The impact of sample processing on inflammatory markers in serum: Lessons learned

Abstract Objectives: To investigate the effect of sample handling on inflammatory cytokines in serum and highlight challenges with using samples pre-collected from biobanks for biomarker research. Methods: Cytokine concentrations (IL-1β, IL-2, IL-6, IL-8, IL-10, TNFα, and IFNγ) were measured in serum samples of 205 patients with bipoldar disorder (BD) from the Mayo Clinic Bipolar Disorder Biobank and 205 non-psychiatric controls from the Mayo Clinic Biobank. As cytokine concentrations varied by recruitment site, post-hoc models were used to test the effect of clinical variables and pre-processing time on cytokines. To evaluate the effect of pre-processing time experimentally, cytokines were assayed in serum and plasma from 6 healthy volunteers processed at different time points. Results: Cytokine levels were significantly higher in the BD group. However, both cytokine levels and pre-processing times differed by recruitment site, and post-hoc analyses revealed that pre-processing time was significantly associated with several cytokines. An experiment using samples from healthy volunteers confirmed that concentrations for most cytokines increased with longer pre-processing times. Conclusions: Delays in processing influence cytokine concentrations in blood samples. Given the increasing use of biobanks in research, this study highlights the need to carefully evaluate sample collection and handling methods when designing biomarker studies.