Protégé: Teplizumab fails in type 1 diabetes prevention; endpoint may have been poor choice

Teplizumab, an anti-CD3 monoclonal antibody designed to preserve beta cell function and decrease insulin needs in patients with recent-onset type 1 diabetes, was no better than placebo at meeting a composite endpoint in the phase III international Protégé trial.
     Among 516 patients randomized to receive a 14-day full-dose or low-dose course of teplizumab, 6-day full-dose course of teplizumab, or placebo, the percentage of patients who at 1 year met the primary endpoint of glycosylated hemoglobin (HbA_1c) less than 6.5% and insulin use less than 0.5 U/kg/day was 19.8% in the 14-day full-dose group, 13.7% in the 14-day low-dose group, 20.8% in the 6-day full-dose group, and 20.4% in the placebo group, reported Nicole Sherry, MD, Director of the Diabetes Center at Massachusetts General Hospital for Children in Boston, on behalf of study colleagues.
     But it may be the new primary endpoint, and not the drug itself, that failed, Sherry suggested. When they reanalyzed their data in exploratory analyses using an endpoint that has been used in other trials-preservation of C-peptide secretion and lower insulin use-they found a significant advantage for 14-day full-dose teplizumab over placebo (area under the curve [AUC] of C-peptide from baseline of -0.96 mmol/L/minute for the antibody, vs. -0.14 mmol/L/minute for placebo; P=.046).
     The study was published simultaneously online in The Lancet. In an accompanying editorial, Jean Francois Bach, MD, from Hopital Necker-Enfants Malades in Paris, France, suggested that starting teplizumab within a few days of diagnosis, rather than after weeks or months, might provide better protection of remaining beta cells.
     Teplizumab is a humanized monoclonal antibody that has been mutated to reduce Fc receptor and complement binding, theoretically protecting the approximately 30% of endogenous beta cells that remain when a clinical diagnosis of type 1 diabetes is made.
     The Protégé investigators enrolled patients aged 8 to 35 years who had been diagnosed with type 1 diabetes within the past 12 weeks. The patients were randomly allocated on a 2:1:1:1 basis to the treatment groups as described above, with the assigned treatment given in infusions at baseline and at week 26 for the specified schedules.
     Although the study did not meet its primary endpoint, 5% of patients in the combined teplizumab group were not taking insulin at 1 year, compared with no patients in the placebo group.
     In addition, among all patients on teplizumab, 40% had preservation of C-peptide, compared with 20% of those on placebo. C-peptide was also preserved in children (55% vs. 29% of adults), patients treated in the United States (33% vs. 13% treated elsewhere), and those who had been diagnosed within 6 weeks. These findings suggest that future studies of teplizumab should consider specific patient populations, Sherry said.
     Adverse events occurred in 99% of patients in both the teplizumab and placebo groups. The most common clinical adverse event among patients on teplizumab was rash, occurring in 53% of patients, compared with 20% on placebo.