(T1D) have had mixed results, with some therapies—anti-CD3 monoclonal antibodies targeting T cells (1–7), anti-CD20 monoclonal antibodies targeting B cells (8), and costimulation blockade (9)—showing promise, with at least transient improvement in β-cell function compared with randomized control groups. In the current issue, Herold et al.(10) from the Immune Tolerance Network (ITN) report the results of the Autoimmunity-Blocking Antibody for Tolerance in Recently Diagnosed Type 1 Diabetes (AbATE) trial. This is the fifth trial with anti-CD3, the fourth with the monoclonal antibody teplizumab, demonstrating preservation of β-cell function. The article describes a group of “responders” to treatment, identified by the label “responders” for those who maintained C-peptide better than the randomized but untreated comparison group at 24 months. Responders, defined in this way, constituted 45% of the subjects treated with anti-CD3. When examining β-cell function over time in the trial, it was evident that the responders had maintained β-cell function for 2 years, whereas the nonresponders had lost β-cell function at a rate similar to the control group. This is a crucial observation, because in an analysis that includes both responders and nonresponders, the profound retention of C-peptide in nearly half of subjects can be missed. The fundamental question is why some subjects failed to respond. It could be that the immunotherapy was ineffective (at least at the dose used), that the immunologic process—perhaps a relapsing and remitting one—was in a latent period at the time of drug administration, that β-cell mass or function had already deteriorated to a point of no return, that the immunologic processes damaging β-cells are different among individuals, or for some other reason.

It turns out that at baseline, prior to treatment, the responders had lower HbA1c levels and used less insulin than the nonresponders. Unfortunately, there is not an unambiguous demarcation of HbA1c level or of insulin dose to identify responders a priori, but rather there is overlap of HbA1c levels and of insulin dose between responders and nonresponders. However, the lower HbA1c levels and lower insulin doses imply that the responders may have had a milder disease or be earlier in the course of the disease, consistent with comments by Jean-Francois Bach (11) that:“Ideally, type 1 diabetes should be regarded as a medical emergency and treatment with teplizumab could be started within a few days after diagnosis, as compared