Type 1 diabetes results from T cell mediated destruction of beta cells. We conducted a trial of antithymocyte globulin (ATG) in new-onset type 1 diabetes (the Study of Thymoglobulin to ARrest T1D [START] trial). Our goal was to evaluate the longer-term safety and efficacy of ATG in preserving islet function at 2 years.

A multicentre, randomised, double-blind, placebo-controlled trial of 6.5 mg/kg ATG (Thymoglobulin) vs placebo in patients with new-onset type 1 diabetes was conducted at seven university medical centres and one Children’s Hospital in the USA. The site-stratified randomisation scheme was computer generated at the data coordinating centre using permuted-blocks of size 3 or 6. Eligible participants were between the ages of 12 and 35, and enrolled within 100 days from diagnosis. Subjects were randomised to 6.5 mg/kg ATG (thymoglobulin) vs placebo in a 2:1 ratio. Participants were blinded, and the study design included two sequential patient-care teams: an unblinded study-drug administration team (for the first 8 weeks), and a blinded diabetes management team (for the remainder of the study). Endpoints assessed at 24 months included meal-stimulated C-peptide AUC, safety and immunological responses.

Fifty-eight patients were enrolled; at 2 years, 35 assigned to ATG and 16 to placebo completed the study. The pre-specified endpoints were not met. In post hoc analyses, older patients (age 22–35 years) in the ATG group had significantly greater C-peptide AUCs at 24 months than placebo patients. Using complete preservation of baseline C-peptide at 24 months as threshold, nine of 35 ATG-treated participants (vs 2/16 placebo participants) were classified as responders; nine of 11 responders (67%) were older. All participants reported at least one adverse event (AE), with 1,148 events in the 38 ATG participants vs 415 in the 20 placebo participants; a comparable number of infections were noted in the ATG and placebo groups, with no opportunistic infections nor difficulty clearing infections in either group. Circulating T cell subsets depleted by ATG partially reconstituted, but regulatory, naive and central memory subsets remained significantly depleted at 24 months. Beta cell autoantibodies did not change over the 24 months in the ATG-treated or placebo participants. At 12 months, ATG-treated participants had similar humoral immune responses to tetanus and HepA vaccines as placebo-treated participants, and no increased infections.

A brief course of ATG substantially depleted T cell subsets, including regulatory cells, but did not preserve islet function 24 months later in the majority of patients with new-onset type 1 diabetes. ATG preserved C-peptide secretion in older participants, which may warrant further study.

ClinicalTrials.gov NCT00515099

START datasets are available in TrialShare www.itntrialshare.org

National Institute of Allergy and Infectious Diseases (NIAID) of the National Institutes of Health (NIH). The trial was conducted by the Immune Tolerance Network (ITN).