Analysis of cytokine release assay data using machine learning approaches

The possible onset of Cytokine Release Syndrome (CRS) is an important consideration in the development of monoclonal antibody (mAb) therapeutics. In this study, several machine learning approaches are used to analyze CRS data. The analyzed data come from a human blood in vitro assay which was used to assess the potential of mAb-based therapeutics to produce cytokine release similar to that induced by Anti-CD28 superagonistic (Anti-CD28 SA) mAbs. The data contain 7 mAbs and two negative controls, a total of 423 samples coming from 44 donors. Three (3) machine learning approaches were applied in combination to observations obtained from that assay, namely (i) Hierarchical Cluster Analysis (HCA); (ii) Principal Component Analysis (PCA) followed by K-means clustering; and (iii) Decision Tree Classification (DTC). All three approaches were able to identify the treatment that caused the most severe cytokine response. HCA was able to provide information about the expected number of clusters in the data. PCA coupled with K-means clustering allowed classification of treatments sample by sample, and visualizing clusters of treatments. DTC models showed the relative importance of various cytokines such as IFN-γ, TNF-α and IL-10 to CRS. The use of these approaches in tandem provides better selection of parameters for one method based on outcomes from another, and an overall improved analysis of the data through complementary approaches. Moreover, the DTC analysis showed in addition that IL-17 may be correlated with CRS reactions, although this correlation has not yet been corroborated in the literature.