Immune profiling of plasma and cervical secretions using recycling immunoaffinity chromatography

Small volumes of cervical secretions have limited measurements of immunity at the cervix, which may be important to studies of human papillomavirus (HPV). We report the use of recycling immunoaffinity chromatography to efficiently study immune profiles in cervical secretions. Frozen pairs of plasma and cervical secretions (collected on ophthalmic sponges) were selected randomly from women with normal cervical cytology (n = 50) participating in a natural history study of HPV in Guanacaste, Costa Rica. Single 25- micro l aliquots of plasma and (diluted) cervical secretions were assayed for interleukin (IL) -1 beta, -2, -4, -6, -8, -10, -12, -13, -15, IFN-alpha, -beta, -gamma, tumor necrosis factor-alpha, -beta, RANTES (regulated on activation normal T-cell express and secreted), MCP-1 (monocyte chemoattractant protein), -2, -3, macrophage inflammatory protein-1 alpha, -1 beta (regulated on activation normal T-cell express and secreted), macrophage colony-stimulating factor, IgG, IgA, and cyclooxygenase 2. All of the specimens were tested as blind replicates, and refrozen plasma was retested 4 months later. To evaluate the reproducibility of the repeat measurements and to examine the correlation between plasma and cervical secretions, we calculated kappa values with 95% confidence intervals among categorized analyte values and Spearman correlation coefficients (rho) among detectable, continuous analyte values. Measurements of all of the analytes in either plasma or cervical secretions were highly reproducible, with all of the kappa > or = 0.78 (70% above 0.90), and all of the rho > or = 0.88 (96% above 0.90). Only IL-1 beta (kappa = 0.60 and rho = 0.82) and IL-6 (kappa = 0.50 and rho = 0.78) levels were strongly correlated between plasma and cervical secretions. IFN-gamma, tumor necrosis factor-beta, RANTES, MCP-1, MCP -2, macrophage inflammatory protein-1 alpha, and macrophage colony-stimulating factor levels were especially poorly correlated between plasma and cervical secretions (kappa < or = 0.25 and rho < or = 0.25). We conclude that recycling immunoaffinity chromatography is a reproducible method of measuring immune profiles from biological specimens, and immune profiles are not well correlated between plasma and cervical secretions, perhaps necessitating cervical collections to study cervix-specific immunity in HPV natural history studies.