Purpose of Review <p>Preclinical models are vital to translational breast cancer research. This review compares cell-line-derived xenografts (CDX) and patient-derived xenografts (PDX), summarising methodological advances, strengths and weaknesses, and their applications in precision oncology.</p> Recent Findings <p>Since 2015, the field has shifted from CDXs toward PDXs. CDX publications peaked in 2019 before declining, whereas PDX use has grown rapidly because these models better preserve tumour heterogeneity and predict clinical response. New approaches, including orthotopic implantation, organoid co-culture, single-cell sequencing, humanised PDXs and artificial intelligence, are improving engraftment and characterisation across triple-negative, HER2-positive and hormone-receptor-positive breast cancers.</p> Summary <p>Evidence indicates that PDXs complement but often outperform CDXs by more faithfully replicating human tumours. Overcoming challenges in engrafting hormone-receptor-positive tumours, preserving human stroma and immune components, and integrating computational tools will further enhance xenograft models and accelerate the development of personalised therapies. Our review synthesises these trends to guide researchers in selecting appropriate models and highlights the growing role of PDXs in functional precision oncology.</p> Graphical Abstract <p>This graphical abstract was created by the authors and is original; no permissions were required.</p> <p></p>