Megakaryocytopoiesis gives rise to platelets by proliferation and differentiation of lineage-specific progenitors, identified in vitro as Colony Forming Unit-Megakaryocytes (CFU-Mk). The aim of this study was to refine and optimize the in vitro Standard Operating Procedure (SOP) of the CFU-Mk assay for detecting drug-induced thrombocytopenia and to prevalidate a model for predicting the acute exposure levels that cause maximum tolerated decreases in the platelets count, based on the correlation with the maximal plasma concentrations (C max) in vivo. The assay was linear under the SOP conditions, and the in vitro endpoints (percentage of colonies growing) were reproducible within and across laboratories. The protocol performance phase was carried out testing 10 drugs (selected on the base of their recognised or potential in vivo haematotoxicity, according to the literature). Results showed that a relationship can be established between the maximal concentration in plasma (C max) and the in vitro concentrations that inhibited the 10-50-90 percent of colonies growth (ICs). When C max is lower than IC10, it is possible to predict that the chemicals have no direct toxicity effect on CFU-Mk and could not induce thrombocytopenia due to bone marrow damage. When the C max is higher than IC90 and/or IC50, thrombocytopenia can occur due to direct toxicity of chemicals on CFU-Mk progenitors.