Induction of oral tolerance in human autoimmune thyroid disease

Our laboratory has reported suppression of experimental autoimmune thyroiditis in mice by oral feeding with antigen. Based on these data, we considered it possible that oral feeding of animal thyroglobulin (TG) might induce tolerance to antigen in human autoimmune thyroid disease (AITD). Thirteen patients receiving thyroid hormone replacement with synthetic thyroxine (T4) (five patients with Graves' disease, treated with radioiodine 4 to 11 years ago and eight patients with Hashimoto's thyroiditis) were randomly assigned to a test group (switched to replacement with desiccated thyroid from porcine thyroids) and a control group (maintained on synthetic T4). Humoral and cellular immunologic parameters were evaluated in addition to clinical parameters before and every 3 months after the onset of study for a year. At the onset of study, there was no difference in clinical parameters, or humoral and cellular immunity to thyroid autoantigens, except a finding that one thyroid peroxidase (TPO) peptide (100 approximately 119) appeared to stimulate peripheral blood mononuclear cells (PBMC) during in vitro microproliferation assay more in the test group than control group (p = 0.051 by t test). Additionally, almost all of TPO and thyrotropin receptor extracellular domain (TSHR) peptides were slightly more stimulatory to PBMC from the test group than the control group, although this was not statistically significant. After treatment, all variables were analyzed at each time point between groups (t test), and also were analyzed over time in each group (analysis of variance, ANOVA). Among the clinical parameters, thyrotropin (TSH) levels were unchanged and equal. Total serum T4 levels (p < 0.05 at 6 and 12 months after treatment) and free thyroxine indices (FT4I) (p < 0.05 at all time points after treatment) were lower in the test group than the control group. This is an expected result of treatment with desiccated thyroid. We found no change over time nor any difference between groups at time points for titers of antibodies to thyroid autoantigens, ie, human TG, human TPO, and recombinant human TSHR from Escherichia coli. However, cellular immunity, measured by in vitro microproliferation of PBMC to peptides of TPO or TSHR, showed significant differences between groups. At 12 months, stimulatory indices (SI) of PBMC to six peptides, containing the indicated amino acids (764 approximately 95, 100 approximately 119, 110 approximately 129, 261 approximately 275, 441 approximately 448, 708 approximately 727) of 10 TPO peptides, and one peptide (145 approximately 163) of 14 TSHR peptides were lower in the test group than control group (p < 0.05). SI of PBMC to phytohemagglutinin, purified protein derivative from mycobacteria, and tetanus toxoid were not different between groups nor changed over time in any group. In conclusion, treating patients with AITD with an antigen related to the autoantigen TG did not produce changes in humoral immunity parameters, while cellular immunity to certain peptides were apparently suppressed. While the results are both surprising and intriguing, we need more evidence to justify the use of autoantigen as a form of immunospecific therapy in patients with AITD.