The Clinical Heterogeneity of Adult-Onset Diabetic Patients With GAD Autoantibodies in Japan: Response to Fukui et al.

We would like to thank Fukui et al. (1) for the interest in our article (2) and the editor for the opportunity to clarify the several points raised. As stated several times in our article, this is a cross-sectional and hospital-based study to investigate the clinical, autoimmune, and genetic characteristics in adult-onset diabetic patients with GAD antibodies (GADab). All diabetic patients who visited our hospital during 1998 and 1999, including those with typical type 1 diabetes (a slowly progressive form of type 1 diabetes [SPIDDM]) and type 2 diabetes, regardless of their therapy, were recruited into this study. In our Ehime Study, the overall prevalence of GADab in patients with adult-onset diabetes was 3.8%. Among them, the prevalence of GADab in patients with type 1 diabetes was 61% (72 of 118). The prevalence of GADab in patients with non-insulin-dependent diabetes was 2.0%, which is similar to that reported by Abiru et al. (2.4% in patients with non-insulin-deficient diabetes) (3). The lower prevalence of GADab in our study compared with the study by Fukui et al. (4) is ascribable to several points.

First, the proportion of the patients treated with insulin in the two studies is quite different. The number of patients under insulin treatment in Japan has been estimated to be ∼20% (5). However, the proportion of the insulin-treated patients in their study is extremely high (>50%), which probably resulted in the higher prevalence of GADab in their subjects. In fact, the prevalence of GADab in 392 type 2 diabetic patients was 3.1% in their study, which is not significantly different from our data. Second, the GADab assay used in their study is different from our radioassay, which can measure GAD65-specific autoantibodies. They used radioimmunoassay with purified rat brain GAD as antigen that contains both GAD65 and GAD67. It is well known that there are a certain number of patients with type 1 diabetes who are positive for only GAD67ab. Although the predictive power of GAD67ab on future insulin deficiency in GADab-positive non-insulin-dependent diabetes has not been established yet, Fukui et al. might have overestimated the prevalence of GADab in their patients.

In this Ehime Study, we are prospectively following the GADab⁺ patients with non-insulin-dependent diabetes to clarify the factors that distinguish the nonprogressors from those who develop the insulin-dependent state. Therefore, with the successful completion of the study, the clinical characteristics, humoral autoimmunity to other organ-specific autoantibodies, and HLA class II genes in patients with slowly progressive form of type 1 diabetes will be clarified.

Fukui et al. misread in part our results on the frequency of HLA-DRB1 haplotypes shown in Table 3 of our article. They pointed out that the prevalence of HLA-DRB1*0405 in GADab⁺ patients with non-insulin deficiency was significantly lower than GADab⁺ patients with insulin deficiency. However, as shown in Table 3 of our article, the frequency of DRB1*0405 was increased in both GADab⁺ patients with insulin deficiency (28.5%) and non-insulin deficiency (21.0%) compared with healthy control subjects (11.3%), and no difference was observed between two groups. On the other hand, DRB1*0901 was decreased in frequency in GADab⁺ patients with non-insulin deficiency compared with GADab⁺ patients with insulin deficiency. Then only the DRB1*0405 was frequent in GADab⁺ patients with non-insulin deficiency, which may be one of the characteristics in Japanese non-insulin-deficient patients with anti-islet autoantibodies. We agree that one of the features that distinguishes GADab⁺ non-insulin-deficient patients from insulin-deficient patients is the frequencies of type 1 diabetes-protective HLA haplotypes DRB1*1501 and *1502. We kindly ask Fukui et al. to read our article again to resolve these misinterpretations.

Recently, the results of a randomized controlled clinical trial to prevent the onset of type 1 diabetes in islet cell antibody-positive relatives of patients with type 1 diabetes who were using prophylactic parenteral insulin has been reported; there was no effect on the development of diabetes (6). Although a pilot study in islet cell antibody-positive patients with type 2 diabetes has suggested that the small doses of subcutaneous insulin improve the serum C-peptide response (7), the large randomized controlled trials are necessary to prove whether prophylactic insulin administration can alter the course of development of insulin deficiency.