The use of residual streams from agricultural production and food consumption containing animal proteins entails the risk of disease transmission as illustrated by the epidemics of bovine spongiform encephalopathy (BSE) and African swine fever. To combat this risk, the use of animal proteins in livestock feed was banned in the European Union, resulting in a drain of valuable proteins from the agricultural system. With an increasing call for a circular food system, the use of residual streams as a feed ingredient needs to be reconsidered with the associated disease risks being assessed and mitigated where needed. In this study, we assessed the BSE risk of bovine spray-dried red blood cells (SDRBC) as an ingredient of aquafeed. Fish fed with bovine SDRBC could indirectly result in exposure of ruminants to BSE infectivity because one of the exemptions of the feed ban is the use of fishmeal as an ingredient in calf milk replacers. A quantitative risk model was built to evaluate the BSE infectivity present in blood sourced from a slaughtered BSE-infected cow and the reduction of infectivity due to processing steps along the production chain. The end point of the model was the BSE infectivity, expressed in cattle oral ID50 (CoID50), reaching calves fed calf milk replacer containing fishmeal, and the corresponding probability that this will result in at least one new BSE infection.The expected BSE infectivity in blood from a BSE-infected cow at the clinical end state of infection is 0.75 CoID50 (median value). Infectivity in blood mainly results from cross-contamination with brain tissue during stunning at the slaughterhouse. The initial infectivity is reduced along the pathway from slaughtered cow to calf milk replacer, with the highest reduction achieved by clearance of infectivity by fish fed bovine SDRBC as an ingredient of aquafeed, although this parameter has high uncertainty. The final infectivity reaching calves via inclusion of fishmeal in calf milk replacer is estimated to be very low (median value: 1.1 × 10−5 CoID50). Assuming an exponential dose-response model, this corresponds with an expected probability that < 10 out of a million slaughtered BSE-infected cows will result in new BSE infections, which is far below the threshold value of 1 for the basic reproduction number (R0) to initiate a new epidemic. We thus conclude that it is very unlikely that the use of bovine SDRBC as ingredient of aquafeed will result in a new BSE epidemic in cattle. What-if analysis indicated that this conclusion is robust, despite high uncertainty for some input parameters.