Performance evaluation of microbead and ELISA assays for follicular G-CSF: a non-invasive biomarker of oocyte developmental competence for embryo implantation.

G-CSF in individual follicular fluids correlates with the potential of the corresponding embryo to result in a live birth after transfer in IVF. To evaluate the requirements for routine follicular fluid G-CSF quantification, we compared follicular fluid G-CSF measurements made with two multiplexed microbead assays purchased from Bio-Rad Laboratories and R&D Systems, and a commercial G-CSF ELISA (R&D Systems). Individual follicular fluids (n=139) associated with transferred embryos were analysed to determine cytokine profile and the fate of each transferred embryo was recorded. The effect of multiplexing as well as comparison of the respective performances of the microbead assay with a flow cytometry assay was explored. Multivariable logistic regression analysis was performed and receiver operating characteristic (ROC) analysis was used to determine the performance and sensitivity/specificity of each method for individual follicular fluids. Covariate factors known to influence IVF outcome such as age, serum oestradiol and embryo score were systematically integrated in each analysis. The quantification of follicular fluid G-CSF using microbead assay methodologies, but not ELISA, yielded results showing the utility of follicular fluid G-CSF as a biomarker predictive of a successful delivery (Au(roc): 0.77 [0.68-0.84] (p=0.003) and 0.75 [0.66-0.82] (p=0.004) for Bio-Rad and R&D Systems microbead assays respectively), whereas follicular fluid G-CSF values quantified by ELISA were not predictive (Au(roc):0.61 [0.52-0.70] p=0.84). Microbead assay and flow cytometry appeared similarly efficient for quantifying follicular fluid G-CSF and multiplex versus single-plex assays did not influence the reliability of quantification.