Three dimensional and microphysiological bone marrow models detect in vivo positive compounds.

Micronucleus (MN) assessment is a valuable tool in safety assessment. However, several compounds are positive in the in vivo bone marrow (BM) MN assay but negative in vitro, reflecting that BM complexity is not recapitulated in vitro. Importantly, these compounds are not genotoxic; rather, drug-driven pharmacological-effects on the BM increase MN, however, without mechanistic understanding, in vivo positives stop drug-progression. Thus, physiologically-relevant BM models are required to bridge the gap between in vitro and in vivo. The current study aimed to investigate the utility of two human 3D BM models (fluidic and static) for MN assessment. MN induction following treatment with etoposide and Poly-ADP Ribose Polymerase inhibitor (PARPi) and prednisolone (negative in vitro, positive in vivo) was determined in 2D L5178Y and human BM cells, and the 3D BM models. Etoposide (0-0.070 µM) and PARPi (0-150 µM) induced MN in both 3D BM models indicating their utility for genotoxicity testing. Interestingly, PARPi treatment induced a MN trend in 3D more comparable to in vivo. Importantly, prednisolone (0-1.7 mM) induced MN in both 3D BM models, suggesting recapitulation of the in vivo microenvironment. These models could provide a valuable tool to follow up, and eventually predict, suspected pharmacological mechanisms, thereby reducing animal studies.