On November 2015, Samarco tailings dam in Mariana, Minas Gerais, Brazil, collapsed, releasing 62 million tons of tailings that advanced through 668 km of the Doce River and adjacent floodplain. Although the collapse was the worst environmental disaster in Brazil, little is known about its consequences to aquatic biota. Here we evaluate the effects of the tailings mudflow on metal and As concentration in fish and how concentration correlates with water and fish characteristics. We quantified semitotal amounts of Ag, Al, As, Cd, Cr, Cu, Fe, Hg, Mn, Ni, Pb, and Zn in fish muscle tissue using inductively coupled plasma mass spectrometry (ICP‐MS) in 255 individuals (34 species) sampled in unaffected and affected areas along the Doce River basin. Arsenic and Hg were higher in fish from affected sites, likely due to turbulent mixing of previously sedimented material by the giant tailings wave. Silver and Zn concentrations were higher in unaffected sites. Arsenic concentration in Geophagus brasiliensis decreased with increasing fish weight. Copper and Zn decreased with increasing fish weight considering the whole assembly of fish. The tailings mudflow increased water conductivity, and conductivity increased Al concentration in fish, so we expected a larger Al concentration in fishes from affected sites. However, the observed Al concentration in fishes from affected sites was lower than expected by water conductivity. Thus, the tailings mudflow reduced Al uptake or accumulation in fishes. Mercury decreased with increasing water conductivity in both unaffected and affected sites considering all species and in G. brasiliensis alone. Despite the relatively low concentration range of metals and As found in fish, fishes from sites affected by the Fe ore tailings mudflow showed higher As and Hg concentration, compared to fishes from unaffected sites. The higher As and Hg in affected sites require further detailed monitoring to ensure safeguards of human health by fishing activity along the Doce River. Integr Environ Assess Manag 2020;16:622–630. © 2020 SETAC