Cross Laboratory Validation of An In Vitro Screen for Cytokine Release Syndrome

Abstract 2778

Monoclonal antibodies (mAb) are widely used in anti-inflammatory and tumor therapy. They are highly effective in certain diseases, but can cause a range of adverse effects. Toxicity may result from the expected pharmacological effects of the antibody and from interactions with antigen expressed on tissues other than the intended target. The results range from mild infusion reactions to cytokine release syndrome (CRS). A number of mAbs are associated with CRS on the first dose in humans (including: anti-CD3, anti-CD52; anti-CD20, anti-CD25 and anti-CD28 superagonist) that were not predicted by preclinical studies in animals. To address this issue, we have developed an in vitro screen using cultured human whole blood that assess the potential for mAb to cause CRS. The purpose of these experiments was to determine if the assay can be transferred to different laboratory and conducted using a different analytical platform and different donors. mAbs are bound to Protein A coated polystyrene beads (Spherotech) and incubated with whole blood from 6 normal human volunteers diluted in media (1:10) in 96 well plates. PBS (no beads) and Autologous Plasma are negative controls, while LPS (1 ug/mL); LPS and superagonist anti-CD28 (ANC28.1/5D10, Ancell) serve the positive controls. After incubation for 48 hr, the supernatants are collected and cytokines (IFN-γ, TNFα, IL-1, IL-2, IL-6, IL-8, IL-10 and IL-12) assayed using SearchLight® (Aushon Biosystems) or MSD® (Meso Scale Discovery) platforms. The response to superagonist anti-CD28 was distinctly different from the response to LPS, anti-CD20 and anti-CD52. Although the absolute values for the cytokines differed across the analysis platforms, Spearman Rank Order analysis showed a statistically significant correlation for most cytokines. Moreover, when the average values were normalized to the response to LPS, hierarchical cluster analysis classified the responses measured by the two platforms in an identical manner. In conclusion, this study has shown that an in vitro whole blood based assay can classify mAb according to the potential to cause CRS in humans and suggest that the assay can be transferred successfully across laboratories and analysis platforms.