Background: Effective and safe topical anti-HIV-1 microbicides that have minimal or no impact on the viability, function, and structural integrity of the vaginal and cervical epithelium are sought to reduce the spread of the AIDS pandemic. The safety assessment of these products requires reliable and reproducible assays and sensitive surrogate markers of mucosal toxicity. The presence of the proinflammatory cytokine interleukin (IL)-1 b in human vaginal secretions, as sampled by cervicovaginal lavage, is emerging as a characteristic indicator of pathologic inflammation and compromised mucosal integrity.

Methods: Recovery of human IL-1b from complex biological matrices was determined using seven different immunoassays and three detection platforms (traditional ELISA, Meso-Scale Discovery (MSD) Electrochemiluminescent Imager, and Luminex). Two recombinant human cytokines with defined biological activity (NIBSC 86/552 and 86/680) were spiked into a human vaginal fluid simulant at both pH 4.5 and 7.2, and compared to spiked human blood serum (AB type), PBS, and saline. Twelve participating laboratories assayed unspiked samples of each matrix type as well as the cytokine-spiked samples. In addition, known amounts of assay calibrators were spiked into the same matrices and tested in two assay runs by each laboratory.

Results: The MSD and Biosource assays showed no matrix or pH effects on the recovery of IL-1b from the vaginal fluid (p< 0.001). The Endogen-Pierce and Luminex assays showed a significant difference between the recoveries of the two recombinant cytokines from the vaginal fluid matrix as compared to PBS (p<0.01). A significant effect of vaginal fluid pH on IL-1b recovery was observed in Luminex/Upstate (p<0.001) for both NIBSC and kit calibrator standards and in the QuantiGlo (RDS) assay but only for the 86/680 standard (p<0.001). Overall, the recovery of IL-1b was lower in serum as compared to other matrices (p<0.05) and most significantly reduced in the Luminex/Upstate assay. With the exception of Luminex (p<0.01), there were no statistically significant differences between PBS and saline, both used for cervicovaginal lavages, across assay platforms.

Conclusion: Our investigation presents the first standardized analytic approach for recovery of human IL-1b from a biological matrix simulating human vaginal fluid and thus provides a technological advance for the development of vaginal microbicides.