Selective unresponsiveness to beta cell autoantigens after induction immunosuppression in pancreas transplantation with anti-interleukin-2 receptor antibody versus anti-thymocyte globulin

Abstract

Pancreas transplantation in type 1 diabetes patients could result in (re)activation of allo- and autoreactive T lymphocytes. Anti-thymocyte globulin (ATG) induction treatment is a successful, but broadly reactive anti-lymphocyte therapy used in pancreas and islet transplantation. A more selective alternative is daclizumab, a monoclonal antibody directed against the interleukin-2 receptor (CD25) on activated lymphocytes. We tested the hypothesis that daclizumab is more selective and has less immunological side effects than ATG. Thirty-nine simultaneous pancreas-kidney transplantation patients with type 1 diabetes were randomized for induction therapy with ATG or daclizumab. Auto- and recall immunity was measured cross-sectionally by lymphocyte stimulation tests with a series of auto- and recall antigens in 35 successfully transplanted patients. T cell autoimmunity to islets was low in both groups, except for a marginal but significantly higher reactivity against glutamic acid decarboxylase (GAD)65 in daclizumab-treated patients. The memory responses to recall antigens were significantly higher in the daclizumab-treated group compared to ATG-treated patients, specifically against purified protein derivative (PPD) (anti-bacterial immunity), Haemophilus influenzae virus matrix protein-1 (anti-viral immunity) and p53 [anti-tumour (auto)immunity]. These data imply that daclizumab is more specifically affecting diabetes-related immune responses than ATG. The autoimmunity is affected effectively after daclizumab induction, while memory responses towards bacterial, viral and tumour antigens are preserved.

Fig. 1

Recall responses after anti-thymocyte globulin (ATG) Fresenius (□) daclizumab (▪) induction therapy only (a) and after second antibody therapy with ATG Merieux in ATG Fresenius (○) and daclizumab (•) treated patients (b); the anti-bacterial [purified protein derivative (PPD)] and the anti-viral (Haemophilus influenzae virus matrix protein M1) T cell response was decreased in patients with ATG induction versus daclizumab induction. (a) Significant differences in recall immunity were found for the daclizumab-treated patients with versus without second antibody treatment. Stimulation indices in these treatment groups were, respectively, 2·0 versus 16·9 for PPD (P = 0·002) and 8·3 versus 23·4 for H. influenzae virus matrix protein M1 (P = 0·03) (a and b).

Fig. 2

Recall responses after anti-thymocyte globulin (ATG) Fresenius (□) daclizumab (▪) induction therapy only (a) and after second antibody therapy with ATG Merieux in ATG Fresenius (○) and daclizumab (•) treated patients (b). In both the ATG-treated group and the daclizumab-treated group, low autoreactivity was found. No statistical differences were found between both groups for all auto-antigens except for glutamic acid decarboxylase (GAD)65. Stimulation indices for GAD65 were 1·0 versus 2·1 for ATG Fresenius and daclizumab induction, respectively (P = 0·02) (a). No significant differences were found in proliferative responses upon stimulation with either auto or recall antigens in ATG-treated patients with versus without second antibody therapy, except for p53 stimulation. Stimulation index was 0·9 versus 12·8 (P = 0·02) (a and b).

Fig. 3

Recall responses after daclizumab without second antibody therapy with anti-thymocyte globulin (ATG) Merieux rejection episode treatment (▪) and with second antibody therapy with ATG Merieux rejection episode treatment (•) (a) and auto responses after daclizumab without second antibody treatment (▪) and with second antibody treatment (•) (b); Significant differences in recall immunity were found for the daclizumab-treated patients with versus without second antibody treatment. Stimulation indices in these treatment groups were, respectively, 2·0 versus 16·9 for purified protein derivative (PPD) (P = 0·002) and 8·3 versus 23·4 for Haemophilus influenzae virus matrixprotein M1 (P = 0·03) (a). No significant differences were found in proliferative responses upon stimulation with autoantigens (b). Data from Figs 1 and 2 have been combined for this comparison.

Fig. 4

Interleukin (IL)-2 responses after anti-thymocyte globulin (ATG) Fresenius (□) and daclizumab (▪) induction therapy only and after second antibody therapy with anti-thymocyte globulin (ATG) Merieux in ATG Fresenius (○) and daclizumab (•) treated patients; ATG and daclizumab induction therapy was associated with stimulation indices of: 2·7 and 26·1, respectively (P = 0·0003). After second antibody therapy with ATG Merieux the stimulation index was significantly higher compared to ATG-treated patients without second antibody therapy (SI 31·0 versus 2·7, respectively; P = 0·01). For daclizumab induction, second antibody therapy was associated with lower stimulation indexes of 5·6 compared to 26·1 in daclizumab-treated patients without second antibody therapy (P = 0·03) (a and b).

Fig. 5

Percentages of CD4⁺ T cells (mean ± standard deviation) for the different treatment groups.