Novel Anti-CD3 Antibody Delays Progression to T1D

SAN FRANCISCO — A prophylactic course of teplizumab helped to delay the development of type 1 diabetes in at-risk individuals, a researcher reported here.

In a group of relatives of people with type 1 diabetes, a 14-day course of treatment with the anti-CD3 antibody teplizumab reduced the risk of of developing type 1 diabetes (adjusted HR 0.41, 95% CI 0.22-0.78, P=0.006). The annualized rates of diagnosis of diabetes were 14.9% per year in the teplizumab group and 35.9% per year in the placebo group, according to Kevan Herold, MD, of Yale University in New Haven.

This remained significant even after adjustment for age, baseline oral glucose tolerance test, and the presence of anti-GAD65 antibody in the phase 2 TrialNet study, Herold reported at the American Diabetes Association (ADA) annual meeting. The results were simultaneously published in the New England Journal of Medicine.

In the 60 months following randomization, only 43% of those who received teplizumab were diagnosed with type 1 diabetes compared with 72% of those who received placebo. Each year, the rate of type 1 diabetes diagnosis was about 36% in the placebo group versus 15% in the teplizumab group.

In those who did progress to a diagnosis of type 1 diabetes, development was significantly delayed in those who received teplizumab compared with placebo, with a median time to diagnosis of 48.4 months versus 24.4 months, respectively.

“This is the first trial that has met its clinical endpoint in delaying the onset of type 1 diabetes,” said Herold during an ADA press conference. “The 2-year delay in diagnosis is clinically important. In light of the daily burden of disease management, any time without clinical diabetes has significance. And I think if you were to ask the families of any of the participants in this trial, or even the families of people with diabetes, they would reiterate this point.”

“But it’s particularly an issue for children, who represent about three-quarters of the participants in our study,” he stated.

The study by Herold and colleagues included 76 individuals who were free of diabetes at baseline but had a relative with diagnosed type 1 diabetes, the majority of whom were under age 18 years. Most participants had a sibling with type 1 diabetes.

All individuals were at least age 8 years at the time of the study and had at least two diabetes-related autoantibodies detected in two different samples within the past 6 months. The most common autoantibodies were harmonized anti-GAD65, followed by anti-ZnT8, harmonized anti-IA-2, ICA, and micro insulin.

“We now understand that essentially all close relatives of people with type 1 diabetes and who also have multiple antibodies can be considered as having the early, asymptomatic form of the disease,” said Carla Greenbaum, MD, of Benaroya Research Institute in Seattle and chair of TrialNet, in a statement. “Just as we treat the asymptomatic presence of hypertension to prevent a heart attack or a stroke, these findings provide strong evidence we are approaching a future in which we can identify and treat type 1 diabetes long before symptoms occur.”

All included individuals had dysglycemia prior to the study, marked by a fasting glucose between 110-125 mg/dL, a 2-hour post-prandial plasma glucose ≥140-200 mg/dL, or an intervening postprandial glucose levels >200 mg/dL at two occasions.

Among the group, 44 individuals were randomized to receive teplizumab, which works by reducing the actions of CD8+T lymphocytes on β-cell targets. This intravenous infusion was given as a 51 µgram/m² of body-surface area on at day 0. This was then increased to 207 µg/m² on day 2, 413 µg/m² on day 3, and 826 µg/m² each day from days 4-13.

The first year after teplizumab therapy yielded the largest benefit for patients, reducing the risk for developing type 1 diabetes by 87% compared with the saline placebo (unadjusted HR 0.13, 95% CI 0.05-0.34).

In a subgroup analysis broken down by the presence and absence of some HLA types and autoantibodies, certain patient characteristics were associated with a greater risk reduction for developing type 1 diabetes:

• Presence of HLA-DR4: HR 0.20 (95% CI 0.09-0.45)
• Absence of HLA-DR3: HR 0.18 (95% CI 0.07-0.45)
• Absence of anti-ZnT8 antibodies: HR 0.07 (95% CI 0.02-0.26)

Teplizumab was associated with some adverse events (AE), with 112 events occurring in the study arm versus 23 events reported with placebo. A total of 75% of participants who received teplizumab experienced a form of AE relating to blood or bone marrow, such as transient lymphopenia. Dermatologic AEs, such as rashes, were also significantly more common with teplizumab (36% vs 3%), although the researchers said both these events were expected with the treatment.

In an accompanying editorial, Clifford Rosen, MD, of the Maine Medical Center Research Institute in Scarborough, and Julie Ingelfinger, MD, of Massachusetts General Hospital in Boston called the findings “striking,” but cautioned that these “results should not be taken to imply that immune modulation constitutes a potential curative approach.”

“Rather, these data provide strong albeit indirect evidence about the pathogenesis of beta-cell destruction and the potential to modify the course of type 1 diabetes with newer biologic agents,” they commented, adding that this particular trial “will probably prompt the development of more refined screening criteria for treatment of persons at highest risk, although challenges in using immune modulators for type 1 diabetes remain.”

Rosen and Ingelfinger said questions remain regarding this therapy, such as long-term side effects, the frequency and duration of treatments needed for the best effect, and separating treatment responders from non-responders.

last updated 06.09.2019