Hopes that insulin immune tolerance therapy could help prevent or delay the onset of type 1 diabetes have not been met in trials of oral and intranasal insulin desensitisation.
With insulin being a key autoantigen in type 1 diabetes, early research in animal models has suggested that oral insulin therapies might prevent the autoimmune disease.
However a prospective long-term study of oral insulin in humans at high risk of diabetes due to antibodies and genotype found no overall effect on progression to type 1 diabetes.
Results from the multinational Trialnet TN07 study, which included Australian participants via the Walter and Eliza Hall Institute in Melbourne, were presented by the Institute’s Dr John Wentworth at ADC 2018 in Adelaide.
In the trial, 389 children at high risk of diabetes (at least one relative with the condition) with micro insulin autoantibodies (mIAA) were randomised to treatment with oral insulin 7.5 mg daily (203 participants) or placebo (186 participants).
Over eight years of follow-up, the time to T1D was no different between insulin and placebo groups, (8.8% annual rate for oral insulin vs 10.2% for placebo).
However oral insulin did appear to have some protective or delaying effect in a secondary stratum group of 55 participants with decreased first-phase insulin release (FPIR). In this group, oral insulin appeared to delay diabetes by an average of 31 months, with annualised rates of diabetes of 18.1% for oral insulin and 34.1% for placebo.
Oral insulin was also shown to be safe, with no significant adverse events reported and no difference in minor events compared to placebo.
Negative findings were also seen in a small trial of intranasal insulin, (INIT II) whose results were presented at ADC 2018 by Professor Len Harrison of Melbourne University.
While nasal insulin suppressed insulin antibody responses, it had no effect on rates of development of T1D in a trial of 108 children and young adults with islet autoantibodies that put them at high risk of T1D.
While intranasal insulin induced an immune response consistent with tolerance to insulin, the study found no effect of intranasal insulin in doses of 40IU or 440 IU compared to placebo on rates of development of diabetes.
Diabetes rates by treatment arm were 30% for placebo (13/43), 61% for low dose insulin (11/18) and 36% for high dose insulin (17/47).
Despite the unusually low rate of diabetes in the placebo group, “autoantigen-specific tolerance is unlikely to be effective after the onset of islet autoimmunity,” Professor Harrison concluded.
He added that a trial was now underway in Germany to assess whether high dose intranasal insulin delays autoantibody seroconversion in high risk children under two years of age.