Oral insulin capsules may not be an effective method for preventing type 1 diabetes, according to Diabetes TrialNet researchers.
Oral insulin treatment in relatives of those with type 1 diabetes who were positive for at least two autoantibodies did not have a significant effect on time to diabetes compared with placebo (28.5% insulin group versus 33% placebo, HR 0.87, 95% CI 0.0-1.2, P=0.21), reported Jeffrey P. Krischer, PhD, of the University of South Florida College of Medicine in Tampa, and colleagues in the Type 1 Diabetes TrialNet Oral Insulin Study Group, in the Journal of the American Medical Association.
“We now know that we can identify people long before they have clinical type 1 diabetes [T1D]. Indeed, we now consider that having two or more islet autoantibodies is an early stage of type 1 diabetes. Our aim is now to find therapies that can delay or prevent disease progression from early stages to clinically apparent disease,” explained Carla Greenbaum, MD, director of the Diabetes Program at Benaroya Research Institute in Seattle, Washington, to MedPage Today.
“Oral insulin was first tested as part of the Diabetes Prevention Trial (DPT-1), which started in 1994. In that study, a retrospective analysis suggested that those with higher levels of insulin antibodies and thus potentially a more rapid rate of progression to clinical T1D had a benefit [from] oral insulin. These findings led to the launch of the TrialNet Oral Insulin Prevention Study in 2007,” added Greenbaum, who is also chair of Diabetes TrialNet.
The trial included four groups of participants from the TrialNet Natural History Study (now called the TrialNet Pathway to Prevention study), who were all autoantibody-positive relatives of those with type 1 diabetes. The first group — the main focus of the study — included 389 individuals with higher levels of beta cell function. All participants in this group were also positive for microinsulin auto-antibody (mIAA) and either islet cell autoantibodies (ICA) or both glutamic acid decarboxylase (GAD65) antibody and islet antigen 2 (IA-2) autoantibodies.
The participants were randomly assigned to receive a daily 7.5-mg capsule of recombinant human insulin crystals or placebo pills. The groups were followed up every 6 months for a median of 2.7 years.
For the first of the secondary study groups, 55 individuals were included who had the same autoantibody expression as the first group, but had lower levels of beta cell function. This was measured by a first-phase insulin release — marked by adding 1- and 3-minute insulin values form an IV glucose tolerance test — with thresholds lower than 60 μU/mL for those between ages 3 to 7 years, or lower than 100 μU/mL in those age >8 years.
This group experienced a protective effect with oral insulin capsules, marked by a significantly longer time to diabetes in the treatment group versus placebo (HR 0.45, 95% CI 0.0-0.82, P=0.006). After follow-up, only 48.1% of the treatment group progressed to type 1 diabetes compared to 70.3% of placebo.
“It has long been thought that antigen therapy would be most effective earlier in disease. It is somewhat surprising therefore that in both the post-hoc DPT-1 analysis and in TrialNet’s new study, the benefits were seen in groups with potentially more aggressive disease,” Greenbaum noted. “An alternative explanation is that oral insulin is somehow treating an ‘immune flare’ occurring in those most responsive. If true, this may fundamentally change our understanding of mechanisms of disease progression and how we target therapy.”
In a secondary stratum group of 116 individuals who were positive for the mIAA autoantibody and either the GAD or IA-2 autoantibodies. there were no significant differences in the time to diabetes between the groups (HR 1.03, 95% CI 0.0-2.11, P=0.53). Similarly, time to diabetes was not different when they were compared to all groups (HR 0.83, 95% CI 0.0-1.07, P=0.11).
“Since type 1 diabetes is increasingly understood to be a heterogeneous disease it is possible that oral insulin may be helpful in only a subset of people. It is also possible that oral insulin is more helpful at a certain point in time, when insulin-producing cells start to lose function prior to the onset of disease. TrialNet is currently evaluating the results of mechanistic studies to help answer some of those very questions,” Greenbaum added.
There were no serious adverse events, such as severe hypoglycemic events, reported during the trial. Infection was the most commonly reported event, but did not differ between the treatment and placebo groups (134 events treatment versus 120 events placebo).
In terms of future studies, Greenbaum highlighted the “Exploring the Immune Effects of Oral Insulin” study expected early next year, which aims to assess various regimens and dosage levels of oral insulin, and subsequent affect on immune response. “Two other prevention studies also underway are testing the drugs abatacept [Orencia] and teplizumab to see if they can delay or prevent disease progression prior to clinical diagnosis,” she noted.
Greenbaum also encouraged family members of people with type 1 diabetes to be screened for islet autoantibodies through the Pathway to Prevention study.
Click here for the American Association of Clinical Endocrinologists’ clinical practice guidelines for developing a diabetes mellitus comprehensive care plan.
The trial was supported by the Diabetes TrialNet Study Group.
Greenbaum and a co-author disclosed relevant relationships with Eli Lilly.
Robert Jasmer, MD Associate Clinical Professor of Medicine, University of California, San Francisco and Dorothy Caputo, MA, BSN, RN, Nurse Planner