Effects of Oral Insulin on Prevention of Diabetes in Relatives of Patients with Type 1 Diabetes

Julia Lvovich, Mercer University College of Pharmacy

Type I diabetes, also known as juvenile diabetes, is a chronic disease where the insulin producing cells of the pancreas are destroyed. Currently, there is no cure for this autoimmune disease, but studies have been using immunologic markers to predict the disease onset and predisposition. The diabetes prevention trial-type 1(DPT-1) explored the possibility of preventing type I diabetes in first- and second-degree relatives by preforming two trials; one trial in which patients received parenteral insulin therapy, and in the other trial, patients received oral insulin therapy. [1,2] Both trials failed to show a significant reduction or delay in the development of type I diabetes; however, new developments have since been made in the field of autoantibody tests prompting TrialNet study. [1,2]

TrialNet study used microinsulin autoantibody assay, which requires much less blood than radioimmunoassay that was used in the DPT-1 trials. TrialNet chose a similar group of subjects to those who have shown a trend toward benefit in the oral subgroup in from the DPT-1 trial. [3] However, the difference is that it first identified subjects by positive microinsulin indicators rather than positive islet cell antibodies, which was the baseline eligibility in the DPT-1 trials. By doing so, TrialNet included a subgroup of patients that may have been excluded in the DPT-1 trials. [1-3]

Effect of oral insulin on prevention of diabetes in relatives of patients with type one diabetes
Design Randomized clinical trial; N=560
Objective To determine whether oral insulin delays onset of type one diabetes in autoantibody-positive relatives of patients with type one diabetes
Study Groups Primary Stratum- main study group

·       Received 7.5mg/d oral insulin (n= 203)

·       Received placebo (n= 186)

Secondary Stratum One

·       Received 7.5mg/d oral insulin (n= 28)

·       Received placebo (n= 27)

Secondary Stratum Two

·       Received 7.5mg/d oral insulin (n= 51)

·       Received placebo (n= 63)

Secondary Stratum Three

·       Received 7.5mg/d oral insulin (n= 1)

·       Received placebo (n= 1)

Methods Eligibility:

·       Non-diabetic first-degree relatives of patients with type one diabetes, aged three to 45 and second-or third-degree relatives aged three to 20

·       Normal glucose tolerance on an oral glucose test

·       Confirmed to test positive for micro insulin autoantibody on two sample collections

·       Absence of diabetes-protective human leukocyte antigen haplotypes DQA1*0102 and DQB1*0602,

Duration February 2007 to December 2015, follow up through December 31, 2016
Primary Outcome Measure Time to diabetes in primary analysis stratum
Baseline Characteristics
Oral Insulin (n= 283) Placebo (n= 277)
Age, median (IQR), y 8.2 (5.9-12.5) 8.2 (5.4–11.5)
BMI, median (IQR) 17.1 (15.3- 19.5) 16.9 (15.5-19.2)
Family members with type one diabetes, No. (%)
Sibling 153 (54.1) 162 (58.5)
Identical twin 6 (2.1) 3 (1.1)
Offspring 3 (1.1) 7 (2.5)
Parent 71 (25.1) 57 (20.6)
Offspring and another first-degree relative 2 (0.7) 0
Second-degree relative 33 (11.7) 30 (10.8)
Third-degree or further removed relative 5 (1.8) 5 (1.8)
Autoantibodies positive, No (%)
Glutamic acid decarboxylase 235 (83.0) 236 (85.2)
Insulinoma-associated antigen-2 157 (55.5) 146 (52.7)
Micro insulin autoantibodies 253 (89.4) 241 (87.0)
Islet cell autoantibodies 198 (70.0) 179 (64.3)
Hemoglobin A1C, median (IQR), %d 5.0 (4.8-5.2) 5.1 (4.8-5.2)
BMI, body mass index, calculated as weight in kilograms divided by height in meter squared; IQR interquartile range
Results ·       In the main group diabetes was diagnosed:

o   58 participants (28.5%) in the oral insulin group

o   62 participants (33%) in the placebo group

o   No statistically significant difference was found in time to diabetes (hazard ratio [HR], 0.87; 95% CI, 0-1.2; p = .21)

·       In secondary stratum 1 diabetes was diagnosed

o   13 participants (48.1%) in the oral insulin group

o   19 participants (70.3%) in the placebo group

o   The time to diabetes was significantly longer with oral insulin (HR, 0.45; 95% CI, 0-0.82; P = .006).

·       Not statistical significance was found in

o   Between-group comparisons time to diabetes (n= 116) (HR 1.03; 95% CI, 0-2.11; P = .53)

o   Time to diabetes in the entire cohort (n= 560)(HR 0.83; 95% CI, 0-1.07; P = .11).


Adverse Events
Treatment Group
Severity (G


Oral Insulin
No. of participants (%)

No. of participants (%)

0/1 133 (47.0) 144 (52.0)
2 114 (40.3) 104 (37.5)
3 35


29 (10.5)
4 1 (0.4) 0 (0)
Total 2

3 (100)

277 (100)
Oral Insulin Placebo
No. of events No. of participants n= 283 (%) No. of events No. of participants n= 278 (%)
Infection 134 67 (23.7) 4 4 (1.4)
Musculaskeletal/Soft Tissue 45 38 (13.4) 20 18 (6.5)
Gastrointestinal 30 28 (9.9) 34 25 (9.0)
Pain 14 11 (3.9) 11 11 (4)
Only one category was statistically significant:     Muscular-Skeletal/Soft Tissue with p =0.005 (1-tail test).


The most common adverse event was infection, but it was not a study-related significant adverse event.

Study Author Conclusions Oral insulin, at a dose of 7.5 mg/d compared to placebo, did not delay or prevent development of type one diabetes.

In this trial, researchers attempted to determine if new microinsulin autoantibody assay, which uses insulin antibodies to establish markers for type I diabetes, would be able to show an alternative outcome to the DPT-1 oral insulin trial. In the oral insulin DPT-1 trial, patients who had positive autoantibody assays progressed to diabetes at a faster rate than those with lower, or unconfirmed insulin autoantibodies (hazard ratio 0.566, 95% CI 0.361–0.888; p= 0.015). [2] This subgroup of patients also seemed to benefit from oral insulin. [2] Thus, by using this marker as indicators for oral insulin, researchers may have hoped to find positive outcomes, which did not occur.

Although at risk population was correctly identified, this study had several limitations. First, they failed to incorporate heterogeneity that they did not adjust for genetic background, age at onset, and type of first-appearing diabetes-related autoantibody. Another limitation is that all secondary analyses are exploratory; however, this is common in most studies. Because they are all exploratory, any significant findings should be taken with caution; however, these may be hypothesis generating. Despite not meeting the primary outcomes, the study may add further knowledge to the literature to design future research as there is an unmet need in type I diabetes prevention.


[1] Diabetes Prevention Trial—Type 1 Diabetes Study Group.  Effects of insulin in relatives of patients with type 1 diabetes mellitus.  N Engl J Med. 2002;346(22):1685-1691.

[2] Skyler  JS, Krischer  JP, Wolfsdorf  J,  et al.  Effects of oral insulin in relatives of patients with type 1 diabetes: the Diabetes Prevention Trial—Type 1.  Diabetes Care. 2005;28(5):1068-1076.

[3] Krischer JP, Schatz DA, Bundy B, Skyler JS, Greenbaum CJ. Effect of Oral Insulin on Prevention of Diabetes in Relatives of Patients With Type 1 Diabetes: A Randomized Clinical Trial. JAMA. 2017;318(19):1891-1902.






Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s