Azathioprine and cyclophosphamide in doses of 3 mg/kg were assessed in humans with immunopathic diseases in respect to effects on humoral antibody production and counts of antigen-binding lymphocytes in blood, using flagellin as test antigen. The results for azathioprine-treated patients did not differ significantly from those for nontreated control patients, whereas for cyclophosphamide-treated patients there was significant suppression of humoral antibody to flagellin, especially IgG, and no increase in antigen-binding lymphocytes in blood after immunization. These and other observations, relating to inhibitory effects of azathioprine on T cells and antiproliferative effects of cyclophosphamide on B cells, could have implications for therapy of immunopathic diseases.Drugs which interfere with immune responses are referred to as "cytotoxic" because of their derivation from chemotherapy of leukemia, and "immunosuppressive" because of their capacity to hinder rejection of allografts. However, uncertainties exist concerning the therapeutic action of such drugs because of insufficient knowledge about a) the relative contributions of humoral and cellular responses in various immunopathies and b) the effects in relation to dose of these drugs on immune re- sponses. T h e present study examines relative effects on the humoral response of "conventional" doses of the two most frequently used of these drugs, azathioprine and cyclophosphamide, using production of humoral antibody to flagellin and antigen-binding to peripheral blood lymphocytes.
MATERIALS AND METHODS
Design of StudyPatients who had been receiving azathioprine or cyclophosphamide for varying periods, usually 3 to 12 months, were immunized with monomeric flagellin, and comparisons were made of their immune response by titration of antibody to flagellin and counts of flagellin-binding lymphocytes in peripheral blood. Equivalent observations were made on an age-sex matched contrast group (controls) likewise immunized with Ragellin.
ImmunizationMonomeric flagellin from the nonpathogen Salmonella adelaide, prepared by the method of Ada et a1 (l), was sterilized by Seitz filtration and stored at -20" C. The antigen was stored in phosphate-buffered saline (PBS) at a concentration of 50 pg/ml, and depolymerized immediately before injection (2). A primary immune response was initiated by injecting 5 pg of flagellin subcutaneously, and serum samples for antibody determinations were drawn before, and at 1 , 2, 6, and 10 weeks after immunization. Antibody was ti-