In vitro metal ion bioaccessibility, as a measure of bioavailability, can be used to read-across toxicity information from data-rich, source substances to data-poor, target substances. To meet the data requirements for oral systemic toxicity endpoints under the REACH Regulation in Europe, 12 nickel substances underwent bioaccessibility testing in stomach and intestinal fluids. A read-across paradigm was developed based on the correlation between gastric bioaccessibility and in vivo acute oral toxicity. The oral LD₅₀ values were well predicted by nickel release (R² = 0.91). Samples releasing <48% available nickel (mg Ni released/mg available Ni × 100) are predicted to have an LD₅₀ > 2000 mg/kg; while samples releasing >76% available nickel are expected to have an LD₅₀ between 300 and 2000 mg/kg. The hazard classifications (European Regulation on Classification, Labelling and Packaging of Chemical Substances and Mixtures) for all oral systemic endpoints were evaluated based on read-across from three source nickel compounds (sulfate, subsulfide, oxide). Samples releasing <48% available nickel were read-across from nickel oxides and subsulfide. Samples releasing >76% Ni were read-across from nickel sulfate. This assessment suggests that nickel chloride and dihydroxide should be less stringently classified and nickel sulfamate should receive a more stringent classification for oral systemic endpoints than currently assigned.