No abstract

Mesenchymal stromal cell (MSC)‐derived small extracellular vesicles (sEVs) show therapeutic potential in multiple disease models, including kidney injury. Clinical translation of sEVs requires further preclinical and regulatory developments, including elucidation of the biodistribution and mode of action (MoA). Biodistribution can be determined using labelled sEVs in animal models which come with ethical concerns, are time‐consuming and expensive, and may not well represent human physiology. We hypothesised that, based on developments in microfluidics and human organoid technology, in vitro multi‐organ‐on‐a‐chip (MOC) models allow us to study effects of sEVs in modelled human organs like kidney and liver in a semi‐systemic manner. Human kidney‐ and liver organoids combined by microfluidic channels maintained physiological functions, and a kidney injury model was established using hydrogenperoxide. MSC‐sEVs were isolated, and their size, density and potential contamination were analysed. These sEVs stimulated recovery of the renal epithelium after injury. Microscopic analysis shows increased accumulation of PKH67‐labelled sEVs not only in injured kidney cells, but also in the unharmed liver organoids, compared to healthy control conditions. In conclusion, this new MOC model recapitulates therapeutic efficacy and biodistribution of MSC‐sEVs as observed in animal models. Its human background allows for in‐depth analysis of the MoA and identification of potential side effects.

As kidney diseases affect ∼10% of the world population, understanding the underlying mechanisms and developing therapeutic interventions are of high importance. Although animal models have enhanced knowledge of disease mechanisms, human (patho-)physiology may not be adequately represented in animals. Developments in microfluidics and renal cell biology have enabled the development of dynamic models to study renal (patho-)physiology in vitro. Allowing inclusion of human cells and combining different organ models, such as kidney-on-a-chip (KoC) models, enable the refinement and reduction of animal experiments. We systematically reviewed the methodological quality, applicability and effectiveness of kidney-based (multi-)organ-on-a-chip models, and describe the state-of-the-art, strengths and limitations, and opportunities regarding basic research and implementation of these models. We conclude that KoC models have evolved to complex models capable of mimicking systemic (patho-)physiological processes. Commercial chips and human induced pluripotent stem cells and organoids are important for KoC models to study disease mechanisms and assess drug effects, even in a personalized manner. This contributes to the Reduction, Refinement and Replacement of animal models for kidney research. A lack of reporting of intra- and inter-laboratory reproducibility and translational capacity currently hampers implementation of these models.