n the October 17 1940 issue of the New England Journal
of Medicine,
Dr. Charles Best (one of the co-discoverers of insulin) and several coauthors
said that insulin should be tested as protection from diabetes
for
relatives of people with diabetes.
Animal studies that were done in the
1980's looked promising, and a pilot
study in 12 human beings showed that giving insulin
delayed
the onset of diabetes in 7 people, compared with
the onset in 5 people who chose not to
participate.
With this background in mind, the NIH, together with the American Diabetes Association,
the Juvenile Diabetes Research Foundation, and corporate sponsors, banded together
in an effort to answer the question: Can antigen-based therapy
using insulin
delay the onset of diabetes in relatives
of people with
type 1 diabetes?
Screening for participation in the study
The first step in the study was finding people who
were thought to be at high risk of developing
diabetes in the near future.
Eligibility requirements included first-degree relatives
of people with diabetes
who were between the ages of 3 and 45,
or
second-degree relatives
between the ages of 3 and 20.
Screening consisted of measurement of
islet-cell antibodies.
If the antibody test
was positive,
metabolic testing was performed using Intravenous Glucose Tolerance Test (IVGTT)
and Oral Glucose Tolerance Tests (OGTT).
Depending on the results of these, subjects were assigned to one of two study groups.
Design of the Study
The DPT-1 study included two major study groups:
- Injections of insulin (parenteral therapy) to subjects whose risk of developing diabetes in the next 5 years
is thought to be 50% or higher. Results of this part of the study were
presented June 23.
- Oral administration of insulin to
subjects whose risk of developing diabetes in the
next 5 years is thought to be between 26-50%.
This part of the study is ongoing, and indeed recruiting of additional subjects continues.
Patients were enrolled over several years,
with nearly 100,000 people undergoing initial screening of their
antibody status over the course of
seven years, and
after further evaluations (See above),
339 subjects met the study's criteria
for parenteral therapy
(positive antibody testing, abnormal first-phase insulin release, and glucose tolerance
tests that did not show diabetes).
Active treatment was 4 days of intravenous insulin annually, with daily injections of 0.25 units/kg/day
of Ultralente insulin.
The control group received no treatment.
Patients were randomized into either active or control groups.
Results
- Of the
people who were initially screened and had positive antibodies,
11.2% already had diabetes (based on blood sugar testing).
- The results of the parenteral insulin part of the DPT-1 study showed
no difference
in the chances of
people developing diabetes
if they took low-dose insulin by injection,
compared with people who did not receive insulin.
Sixty percent of both the treated group, and the untreated group, developed diabetes within
5 years.
Children under the age of 13 progressed to diabetes more rapidly than kids post-puberty or adults.
- 73% of those who developed diabetes during the DPT-1
were asymptomatic at the time of their diagnosis, and the
diagnosis was picked up only because they were in the study.
This shows that it is indeed possible to diagnose
type 1 diabetes before symptoms
develop.
Speculation on why this study failed to find a difference
centered around the thought
that perhaps the patients studied were too far along in the
process of developing diabetes to
be able to reverse the process.
Or perhaps the dose of insulin was wrong -- only one dose was
tested.
Results from the Oral Part of the Study are still pending
It must be noted that the DPT-1
has reported the results of one arm of the study (insulin-by-injection).
As mentioned above, the second arm of the
study (using oral insulin) is still proceeding.
This arm of the study will be evaluating patients who
are presumably earlier in the beta-cell destruction
process, and therefore might be more likely to
respond to the intervention, and have a delay in the
onset of diagnosis.
Also see
The study is published in
the
New England Journal of Medicine,
at
Effects of Insulin in Relatives of Patients with Type 1 Diabetes Mellitus,
Volume 346:1685-1691 May 30, 2002 Number 22
Insulin Injections Fail To Prevent Type 1 Diabetes
Press Release, June 23, 2001,
from the National Institute of Diabetes and Digestive and Kidney Disease/NIH
Research:
The Diabetes Prevention Trial
at the DIABETES REGISTRY
