The sensitivity of the MDA-kb2 cell in vitro assay in detecting anti-androgenic chemicals--identification of sources of variability and estimation of statistical power.

In vitro assays for anti-androgens have been developed as screening tools for the identification of androgen receptor (AR) antagonists. We explored the usefulness of such assays for experimental purposes that require quantitation of effects in a highly reproducible manner, such as multi-component mixture experiments or evaluation of extracts of complex environmental samples. We have investigated sources of experimental variation in the MDA-kb2 assay for AR-antagonists. By omitting phenol red from culture media, avoiding media changes and extending the period allowed for cell attachment, the dynamic range increased. Variations in luminescence readings decreased, with smaller coefficients of variation within- and between-experiments. Normalisation of luminescence values to positive controls improved experiment-to-experiment reproducibility and allowed pooling of data from independent experiments. We also performed statistical power analyses to determine the minimal suppression of androgenic (DHT) effects by test agents that are detectable as statistically significantly different from positive controls (so-called minimum significant differences, MSD). Using the modified assay protocol extensive concentration-response analyses were conducted with bisphenol A, BDE100 and vinclozolin. Our modified procedure improves considerably the reproducibility of the MDA-kb2 assay.