Tear fluid biomarkers: a comparison of tear fluid retrieval and storage methods

Purpose: Increased research using multiplex technology in the field of ocular surface disease (OSD) has shown its potential in identifying novel biomarkers. These studies, however, have also highlighted differences in sampling across laboratories. This preliminary study compared different tear fluid retrieval and storage methods versus cytokine expression, using magnetic multiplex bead arrays.Methods: Pooled tear samples underwent microarray bead analysis for 7 cytokines (IL1-ß, -2, -6, -8, -17, IFN-¿ and TNF-a) using various tear sampling and storage techniques. A standard method of using 1µl tears in a 50x dilution was used throughout the study, except when looking at the effect of different tear volumes on cytokine detection.Results: Multiple freeze/thaw cycles significantly reduced levels of IL1-ß and -2 (both p<0.05) and showed a trend for a reduction in the other 5 cytokines. No significant deterioration of cytokines was noted for samples stored on ice for 5hrs, however storage for 5hrs at room temperature revealed a significant reduction in all 7 biomarkers (p<0.05). Levels of IL1-ß and TNF-a were significantly diminished for tears stored in standard eppendorf tubes, versus Lo-Bind eppendorfs (p<0.05). Moreover, IL-2, -6, and -17 were significantly reduced in tears collected with a Schirmer strip versus a glass microcapillary (p<0.05). A significant reduction in IL-8 and IFN-g levels was noted in tears retrieved using a minisponge versus a glass microcap. No significant difference was observed for tears stored at -20°C versus -80°C, although there was a trend for slight reductions in cytokine levels at -20°C. There was a general trend for lower cytokine levels as the tear sample size was increased (1µl versus 3 and 5 µl), suggesting a possible matrix effect. This reduction was found to be significant when comparing 1 and 5µl volumes for IL1-ß and 17 (p<0.05).Conclusions: Differences in tear retrieval and storage methods may affect cytokine detection by multiplex bead arrays. This may be due to sample dilution and inherent matrix effects, varying degrees of protein binding affinities of lab plastics used for storage, and sample sublimation. A standardised procedure for tear fluid retrieval and storage would benefit future OSD cytokine analyses. Further work will be carried out to extend and confirm the present results.