Response to Comment on Russell et al. Abatacept for Delay of Type 1 Diabetes Progression in Stage 1 Relatives at Risk: A Randomized, Double-Masked, Controlled Trial. Diabetes Care 2023;46:1005–1013

Cyclosporine studies beginning in the 1980s confirmed that immunosuppressive therapies promote remission of type 1 diabetes (T1D), but as Dr. Pinheiro notes (1), the efficacy was not permanent, in some cases even with ongoing treatment (2,3). TrialNet studies and others have subsequently shown that a variety of immunologic interventions targeting antigens, T cells, B cells, and their downstream cytokines modify the progression of autoimmune diabetes. Most studies have been single-agent trials that focus on a mechanism and have used component analysis to determine the prospect of benefit in children.

Our study, while not meeting its prespecified target of risk reduction, showed that abatacept-treated individuals had delayed progression to stage 2 disease and demonstrated key clinical, β-cell, and immunologic indicators of drug effect, but only for the 1 year of treatment. We speculate that longer therapy might result in prolonged delay in diabetes progression. Preselection of those who are most likely to respond might improve the outcome of this therapy for prevention.

We agree with Dr. Pinheiro that the field is ripe for combination trials. A TrialNet study that combines B-cell– and T-cell–targeted therapy using rituximab, followed by 2 years of abatacept treatment, in individuals with stage 3 T1D will soon recruit participants. However, we differ in our assessment of the efficacy of agents that enhance β-cell function in the setting of T1D. In patients with stage 3 T1D, imatinib showed some efficacy on β-cell function, while exenatide, dipeptidyl peptidase-4 inhibitors, and liraglutide as adjuncts to insulin or immune therapy have had limited effects even when residual function is present (4,5). It is similarly unknown whether any of these would have long-term effects after discontinuation and whether β-cell stress causes immune dysfunction, results from it, or is an independent event. Further mechanistic analysis of our study may reveal changes in β-cell stress markers, such as the proinsulin-to-insulin ratio, which would be an indication for a metabolic treatment. Trials with combinations selected for their likely positive interactions based on mechanistic analysis of data from prior clinical trials are under review at TrialNet and elsewhere.

Dr. Pinheiro’s comments invoke the insights of Best and colleagues (6), who noted that after development of diabetes in partially pancreatectomized dogs, residual islets exhibited “progressive degranulation and finally hydropic degeneration.” They speculated that “overwork resulting from excessive demands on this organ” were responsible. Administration of insulin promoted the health of remnant islets, and they proposed that “the prophylactic administration of insulin to potential diabetic patients may become an accepted clinical procedure in the future.” The admonition of Best and associates rings true, in that promotion of β-cell health in the failing diabetic pancreas is an important goal as we continue our efforts to curb the immune attack, but this enthusiasm itself must be combined with a strategic approach that builds each new trial on specific mechanistic insights of prior trials. In 1940 there was little understanding that immune-mediated islet damage initiated disease in young patients with diabetes, but their speculations on ameliorating β-cell damage, whatever the originating insult, remain relevant.