Cast steel components are affected by manufacturing process-based imperfections, which alter the local stress state leading to significantly reduced fatigue life. Due to the arbitrary spatial shape of intrinsic defects, the application of local fatigue assessment concepts is encouraged to ensure safe operation of defect-afflicted components. This research work focusses on the investigation of the fatigue life limiting effect of casting defects. Based on extensive experimental work, the material behavior of the cast steel alloys G12MnMo7-4+QT and G21Mn5+N is characterized utilizing plain and notched small-scale specimens made from near defect-free bulk material. Additional high-cycle fatigue tests of bulk and surface defect-afflicted large-scale samples provide experimental assertions on the adverse effect of immanent imperfections. In order to quantify the influence of the defect geometry, the investigated imperfections are geometrically characterized by non-destructive testing methods and fracture surface analyses. The obtained results serve as basis for the fatigue analysis of defect-afflicted specimens. Based on the theory of critical distances and the strain energy density concept, a statistical procedure is presented which facilitates the fatigue assessment in the finite- and long-life fatigue regime, considering the local defect size, shape and orientation as well as both size and mean stress effect. The devised numerical framework bases on planar defect projections in the form of radiographs, emphasizing the application-oriented focus of the present work. In order to investigate the effect of local plastic deformations, a novel approximation framework is introduced, facilitating the evaluation of the elastic-plastic strain energy density based on linear-elastic finite element analyses. The resulting holistic fatigue assessment concept merges the experimental results of plain and notched small- as well as bulk and surface defect-afflicted large-scale specimens into a unique narrow scatter band. The novel elastic-plastic concept possesses enhanced prediction accuracy compared to linear-elastic approaches, which is attributed to the consideration of cyclic mean stress relaxation and the resulting formation of a locally effective stress ratio. As the established fatigue assessment framework solely bases on planar defect projections and linear-elastic finite element analyses, numerical efficiency is significantly enhanced, confirming the holistic concept as engineering-feasible approach to facilitate the comprehensive assessment of defect-afflicted cast steel components.