Background: The T1DAL trial1,2 is an ongoing US clinical trial evaluating the ability of alefacept to preserve beta cell function in new-onset Type 1 diabetes (T1D). Alefacept (LFA3-Ig) binds to cells highly expressing CD2 and has been shown to delete memory T cells. The study has reached its 12-month primary endpoint: although the 2h C-peptide AUC was not different, the 4h AUC was significantly greater and insulin use and hypoglycemia were significantly lower in the alefacept vs. placebo group, suggesting alefacept may be able to preserve beta cell function. To better understand the mechanism of alefacept in T1D, we used multicolor flow cytometric analyses to evaluate the mean-florescence intensity (MFI) of CD2 on peripheral blood lymphocytes and changes in specific lymphocyte subsets in alefacept- and placebo-treated patients.
Results: At baseline, CD2 expression (MFI) was highest on CD8 and CD4 effector-memory (Tem; CD45RA-CCR7-) and CD8 cental-memory (Tcm; CD45RA-CCR7+) T cells and CD56hi/CD16lo NK cells; intermediate on CD4 Tcms; low on CD4 and CD8 naïve (Tn; CD45RA+CCR7+) T cells, Tregs (CD4+FoxP3+CD127lo), and lytic (CD56lo/CD16hi) NK cells; and very low on naïve (CD27-) and memory (CD27+) B cells. Through 12 months, alefacept subjects had significant, sustained reductions of CD4 Tems and CD4 and CD8 Tcms. CD56hi/CD16lo NK cells were substantially depleted during therapy, but quickly rebounded. There were no significant changes in CD8 Tems, CD4 or CD8 Tns, Tregs, lytic NK cells or native or memory B cells.
Conclusions: Declines in CD4 and CD8 Tcms, CD4 Tems, and CD56hi/CD16lo NK cells and sparing of Tregs, CD4 and CD8 Tns, B cells and lytic NK cells correlated with CD2 expression. It is unclear why CD8 Tems were not depleted. Depleting memory T cells and possibly CD56hi/CD16lo NK cells, while sparing Tregs, may be a signature of beneficial therapies in T1D.