The fatigue behavior and FEA analysis of different ceramic materials cemented over distinct substrates for implant-supported crowns were evaluated in this study. Discs of 10 mm in diameter of both restorative and substrate materials were made and randomly allocated into pairs (n = 15) considering the two study factors: ‘restorative ceramic material’ (1 mm thickness) – polymer-infiltrated ceramic network (PICN), lithium disilicate (LD), zirconia-reinforced lithium silicate (ZLS), or translucent zirconia (TZ); and ‘foundation substrate’ (2 mm thickness) – polyetheretherketone (Peek) or yttrium-stabilized zirconia (YZ). Adhesive cementation was made with a dual cure resin cement. Fatigue testing was run using the step-stress methodology: initial load of 200 N for 5000 cycles, followed by steps of 10,000 cycles starting at 400 N up to 2800 N or until failure, step size of 200 N, frequency of 20 Hz. Data were analyzed by the Kaplan Meier and log-rank post-hoc tests. Fractography analysis (stereomicroscope and SEM) and FEA were also performed. Both factors under study and their interaction statistically influenced the fatigue failure load (FFL), cycles for failure (CFF) and survival rates (p < 0.001). The restorative materials bonded to YZ had higher FFL and CFF than when adhering to Peek, while restorative materials with more crystalline content (TZ and ZLS) showed higher FFL and CFF than LD and PICN. The fractography analysis showed that all materials bonded to YZ resulted in failures starting at the occlusal surface (Hertzian cone cracks), while materials bonded to Peek had radial cracks from the ceramic-cement intaglio surface. FEA analysis showed that tensile stress concentration decreased in the intaglio surface when testing the restorative material over a stiffer (YZ) foundation substrate. In addition, the higher the restorative material's crystalline content, the more the stress is concentrated within the material (TZ > ZLS ≥ LD > PICN) when bonded to the same foundation substrate. Thus, it concluded that a stiffer foundation substrate (YZ) enhances the load-bearing capacity under fatigue of the restorative set; that restorative materials with higher crystalline content results in higher fatigue performance of the set, regardless of the foundation used; and that the foundation material influences the failure pattern of the restorative set. •Stiffer foundation substrate enhances the fatigue performance of a restorative set.•Restorative materials with higher crystalline content show higher fatigue strength.•The foundation substrate influences the failure pattern of a restorative set.