Abstract 73: Evaluation of bead-based assays of inflammation markers for epidemiologic studies.

Given finite sample availability in epidemiologic studies, minimizing assay volumes is an important consideration in studies of biomarker-disease relationships. Consequently, there is great interest in multiplex assays that use less volume than standard assays and provide data on multiple markers in one assay. This is particularly true for markers of inflammation, a key factor in cancer and other age-related diseases. Therefore, we conducted pilot studies of three bead-based, multiplex, inflammatory marker panels at three laboratories. The first pilot evaluated assay reproducibility by calculating the coefficient of variation (CV) using 12 blinded replicate samples, and the second evaluated the role of delayed processing using samples from 12 participants who had one-third of their sample processed immediately after blood draw, and one-third each processed 24 and 48 hours later. Minimum criteria for an analyte to pass the pilots were: (1) >50% of samples had levels above the limit of detection (LOD), (2) a CV ≤20%, and (3) an intraclass correlation across processing times >0.80. For Panel 1, 200uL of plasma was used to assess 10 biomarkers in 2007. For Panel 2, 50uL of plasma was used to assess 17 markers in 2011. For Panel 3, 250uL of plasma was used to assess 12 markers in 2012. More than half of the biomarkers in two of the panels (1 and 3) had >50% of samples with levels less than the LOD; in Panel 2 29% of biomarkers had >50% of samples below the LOD. Of the biomarkers with more than half the samples above the LOD, all had CVs <20%. Overall, 3 (30%) of markers passed the LOD and CV criteria in Panel 1, 12 (71%) passed in Panel 2, and 4 (33%) passed in Panel 3. Among these, none of the markers in Panels 1 and 3 and 7 markers in Panel 2 were stable with delayed processing, resulting in an overall pass rate of 0% in Panels 1 and 3 and 41% in Panel 2. Further, for Panel 2, we observed that the correlation between the multiplex and the standard ELISA for IL-6 was only 0.25. Our results suggest that assessing inflammatory markers on bead-based multiplex platforms may not be appropriate for epidemiologic studies, possibly reflecting poor assay optimization for specific biomarkers and thus lowered overall assay sensitivity and specificity.

Citation Format: Elizabeth M. Poole, Mary K. Townsend, A Heather Eliassen, Janine Neville-Golden, Shelley S. Tworoger. Evaluation of bead-based assays of inflammation markers for epidemiologic studies. [abstract]. In: Proceedings of the AACR Special Conference on Post-GWAS Horizons in Molecular Epidemiology: Digging Deeper into the Environment; 2012 Nov 11-14; Hollywood, FL. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2012;21(11 Suppl):Abstract nr 73.