Diabetes Research Centre

Melbourne, Australia

What is Diabetes?

Clinical Trials

In the Lab

Links

Contacts

Site History

Intranasal insulin trial (INIT I) in pre-clinical type 1 diabetes

Autoimmune disease results from dysregulated immune responses to self-antigens. Type 1 diabetes is an autoimmune disease in which immune cells target and destroy the insulin-secreting beta cells in the islets of the pancreas. Individuals at-risk for type 1 diabetes can now be identified by the presence of circulating antibodies to the islet antigens insulin, glutamic acid decarboxylase (GAD) and tyrosine phosphatase-like islet antigen 2 (IA-2).

The ability to use self-antigens as tools to induce protective immunity, free from the side-effects of conventional non-specific immunosuppressive agents, is the 'Holy Grail' of autoimmune disease therapy. Administration via 'tolerogenic' mucosal routes in animal models has provided proof-of-concept for such antigen-specific therapy. For example, we have shown that regulatory T cells that prevent diabetes can be induced by administering insulin to the naso-respiratory tract of diabetes-prone mice.
On this background, we conducted a trial of intranasal insulin in 38 humans (median age 10.8 years) at risk for type 1 diabetes to determine the effect of intranasal insulin on immune and metabolic markers and to ensure that this potentially immunoprotective strategy was safe. The intranasal insulin trial (INIT) was randomized, placebo-controlled and double-masked, with a crossover at six months. A separate cohort of 13 subjects (median age 13.0 years), was followed to symptomatic diabetes over 1.0-7.7 years up to the time the trial commenced.

Subjects had blood samples taken monthly during the trial for measurement of blood glucose concentration, insulin antibodies (IAb), GADAb and IA2Ab, as well as T-cell proliferative responses to tetanus toxoid and denatured human insulin. In addition, beta-cell function was assessed by measuring first-phase insulin response (FPIR) to intravenous glucose, at randomisation, six months and 12 months. After completion of the trial, subjects were followed to diabetes, or otherwise at least yearly, with repeat testing.

Intranasal insulin was associated with increases in antibody and decreases in T-cell responses to insulin, similar to the changes observed in mice. The reciprocal changes in antibody and T-cell responses are consistent with deviation from Th1 towards Th2 immunity, a phenomenon known to accompany induction of mucosal tolerance. No local or systemic adverse effects were observed.

With follow-up (median 3.0 years), only subjects with markedly impaired beta-cell function at entry progressed to diabetes. The remaining 26 subjects exhibit stable beta-cell function, in contrast to a contemporaneous group of 13 untreated at-risk subjects in whom beta-cell function has progressively deteriorated. Their predicted and observed times to diabetes were identical. These observations are consistent with a protective effect of intranasal insulin on the natural history of autoimmune beta-cell destruction. At a minimum, the INIT demonstrates the utility of immune and metabolic markers to evaluate antigen-specific mucosal tolerance in humans and identifies a dose of intranasal insulin that is active and apparently safe in young subjects with residual beta–cell function at-risk for T1D. These findings justify a formal trial to determine if intranasal insulin is immunotherapeutic and retards progression to clinical diabetes.

Download PDF of INIT I paper (Published in Diabetes Care, 2004)

Visit INIT II Page

Return to Clinical Trials Page