Ideally, the combination of clinical imaging techniques with additive manufacturing processes enables the fabrication of patient-specific regenerative implants that precisely fit into the defect site, promoting native tissue restoration while gradually degrading. Osteochondral defects, affecting both cartilage and subchondral bone in joints are best visualized using magnetic resonance imaging (MRI). In this study, a workflow for computer-aided manufacturing of patient-specific osteochondral implants based on geometrical data obtained from MRI scans was evaluated in a clinically relevant setting. Artificial osteochondral defects were created in femoral condyles of human body donors and scanned with MRI. ‘Computer-Aided Design’ (CAD) models for bone and cartilage components served as basis for designing defect-specific trizonal implants consisting of (i) a bone, (ii) an interlocking, and (iii) a cartilage zone. These implants were fabricated using multi-channel 3D extrusion printing, using a calcium phosphate cement as a bone substitute and an alginate-based hydrogel as a cartilage substitute material - with both materials alternately printed in the interlocking zone. After fabrication, the constructs were implanted into the corresponding defects, and assessed for fit accuracy via clinical imaging. The entire process chain was successfully conducted under near-clinical conditions by an interdisciplinary team of engineers, radiologist and surgeons, during which critical points were identified. Due to the inherent resolution limitations of clinical MRI and extrusion-based 3D printing, inaccuracies in implant fitting occurred; strategies to address these challenges were identified by integrating design tolerances and applying minor intraoperative adjustments.