Abstract Background With ever-increasing exposure to engineered nanomaterials (NMs), there is an urgent need to evaluate the probability of consequential adverse effects. The potential for NM translocation to distal organs is a realistic prospect, with the liver being one of the most important target organs. Traditional in vitro or ex vivo hepatic toxicology models are often limiting (i.e. short life-span, reduced metabolic activity, lacking important cell populations, etc.). In this study, we scrutinize a 3D human liver microtissue (MT) model (composed of primary hepatocytes and non-parenchymal cells). This unique experiment benefits from long-term (3 weeks) repeated very low exposure concentrations, as well as incorporation of recovery periods (up to 2 weeks), in an attempt to account for the liver’s recovery capacity in vivo. As a means of assessing the toxicological potential of NMs, cell cytotoxicity (cell membrane integrity and aspartate aminotransferase (AST) activity), pro/anti-inflammatory response and hepatic function were investigated. Results The data showed that 2 weeks of cell culture might be close to limits before subtle ageing effects start to overshadow low sub-lethal NM-induced cellular responses in this test system (adenylate kinase (AK) cytotoxicity assay). We showed that in vitro AST measurement are not suitable in a nanotoxicological context. Moreover, the cytokine analysis (IL6, IL8, IL10 and TNF-α) proved useful in highlighting recovery periods as being sufficient for allowing a reduction in the pro-inflammatory response. Next, low soluble NM-treated MT showed a concentration-dependent penetration of materials deep into the tissue. Conclusion In this study the advantages and pitfalls of the multi-cellular primary liver MT are discussed. Furthermore, we explore a number of important considerations for allowing more meaningful in vitro vs. in vivo comparisons in the field of hepatic nanotoxicology.