Anti-cytokine autoantibodies are associated with opportunistic infection in patients with thymic neoplasia

Abstract

Patients with thymic malignancy have high rates of autoimmunity leading to a variety of autoimmune diseases, most commonly myasthenia gravis caused by anti-acetylcholine receptor autoantibodies. High rates of autoantibodies to cytokines have also been described, although prevalence, spectrum, and functionality of these anti-cytokine autoantibodies are poorly defined. To better understand the presence and function of anti-cytokine autoantibodies, we created a luciferase immunoprecipitation system panel to search for autoantibodies against 39 different cytokines and examined plasma from controls (n = 30) and patients with thymic neoplasia (n = 17). In this screen, our patients showed statistically elevated, but highly heterogeneous immunoreactivity against 16 of the 39 cytokines. Some patients showed autoantibodies to multiple cytokines. Functional testing proved that autoantibodies directed against interferon-α, interferon-β, interleukin-1α (IL-1α), IL-12p35, IL-12p40, and IL-17A had biologic blocking activity in vitro. All patients with opportunistic infection showed multiple anti-cytokine autoantibodies (range 3-11), suggesting that anti-cytokine autoantibodies may be important in the pathogenesis of opportunistic infections in patients with thymic malignancy. This study was registered at http://clinicaltrials.gov as NCT00001355.

Figure 1

Quantitative immunoglobulins (Igs) and lymphocyte phenotyping for all patients enrolled on study. (A) IgG, IgA, and IgM levels. Lymphocyte subsets in (B) absolute and (C) percentage of total lymphocyte count; memory lymphocyte subsets in (D) absolute and (E) percentage of total lymphocyte count based on complete blood count with differential performed on the same day.

Figure 2

Anti-cytokine autoantibodies in patients with thymic neoplasia. Shown are results from 30 healthy volunteers (HVs), 12 patients with thymic neoplasia without opportunistic infection (Thym/OI−), and 5 patients thymic neoplasia with opportunistic infection (Thym/OI+). Each symbol represents a sample from one patient. The antibody titers in LU for (A) anti–IFN-γ, (B) anti–IFN-α, (C) anti–IFN-ω, (D) anti–IFN-λ, (E) anti–IFN-β, and (F) anti–IFN-ε, (G) anti–IL-12p40 (H) anti–IL-12p35, (I) anti–IL-17, (J) anti-BAFF, (K) anti–IL-1α, and (L) anti–TNF-α antibody titers are plotted on the y-axis using a log₁₀ scale. The geometric mean antibody titer for each group is shown by the short line. The dashed and solid lines represent the cutoff levels for determining seropositivity and are derived from the mean + 3 SDs and mean + 5 SDs of the antibody titer of the 30 controls, respectively. P values for the different groups were calculated using the Mann-Whitney U test.

Figure 3

Heatmap analysis of anti-cytokine autoantibody titers for plasma from 17 patients with thymic neoplasia and 30 healthy volunteers. Shown are the 16 cytokines that had any positive result of the 38 cytokines tested. Titer values greater than the mean plus 3 SDs of the 30 healthy volunteers were color-coded from green to red to signify the relative number of SDs above these reference values. The patient codes colored blue indicate the thymoma patients with OI: patients 1 and 5 had CMC, patient 2 had CMC pulmonary M avium and sinopulmonary S apiospermum, patients 1 and 17 had dVZV, and patient 9 had disseminated cryptococcosis.