Redosing of anti-CD3 antibodies in NOD mice with new-onset diabetes does not alter the effect of a single treatment course.

aims/hypothesisUncertainty exists regarding the effectiveness of redosing teplizumab, a humanised anti-CD3 monoclonal antibody, in individuals with stage 2 and 3 type 1 diabetes. This study aimed to investigate the impact of redosing an anti-CD3 antibody on disease reversal and immune responses in NOD mice.

methodsNew-onset diabetic NOD mice were treated with a short course (days 0-4) of low-dose anti-CD3 (2.5 µg/day i.v.), followed by a second treatment course administered either early (days 14-18) or late (days 43-47) following the initial course. Diabetes outcomes, anti-drug antibody titres, inflammatory markers and the profiles of peripheral blood and pancreatic immune cells were assessed.

resultsA single course of anti-CD3 therapy reversed type 1 diabetes in 17 out of 40 treated mice (43%). However, neither early nor late redosing improved therapeutic outcomes in non-responder mice or disrupted re-established tolerance in responder mice. Early redosing, administered at the time of CD3⁺ T cell repopulation, and late redosing, timed when anti-drug antibody titres declined after their peak at 14 days, failed to induce remission in non-responder mice (0/10 and 0/5, respectively). Importantly, redosing did not trigger relapse in mice that had achieved remission following the initial treatment course, with all remaining normoglycaemic (5/5 with early redosing; 23/23 with late redosing). Inflammatory markers increased transiently, regardless of treatment response or regimen. In the pancreas, responder mice had a pronounced shift from an effector (CD62L⁻CD44⁺) to a naive (CD62L⁺CD44⁻) phenotype in both CD4⁺ and CD8⁺ T cells, independent of treatment regimen, indicative of a more regulated immune environment. These mice also exhibited elevated frequencies of CD4⁺ and CD8⁺ T cells with an exhausted (programmed cell death protein 1 [PD1]⁺killer cell lectin-like receptor subfamily G, member 1 [KLRG1]⁺) profile in their pancreas, a trait not observed in non-responder mice. Furthermore, the proportion of natural killer (NK) (NKp46⁺CD3⁻) cells was enriched in the pancreas of non-responder mice. CONCLUSIONS/

interpretationThese findings suggest that while a single course of anti-CD3 therapy can be effective in a subset of mice, redosing does not enhance efficacy yet is well tolerated in previously reversed mice.