Onset age-dependent variations of three islet specific autoantibodies in Japanese IDDM patients

Diabetes Research and Clinical Practice 39 (1998) 211 – 217

Onset age-dependent variations of three islet specific autoantibodies in Japanese IDDM patients Hitomi Yamada a, Yasuko Uchigata a,*, Eiji Kawasaki b, Nobuo Matsuura c, Toshika Otani a, Akiko Sato a, Kazuko Mutoh a, Tadasu Kasahara a, Naoki Fukushima d, Akemi Koike e, Michiyo Mizota a, Junnosuke Miura a, Hikaru Kubo f, Akio Yamaguchi f, Shigenobu Nagataki b, Yasue Omori a, Yasuhiko Iwamoto b a

Diabetes Centre, Tokyo Women’s Medical College, 8 -1 Kawada-Cho Shinjuku-ku, Tokyo 162, Japan The First Department of Internal Medicine, Nagasaki Uni6ersity School of Medicine, Nagasaki, Japan c The Department of Pediatrics, Kitasato Uni6ersity School of Medicine, Kanagawa, Japan d The Department of Pediatrics, Sapporo City Hospital, Hokkaido, Japan e The Department of Pediatrics, Tonan Hospital, Hokkaido, Japan f The Autoimmune Project, Drug Disco6ery Research Laboratories, Pharma Research and De6elopment Di6ision, Hoechst Japan, Tokyo, Japan b

Received 18 June 1997; received in revised form 12 December 1997; accepted 26 January 1998

Abstract The age related incidence rate of insulin-dependent diabetes mellitus shows a bimodal distribution, not only in Caucasians but also in Japanese. To evaluate the onset age-related autoimmune profile at presentation in insulin-dependent diabetes mellitus (IDDM), glutamic acid decarboxylase (GAD) autoantibody, islet cell antibody (ICA), and insulin autoantibody (IAA) were measured in 137 newly diagnosed Japanese IDDM patients with onset ages between 0–29 years. The prevalence of GAD autoantibody was significantly increased from the lowest (32%) in the 0 – 5 years onset age group to 75% in the 13–19 years onset age group (PB 0.05), whereas the IAA prevalence significantly decreased from the peak (48%) in the 6–12 years onset age group to 10% in the 20 – 29 years onset age group (PB 0.05). The ICA prevalence was increased from the lowest (32%) in the 0 – 5 years onset age group to the highest (53%) in the 20–29 years onset age group similar to that for the GAD autoantibody. Such results demonstrate that there was age-related autoimmune characteristics at presentation of IDDM in Japanese as well as in Caucasians. © 1998 Elsevier Science Ireland Ltd. All rights reserved.

Abbre6iations: GAD, glutamic acid decarboxylase; IAA, insulin autoantibody. * Corresponding author. Tel.: + 81 3 33538111, ext. 27011; fax: + 81 3 33581941. 0168-8227/98/$19.00 © 1998 Elsevier Science Ireland Ltd. All rights reserved. PII S0168-8227(98)00008-4

212

H. Yamada et al. / Diabetes Research and Clinical Practice 39 (1998) 211–217

1. Introduction

2. Experimental design and methods

It has been reported that the incidence rate distribution for type 1 (insulin-dependent) diabetes mellitus (IDDM) in Caucasian populations peaks twice, once at puberty (the first peak), and again after the age of 40 (the second peak) [1]. The incidence rate of IDDM in Japanese patients also has ages corresponding to the two peaks which differ from those of Caucasian populations: one peak around the age of 4 (the first peak) with the other peak at puberty (the second peak) [2]. In the second peak of Japanese IDDM population, the incidence of non-insulin-dependent diabetes mellitus (NIDDM) in Japanese begins to increase after 10 years of age [2], whereas the incidence of NIDDM in Caucasians begins to increase after 40 years of age [3]. It may raise the age-related characteristics of IDDM between Japanese and Caucasian population. Comparisons of childhood-(the first peak) and adult-(the second peak) onset Caucasian IDDM patients have revealed differences in the prevalences of islet cell autoantibody (ICA) and insulin autoantibody (IAA) [4], and differences in the HLA-DR3/4 frequency [5]. The mode of presentation also seems to differ between the two groups; the childhood-onset IDDM patients seem to present with ketoacidosis of abrupt onset rather than ketosis with mild onset, observed more commonly in adult-onset IDDM patients [4,6]. The pubertyonset (the first peak) and adult-onset (the second peak) Caucasian IDDM patients probably correspond to the 4 year old onset (the first peak) and puberty-onset (the second peak) Japanese IDDM patients, respectively. The objectives of the present study were to evaluate the autoimmune characteristics at presentation in Japanese IDDM patients with onset age between 0–29 years, as the frequencies of HLA types as one of the genetic characteristics have already been reported in a large-scale study on 340 Japanese IDDM patients with the onset age between 0–29 years [7].

2.1. Patients We studied all patients who were newly diagnosed with IDDM in the diabetes center of Tokyo Women’s Medical College, the Department of Pediatrics of Sapporo City Hospital, and Tonan Hospital from 1984 to 1993, according to the criteria for classification of IDDM published by the National Diabetes Data Group [8]. All the patients included in this study were B 30 years of age at diagnosis, had an acute onset, were ketosisprone and non-obese, required a daily insulin dose ] 0.5 U/kg, and showed a daily urinary C-peptide value of B 6.6 nmol/day. Serum samples obtained within 1 week after the initial assessment were stored in the freezer and were subsequently used for GAD autoantibody, ICA and IAA assays. The data obtained for the study groups are shown in Table 1 and Table 2. As the age-dependent variation of HLA class I and II serological types has been shown in four groups according to onset age of IDDM [7], the 137 patients in this study were divided into the same four groups; 0–5 years (n=19); 6–12 years (n= 48); 13–19 years (n= 40); and 20–29 years (n= 30). Diagnosis of autoimmune thyroid disease was based on the clinical symptoms of thyroid dysfunction, goiter presence, serum levels of thyroid hormones and thyroid-stimulating hormone, and the presence of antithyroid autoantibodies. Table 1 Characteristics of 137 IDDM patients in this study IDDM patients

n Age at onset (years) Autoimmune thyroid disease Anti-thyroglobulin Ab+ Anti-thyroid peroxidase Ab+ GAD autoantibody+ ICA+ IAA+

P

Male

Female

51 13.2 9 6.0 0 (0%)

86 12.5 95.9 3 (3%)

\0.5 \0.5

5 (11%)

7 (8%)

\0.5

5 (11%)

14 (16%)

\0.5

25 (56%) 22 (49%) 13 (29%)

55 (64%) 39 (45%) 37 (43%)

\0.1 \0.5 \0.2

H. Yamada et al. / Diabetes Research and Clinical Practice 39 (1998) 211–217

213

Table 2 Characteristics of four groups Age at onset

0–5

6 – 12

13 – 19

20 – 29

P

n Female (%) Autoimmune thyroid disease (n) Anti-thyroglobulin Ab+ (%) Anti-thyroid peroxidase Ab+ (%) GAD autoantibody+ (%) ICA+ (%) IAA+ (%)

19 11 0 0 2 6 6 7

48 26 0 4 5 26 23 23

40 30 2 7 8 30 18 17

30 19 1 1 4 16 16 31

\0.5 \0.3 \0.2 \0.8 B0.05 \0.3 B0.05

(57)

(10) (32) (32) (37)

(54) (8) (10) (54) (48) (48)

(75) (17) (20) (75) (45) (43)

(63) (3) (13) (53) (53) (10)

Figures in circles represent percentage.

2.2. Autoantibodies to GAD

2.4. Insulin autoantibody (IAA) assay

A GAD antibody RIA was performed using a ‘RIP Anti-GAD Hoechst’ RIA kit, developed by Hoechst Japan (Tokyo, Japan). GAD was purified from fresh pig brain according to Wu et al. [9] and Gottlieb et al. [10]. The details of the assay have been described previously [11,12]. Sera were considered GAD antibody-positive if they contained \5 U/ml of the antibody, which was \ 3 S.D. above the mean of 323 healthy individuals [12]. The inter- and intra-assay coefficients of variation (CV) were 3.5 – 8.9% and 2.4 – 5.1%, respectively.

Serum levels of insulin autoantibodies (IAA) were determined by a modification of the liquidphase radiobinding assay as described by Soeldner [14]. The results were expressed in terms of the radioactivity bound in the precipitates, as a percentage of the total count in the assay with the unlabeled human insulin subtracted from the percentage of that without unlabeled human insulin. Values \ 3 S.D. above the age-matched controls of IAA \ 1.35 d percent binding, were considered positive. The inter- and intra-assay CV were 5.5 and 4.3%, respectively.

2.3. Islet cell autoantibody (ICA) assay

2.5. Assay for antithyroid autoantibodies

ICAs were detected by immunoenzymatic staining using peroxidase-labeled protein A on human pancreatic fresh-frozen sections on serial dilutions of serum samples, as described previously [13]. Titers were converted to Juvenile Diabetes Foundation Units (JDF U) by comparison with a standard curve constructed using a JDF standard reference serum. Positive results were defined as \2.5 JDF U. Our ICA assay was presented for the ICA proficiency tests organized by the International Diabetes Workshop, and the results for lab validity, consistency, sensitivity and specificity in the 5th ICA proficiency test results were 95, 95, 85 and 100%, respectively.

Anti-thyroglobulin antibody and anti-thyroid peroxidase antibody were measured by hemagglutination tests using commercial kits (Serodia, Fujirebio, Tokyo, Japan). The cut-off values were 1:100 for both antibodies.

2.6. Statistical analysis Categorical variables were analyzed for statistical significance using the x 2 test or Fisher exact test and continuous variables was analyzed using unpaired t-test. All analyses were run on the personal computer statistics package SPSS for windows version 6.0.

214

H. Yamada et al. / Diabetes Research and Clinical Practice 39 (1998) 211–217

Table 3 Prevalence of GAD autoantibody, ICA, IAA positivies of male and female IDDM patients Group

0–5

6 – 12

13 – 19

20 – 29

GAD autoantibody

Male Female

1/8 (13) 5/11 (46)

11/22 (50) 15/26 (58)

7/10 (70) 23/30 (77)

4/11 (37) 12/19 (63)

ICA

Male Female

3/8 (46) 3/11 (27)

11/22 (50) 12/26 (46)

4/10 (40) 14/30 (47)

6/11 (55) 10/19 (53)

IAA

Male Female

0/8 (0)* 7/11 (64)

10/22 (45) 13/26 (50)

3/10 (30) 14/30 (47)

0/11 (0) 3/19 (16)

Figures in circles represent percentage. * PB0.001 vs. IAA in female of 0–5 year group.

3. Results

3.3. Age-dependent 6ariation of ICA pre6alence

3.1. Characteristics of subjects

ICA was detected in 46% of the total 137 subjects, lower than the GAD autoantibody prevalence. The percentages of ICA-positive patients were 31, 48, 45 and 53% (P\0.3) in the 0–5, 6–12, 13–19 and 20–29 years groups, respectively. Age segregation for the ICA prevalence was similar to that for the GAD autoantibody prevalence. The age segregation for the ICA prevalence had no gender differences (Table 3).

As shown in Table 1, there was a dominance of female IDDM patients, and this dominance was observed in each group (Table 2). Female dominance in Japanese IDDM patients has already been reported by Tajima et al. [15] and Matsuura et al. [16] as well as Otani [2]. There were no gender differences in the GAD autoantibody, IAA, ICA and thyroid specific antibody prevalences (Table 1).

3.2. Age-dependent 6ariation of GAD autoantibody pre6alence A total of 80 out of 137 (58%) patients with newly-diagnosed IDDM were positive for GAD autoantibody. The prevalence of GAD autoantibody was analyzed in each group (Table 2). The percentages of GAD antibody-positive patients in the 0–5, 6–12, 13– 19 and 20 – 29 years groups were 32, 54, 75 and 53%, respectively (P B 0.05). The prevalence increased with increasing onset age and reached a peak (75%) in the 13 – 19 years group. Table 3 shows the segregation for GAD autoantibody prevalence in male and female IDDM patients. There was no gender difference in the age segregation for GAD autoantibody prevalence.

3.4. Age-dependent 6ariation of IAA pre6alence The results of the IAA assay were positive in 36% of the total 137 subjects, the lowest percentage of prevalence among the three autoantibodies studied. The percentages of patients testing positive for IAA in the 0–5 to the 20–29 year groups were 37, 48, 43 and 10%, respectively (PB 0.05), showing a peak in the 6–12 years group and decreasing from this group to the 20–29 years group. The onset age variation of IAA prevalence was remarkably different from that for the GAD autoantibody and ICA prevalences. The age segregation for IAA prevalence showed a more obvious inverse relationship in females than in the male subjects (Table 3). The gender difference was present in only the 0–5 years group (Table 3).

H. Yamada et al. / Diabetes Research and Clinical Practice 39 (1998) 211–217

215

Fig. 1. Relationship between GAD autoantibody, ICA, and IAA prevalence in four groups. The percentage of individuals positive for islet specific autoantibody in each group was shown.

3.5. Relationship between GAD autoantibody, ICA, and IAA Fig. 1 shows the percentages of individuals positive for each islet-specific autoantibody segregated into the four groups. Under age 5, none of the ICA- and IAA-negative patients were GAD autoantibody-positive. Between ages 6 and 12, GAD autoantibody contributed only 8% antibody-positive subjects in addition to ICA and IAA positivity. In patients with onset between ages 13 and 19, 30% were GAD autoantibodypositive and ICA- and IAA-negative. However, the percentage of GAD autoantibody prevalence decreased after the age of 20.

4. Discussion In this study, we demonstrated onset age-de-

pendent variations in the prevalences of GAD autoantibody, ICA and IAA in Japanese IDDM patients. There were significant onset age-dependent differences in prevalence of GAD autoantibody (PB 0.05) and IAA (PB 0.05). The GAD autoantibody did not complement IAA and ICA as IDDM markers in the 0–5 years group, but was a more objective marker for IDDM in the \ 13 years group. This is partly in agreement with a study by Vandewalle et al. [17], and a study by Zimmet et al. [18] which showed the prevalences of GAD autoantibody and IAA in IDDM patients positive for ICA, and with a study by Verge et al. [19]. In the reports of Vandewalle et al., onset age-dependent segregation in the prevalence of the GAD autoantibody was not apparent, although those of IAA and ICA were clearly distinguished [17]. The report of Verge et al. showed onset age-dependent segregation in the prevalence of IAA, but did not in the prevalence of GAD

216

H. Yamada et al. / Diabetes Research and Clinical Practice 39 (1998) 211–217

autoantibody and ICA [19]. The age-related difference in the prevalences of both GAD autoantibody and IAA in this study may suggest heterogeneity in the pathogenesis of IDDM which could, on the basis of onset age, be mainly divided into the younger group, the puberty group and the twenties group in Japanese, or into a childhood-onset group and an adult-onset group in Caucasians. The prevalences of IAA in Japanese IDDM patients had a different trend from that in Caucasian IDDM patients which had the highest prevalence of IAA in the youngest group and the decline with age [19], which may be not surprising from the following thoughts; there was no evidence of autoimmune insulitis in half of the Japanese IDDM patients who received pancreas biopsy [20], and Japanese IDDM patients in the 0 – 5 years group may be developed before making detectable IAA, although T-cell reactivity to insulin are present [21]. There were no significant gender differences in the onset age-dependent prevalences in GAD autoantibody, ICA, and IAA except in the IAA prevalence in the 0 – 5 years group, although a female dominance in the subjects was observed in this study. There was a tendency for a higher prevalence of the GAD autoantibody in female patients. Many reports on the Caucasian IDDM population [18,22] showed gender differences in the prevalence of islet-specific antibodies. This ethnic difference may be partly because of the low incidence of IDDM in Japanese patients studied, especially of male IDDM patients (The incidence of IDDM in Japan is  1.0 – 2.0/100000 general population [23] and male:female= 1:2 – 2:3 [15,16]). Another reason may be the small number of IDDM patients with autoimmune thyroid diseases in the present study. The prevalence of GAD autoantibody measured using ‘RIP Anti-GAD Hoechst’ RIA kit [24,25] in Japanese IDDM was definitely lower compared with that reported in Caucasians with newly-diagnosed IDDM [26,27]. If the prevalence of GAD autoantibody in this study is still considered to be low, one reason could be that half of those patients, who did not show a high prevalence of GAD antibody, (32% of the 0 – 5 years

group and 54% of the 6–12 years group), were B 12 years of age, and the small number of IDDM patients with autoimmune thyroid diseases. The prevalence of GAD autoantibody in our study agrees with that in other reports of Japanese IDDM patients [12,26,27]. As Kawasaki et al. [27] already reported, the prevalence of GAD autoantibody in this study was also higher than that of ICA. It supports that the sensitivity of GAD autoantibody detection is greater than that of ICA detection. The lower prevalence of ICA in Japanese IDDM patients compared with other reports [28] may be explained by the same reasons described according to the prevalence of GAD autoantibody. Aclearly high prevalence of IAA was indeed observed in the youngest age group and the GAD antibody dominance appeared with increasing onset age, although the percentage of patients positive for both ICA and GAD autoantibody was higher in the groups with onset age \ 20 years. These results support the inference that there was age-related autoimmune characteristics at presentation of IDDM in Japanese as well as in Caucasians. References [1] A.S Krolewski, J.H Warram, L.I Rand, C.R. Kahn, Epidemiologic approach to the etiology of Type 1 diabetes mellitus and its complications, New Engl. J. Med. 317 (1987) 1390 – 1398. [2] T. Otani, H. Yokoyama, Y. Higami, T. Kasahara, Y. Uchigata, Y. Hirata, Age of onset and type of Japanese younger diabetes in Tokyo, Diabetes Res. Clin. Pract. 10 (1990) 241 – 244. [3] M. Laakso, K. Pyorala, Age of onset and type of diabetes, Diabetes Care 8 (1985) 114 – 117. [4] J. Karajalainen, P. Salmela, J. Ilonene, H.-M. Surcel, M. Knip, A comparison of childhood and adult type 1 diabetes mellitus, New Engl. J. Med. 320 (1989) 881 – 886. [5] S. Caillat-Zucman, H.-J. Garchon, J. Timsit, et al., Agedependent HLA genetic heterogeneity of type 1 insulindependent diabetes mellitus, J. Clin. Invest. 90 (1992) 2242 – 2250. [6] T. Kobayashi, H. Kajio, K. Takemoto, et al., Immunogenetic and clinical characterization of slowly progressive IDDM, Diabetes Care 16 (1993) 780 – 788. [7] M. Mizota, Y. Uchigata, S. Moriyama, et al., Age-dependent association of HLA-A24 in Japanese IDDM patients, Diabetologia 39 (1995) 371 – 372.

H. Yamada et al. / Diabetes Research and Clinical Practice 39 (1998) 211–217 [8] National Diabetes Data Group, Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance, Diabetes 28 (1979) 1039–1057. [9] J.Y. Wu, L. Genner, C.T. Lin, G. Song, L-glutamate decarboxylase from brain, Methods Enzymol. 113 (1985) 3– 17. [10] D.I. Gottieb, Y.C. Chang, J.E. Schwob, Monoclonal antibodies to glutamic acid decarboxylase, Proc. Natl. Acad. USA 83 (1986) 8808–8812. [11] E. Kawasaki, H. Takino, M. Yano, et al., Autoantibodies to glutamic acid decarboxylase in patients with IDDM and autoimmune thyroid disease, Diabetes 43 (1994) 80– 86. [12] A. Tsuruoka, I. Matsuba, T. Toyota, G. Isshiki, S. Nagataki, Y. Ikeda, Antibodies to GAD in Japanese diabetic patients: A multicenter study, Diabetes Res. Clin. Pract. 28 (1995) 191 – 199. [13] A. Takahashi, M. Tsujihata, A. Yokota, et al., A new method of detection of islet cell antibodies (ICA) using peroxidase-labeled protein A and incidence of ICA in type 1 (insulin-dependent) diabetes, Diabetologia 29 (1986) 378 – 382. [14] P. Vardi, S.A. Dib, M. Tuttleman, et al., Competitive insulin autoantibody assay. Prospective evaluation of subjects at high risk for development of type 1 diabetes mellitus, Diabetes 36 (1987) 1286–1987. [15] N. Tajima, R.E. Laporte, I. Hibi, T. Kitagawa, H. Fujuta, A.L. Drash, A comparison of the epidemiology of youth-onset insulin-dependent diabetes mellitus between Japan and the United States, Diabetes Care 8 (1985) 17 – 23. [16] N. Matsuura, N. Fukushima, H. Fujita, et al., Epidemiologic survey of juvenile-onset insulin-dependent diabetes mellitus (IDDM) in Hokkaido, Japan, 1973–1981, Tohoku J. Exp. Med. 141 (Suppl.) (1983) 181–189. [17] The Belgian Diabetes Registry, C.L. Vandewalle, A. Falorni, S. Avanholm, A. Lernmark, D.G. Pipeleers, F. Gorus, High-diagnostic sensitivity of glutamic acid decarboxylase autoantibodies in insulin-dependent diabetes mellitus with clinical onset between 20 and 40 years, J. Clin. Endocrinol. Metab. 80 (1995) 846–851. [18] P. Zimmet, R. Elliott, I. Mackay, et al., Autoantibodies to glutamic acid decarboxylase and insulin in islet cell antibody positive presymptomatic type 1 diabetes mellitus: Frequency and segregation by age and gender, Diabetic Med. 11 (1994) 866–871. [19] C.F. Verge, N.J. Howard, M.J. Rowley, et al., Anti-glutamate decarboxylase and other antibodies at the onset of

.

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

217

childhood: A population-based study, Diabetologia 37 (1994) 1113 – 1120. N. Itoh, T. Hanafusa, A. Miyazaki, et al., Mononuclear cell infiltration and its relation to the expression of major histocompatibility complex antigens and adhesion molecules in pancreas biopsy specimens from newly diagnosed insulin-dependent diabetes mellitus patients, J. Clin. Invest. 92 (1993) 2313 – 2322. N.C. Schloot, B.O. Roep, D. Wegmann, et al., Altered immune response to insulin in newly diagnosed compared to insulin-treated diabetic patients and healthy control subjects, Diabetologia 40 (1997) 564 – 572. R.S. Schmidli, H.J. DeAizpurua, L.C. Harrison, P.G. Colman, Antibodies to glutamic acid decarboxylase in at-risk and clinical insulin-dependent diabetic subjects: Relationship to age, sex, and islet cell antibody status, and temporal profile, J. Autoimmun. 7 (1994) 55 – 66. World Health Organization DIAMOND Project group, M. Karvonen, J. Tuomilehto, I. Libman, R. LaPorte, A review of the recent epidemiological data on the worldwide incidence of type 1 (insulin-dependent) diabetes mellitus, Diabetologia 36 (1993) 392 – 883. R.S. Schmidli, P.G. Colman, E. Bonifacio, G.F. Bottazo, L.C. Harrison, High levels of concordance between assays for glutamic acid decarboxylase antibodies. The first international glutamic acid decarboxylase antibody workshop, Diabetes 43 (1994) 1005 – 1009. Participating Laboratories, R.S. Schmidli, P.G. Colman, E. Bonifacio, Disease sensitivity and specificity of 52 assays for glutamic acid decarboxylase antibodies, Diabetes 44 (1995) 636 – 640. S.W. Serjeantson, M.R.J. Kohonen-Corish, M.J. Rowley, I.R. Mackay, W. Knowles, P. Zimmet, Antibodies to glutamic acid decarboxylase are associated with HLA-DR genetypes in both Australians and Asians with type 1 (insulin-dependent) diabetes mellitus, Diabetologia 35 (1992) 996 – 1001. P.Z. Zimmet, M.J. Rowley, I.R. Mackay, et al., The ethnic distribution of antibodies to glutamic acid decarboxylase: Presence and levels in insulin-dependent diabetes mellitus in European and Asian subjects, J. Diabetes Complications 7 (1993) 1 – 7. E. Kawasaki, H. Takino, M. Yano, et al., Evaluation of islet-specific autoantibodies in Japanese patients with insulin-dependent diabetes mellitus: A comparison between autoantibodies to glutamic acid decarboxylase, autoantibodies to 64 kDa islet cell protein and islet cell antibodies, J. Autoimmun. 7 (1994) 791 – 802.