Update on slowly progressive type 1 diabetes

Type 1 diabetes is a metabolic disease in which the destruction of pancreatic β cells leads to absolute insulin deficiency and chronic hyperglycemia, mainly due to islet cell autoimmunity¹. Different islet cell-associated autoantibodies are important in this diagnosis, including anti-glutamic acid decarboxylase antibody (GADA), anti-insulinoma-associated antigen-2 antibody (IA-2A)², anti-insulin autoantibody (IAA), anti-zinc transporter 8 antibody (ZnT8A), and islet cell antibody (ICA). Type 1 diabetes can be classified into three types: acute-onset, slowly progressive, and fulminant type 1 diabetes^{3, 4}. In recent years, many cases of type 1 diabetes associated with immune checkpoint inhibitors have been reported⁵. Historically, the presence of GADA or ICA in people with diabetes in a non-insulin-dependent state has led to a diagnosis of slowly progressive type 1 diabetes (slowly progressive insulin-dependent diabetes mellitus; SPIDDM). However, the diagnostic criteria for SPIDDM were revised in Japan in 2023, with more strict guidelines⁶. In addition, a statement regarding therapeutic interventions for suspected SPIDDM cases has been issued⁷. This article focuses on these topics in addition to the latest information on SPIDDM pathogenesis.

Persons with SPIDDM are non-insulin-dependent in the early stages of the disease and present a clinical picture similar to that of type 2 diabetes. However, as the disease progresses, their ability to secrete insulin gradually decreases and they finally become insulin-dependent⁶. Since the diagnostic criteria require the presence of islet-related autoantibodies such as GADA, it is inferred that autoimmune responses targeting β cells are involved in the pathological condition. We previously reported that, in approximately 22% of participants with SPIDDM, including suspected cases, mononuclear cells producing interferon-γ (T helper 1 [Th1] response) reactive to the insulin peptide consisting of amino acids 9–23 (B9-23) of the insulin B chain and its adjacent peptides (hereinafter referred to as insulin B9-23-related peptides) were detected in their peripheral blood⁸. In addition, there was a significant negative correlation between the peripheral frequency of these mononuclear cells and the capability to secrete endogenous insulin⁸. These findings suggest that insulin B9-23-related peptides may play an important role as autoantigens in SPIDDM development, with the insulin peptide-specific Th1 responses potentially reflecting disease activity.

Previous studies on SPIDDM demonstrated that people with diabetes in a non-insulin-dependent state with a GADA titer of ≥10 U/mL measured through a radioimmunoassay (RIA) (GADA-RIA) are at a higher risk for future progression to an insulin-dependent state than those with a GADA-RIA titer of <10 U/mL^{9, 10}. However, in December 2015, GADA measurements shifted from RIA to enzyme-linked immunosorbent assays (ELISA) in Japan¹¹. Since a GADA titer of 180 U/mL measured through ELISA (GADA-ELISA) is reportedly equivalent to a GADA-RIA titer of 10 U/mL, we recently conducted a new analysis focusing on the relationship between insulin B9-23-related peptide-specific Th1 responses and GADA-ELISA titers in SPIDDM¹². Subsequently, we found that participants with SPIDDM and a GADA-ELISA titer of ≥180 U/mL showed higher B9-23-related peptide-specific Th1 reactions than those with a GADA-ELISA titer of <180 U/mL, especially those with a diabetes duration of <7 years, suggesting an increased risk of future progression to an insulin-dependent state. Kawasaki et al.^{13, 14} recently showed in a nationwide prospective cohort study with a limited observation period that, unlike GADA-RIA, a positive GADA-ELISA result may indicate a high risk of developing insulin dependence in the future in SPIDDM, regardless of the GADA-ELISA titer. Long-term prospective cohort studies are required to confirm the validity of these findings.

According to histological examinations, in addition to insulitis, precancerous lesions called acinar-to-ductal metaplasia (ADM) and pancreatic intraepithelial neoplasia (PanIN) lesions are frequently observed in the exocrine pancreas of persons with SPIDDM¹⁵. Interestingly, enteroviruses were localized in these endocrine and exocrine lesions, suggesting that chronic and persistent enteroviral infections in the pancreas may be involved in these histological changes. Furthermore, the viral genome-derived protease “2A^Pro” is expressed in enterovirus-infected cells, and this protease may induce chronic islet inflammation, presumably triggering the development of islet-associated antigen-specific autoimmune responses and β cell injury (Bystander mechanism)¹⁵.

The diagnostic criteria for SPIDDM were revised in 2023 (Table 1)⁶. The key points are described below.

In 2023, the Japan Diabetes Society issued a statement on therapeutic interventions for persons with SPIDDM (probable)⁷. In addition to insulin therapy, the use of dipeptidyl-peptidase-4 inhibitors and biguanide was recommended in the early stages⁷. If hyperglycemia is difficult to control, the use of glucagon-like peptide-1 receptor agonists or other oral hypoglycemic drugs can be considered. However, sulfonylureas should be avoided as they increase the risk of insulin dependency in SPIDDM¹⁸. In contrast, persons with SPIDDM (definite) should receive intensive insulin therapy. Some sodium–glucose cotransporter 2 inhibitors covered by insurance for type 1 diabetes can be used in combination with insulin therapy in Japan, with careful attention paid to the onset of diabetic ketoacidosis¹⁹.

Akira Shimada has received lecture fees from Sanofi K.K. The other authors declare that there is no duality of interest associated with this manuscript.

Approval of the research protocol: N/A.

Informed consent: N/A.

Registry and the registration no. of the study/trial: N/A.

Animal studies: N/A.