After en bloc resection of primary pelvic bone tumours, orthopaedic oncologists usually try to perform reconstruction in order to achieve limb salvage. Reconstruction after these types of resections is challenging. Especially when the total ilium and parts of the sacrum needs to be removed. One of the reconstructive options is endoprosthetic fixation to sacral bone with the LUMiC prosthesis, but using the LUMiC prosthesis in the sacrum resulted in poor clinical results. Emerging new production techniques such as 3D printing provide for customised prostheses that could be an adequate reconstructive option in these cases and could complement the LUMiC prosthesis to enable sacral anchoring of the LUMiC prosthesis. One of the challenges of customised prostheses is that they are not ‘off the shelf’ and therefore are all different in terms of geometry and, thus, performance. Therefore, these implants cannot be subjected to stringent regulations and registrations used in conventional arthroplasty. In order to ensure maximal safety of these implants we must perform biomechanical testing before implantation. Unfortunately, the bone models for biomechanical testing currently available are not validated on implant fixation and thus unreliable for prosthesis testing. Therefore, in this thesis a new prosthesis that enables sacral anchoring of the cup of the LUMiC prosthesis is designed and two strategies are developed in order to mechanically evaluate this newly designed prosthesis for the sacral anchoring of the LUMiC cup. First strategy is using Finite element modelling to develop a model of the sacrum with implanted prosthesis. Second strategy is the development of 3D bone models of the sacrum using 3D printing together with performing experiments to mechanically evaluate the newly designed prosthesis. Four different mechanical bone models with different shell thickness and infill density were developed and compared to the FE model. Results showed that the new prosthesis design performed better in terms of micromotions between implant and bone compared to the use of the old LUMiC stem in the FE model. Furthermore, a 3D printed mechanical bone model with shell thickness of 1mm and infill structure of 20% showed most comparable results in terms of micromotions between implant and bone compared to the developed FE model. There could be concluded that the FE model and the development of 3D printed mechanical bone models together with the experiments are good starting points in the development of a protocol to in vitro evaluate newly designed 3D printed patient specific prosthesis for the sacrum after tumour resection. The mechanical bone model with shell thickness 1mm and infill structure 20% is the best option in terms of mimicking real bone. Furthermore, the newly designed prosthesis for sacral anchoring of the LUMiC cup showed promising results and could be considered as a improved alternative for the old LUMiC prosthesis stem. In future research cadaver tests should be performed to further validate the found results.