In silico approaches for enhancing retrieval analysis as a source for discovery of implant reactivity‐related mechanisms and biomarkers

The ability to characterize implant debris in conjunction with corresponding immune and tissue-destructive responses renders retrieval analysis as an important tool for evaluating orthopedic devices. We applied advanced analytics and in silico approaches to illustrate the retrieval-based potential to elucidate host responses and enable discovery of corresponding biomarkers indicative of in vivo implant performance. Hip retrieval analysis was performed using variables based on immunostaining, polarized microscopy, and fretting-corrosion and oxidation analyses. Statistical analyses were performed in R. Hierarchical/k-means clustering and principal component analysis were used for data analysis and visualization. Correlation Engine (CE) and Ingenuity Pathway Analysis (IPA) were employed for in silico corroboration of putative biomarkers. Higher giant cell and histiocyte scores and positivity for CD68 and CD3 indicating infiltration with macrophages and T-cells, respectively, were detected mainly among older generation hips with higher ultra-high-molecular-weight-polyethylene loads. Our in silico analysis using pre-existing data on wear particle-induced loosening substantiated the role of CD68 in implant-induced innate responses and identified the CD68-related molecular signature that can be indicative of development of aseptic loosening and can be further corroborated for diagnostic/prognostic testing in clinical setting. Thus, this study confirmed the great potential of advanced analytics and in silico approaches for enhancing retrieval analysis applications to discovery of new biomarkers for optimizing implant-related preclinical testing and clinical management. (c) 2019 Wiley Periodicals, Inc. J Biomed Mater Res Part B: Appl Biomater 108B:263-271, 2020.