively [
18-
21]. ICA was the first discovered serum autoantibody in LADA patients. But due to the difficulties in the standardization of its assay, current clinical application of ICA is limited. GADA is now recognized as the most sensitive immune parameters for the diagnosis of LADA, because it appears at the early stage of LADA and has a long duration in serum. On the other hand, the assay of GADA is the most standardized of all autoantibodies. Moreover, patients with high GADA titer (LADA1) are more similar to type 1 diabetes (T1D) subjects, while with low GADA titer (LADA 2) are more prone to type 2 diabetes subjects with metabolic syndrome [
22,
23]. The determination of GADA is useful for clinical classification of diabetes. First, with GADA, we can discriminate autoimmunity-mediated LADA from phenotypic type 2 diabetic subjects. Second, the titer of GADA could help us determine the early intervention of LADA. Subjects with high GADA titers could benefit from early insulin intervention for their high risk of β cell failure. GADA titers decrease in the disease duration, but can be found positive after more than 10-20 years [
24]. IAA, IA-2A, and ZnT8A were also found to be alternative immune parameters for the diagnosis of LADA. Multiple autoantibodies in combination are able to improve the positivity of LADA [
25]. The radioligand assays for islet autoantibodies have the highest sensitivity and specificity among the currently established islet autoantibodies assays [
26].
Time to insulin dependence
A minimum period of insulin independence after diagnosis is meant to distinguish LADA from classic type 1 diabetes [
4]. But results of Action LADA2 have shown that time to insulin initiation cannot be used in defining LADA, because the time point to initiate insulin treatment is dependent on local clinicians’ judgment, but not on the disease process [
27]. Therefore, instead of “insulin independence,” some researchers described it as “without occurrence of ketosis or ketoacidosis in 6 months of onset of diabetes” [
28]. But this classification standard is also in dispute, because under the effect of certain incentives, such as infection, LADA patients may also develop ketosis or ketoacidosis. Therefore, large prospective study is still needed to resolve this problem.
Age of diagnosis
Age of diagnosis for LADA in the published literatures varied from 15 to 45 years old in different countries. The inclusion criterion of age for LADA in the United Kingdom Prospective Diabetes Study (UKPDS) was 25-65 years old [
20,
29], while the studies conducted by Weber and his colleagues hold that LADA should be diagnosed in patients at or more than 35 years old [
24,
30]. Even in China, the age criterion varied greatly, which was 20-48 years old in Pan
et al.’s study [
14] and more than 40 years old in Chi
et al.’s study [
17]. In Zhou and his colleagues’ study [
16], the cut point was 15 years old, which was the same with the inclusion criterion of age in DiaMond Study (World Health Organization’s Multinational Project for Childhood Diabetes) [
31]. In DiaMond Study, 15 years old was considered as the cut point of children and adults. Moreover, age of onset of type 2 diabetes trends to be younger today, so broadening the age limit is helpful for physicians to screen more LADA patients, so that we can make the correct diagnosis and provide better therapies at the early stage of the disease. Besides, in clinical practice, patients younger than 15 years old are often admitted to pediatric ward, while those who are more than 15 years old are often admitted to internal medicine ward for adults in China. Therefore, considering 15 years old as the cut point is more beneficial to screen for more LADA patients. But IDS hold different views, after analyzing studies on genetic, immunological and metabolic factors, IDS deemed “no more than 30 years old” as the diagnostic criterion of age [
4], and a great majority of scholars worldwide have adopted it. In China, the legal age limit for adult is 18 years old, that is, patients who are younger than 18 years old are termed as minors, while those who are at or older than 18 years are termed as adults. So in China, it seems that regarding 18 years old as the cut point will be more accordant with the definition of “adult” in LADA. In addition, 18 years old was also considered as the cut point of age for pediatric diabetes in many studies [
32,
33].
Other related aspects of LADA diagnostic criteria
β cell function
LADA is similar to type 2 diabetes at onset with insulin independency, it starts at onset of type 2 diabetes without the need of insulin and slowly develops to β cell devastation. Therefore, some scholars adopted β cell function to diagnose LADA [
17,
24,
30]. It was reported that the decreasing rate of islet β cell function in LADA, being highly heterogeneous, was three times higher than that of type 2 diabetes mellitus (T2DM) patients [
34,
35]. Currently, serum C peptide was the chief parameter to evaluate β cell function. But it is rather problematic when applied, because there are no standardized assays. Yet there has not been any published articles regarding β cell function as a diagnostic criterion of LADA.
BMI and chronic complications
Chi
et al. [
17] regarded BMI and chronic complications as referenced criteria to apparent LADA. But results from both Zhou
et al. and Action LADA3 have shown that metabolic syndrome is not a characteristic of LADA [
36,
37]. Except for a lower prevalence and incidence of nephropathy, LADA patients have a similar risk of complications and death to patients with phenotypic type 2 diabetes [
38,
39]. Thus, metabolic parameters and chronic complications cannot differentiate LADA from type 1 or type 2 diabetes.
Genetics
Pan
et al. and Zhou
et al. proposed that susceptible gene should be regarded as a key factor in the diagnosis of LADA [
14-
16]. Studies have found that LADA patients shared genetic characteristics of both type 1 and type 2 diabetics. Alleles of HLA class II genes DQB1, DQA1, and DRB1* are major determinants of genetic predisposition to T1D and in the majority of cases to LADA [
30]. A recently published study indicated LADA patients differed genetically from patients with type 1 diabetes diagnosed after 35 years old [
40]. Bakhtadze
et al. [
41] reported that common variants in the TCF7L2 gene was helpful to differentiate autoimmune from non-autoimmune diabetes in patients aged 15-34 years but not in those aged 40-59 years. However, no specific pathogenic genes that may directly cause LADA has been found yet.
T cells
Some LADA patients had no serum antibodies at the time of diagnosis, but abnormal T cell function could be detected. So these have been linked to the diagnosis of LADA, which quickly becomes a hotspot. Several different approaches have been taken to examine T cell responses in detail. A more recent development is enzyme-linked immunosorbent spot (ELISPOT) analysis, which enables the investigator to determine the qualitative and quantitative antigen-specific immune response on a single-cell level with regard to cytokine secretion [
42]. But without the availability of reference cells and powerful positive controls, there is still a long way to go for the standardization of ELISPOT assays. Some researchers recently proposed the term “T-LADA” to define the autoantibody-negative LADA, the evidence of T cell autoreactivity can be found in them. And antibody positive subjects were termed as B-LADA [
2].
Conclusions
An overview of diagnostic criteria of LADA made by different research groups in China and IDS (2004) are listed in Table 1. Above all, there are still lots of controversies over the diagnostic criteria of LADA. We believe with the effort of all colleagues and industries nationwide, discussion on the diagnostic criteria of LADA will finally reach an agreement and bring evangel to our patients.
Higher Education Press and Springer-Verlag Berlin Heidelberg
PDF
(103KB)
