Keloid disorder is an abnormal fibroproliferative reaction that can occur on any area of skin, and it can impair the quality of life of affected individuals. To investigate the pathogenesis and develop a treatment strategy, a preclinical animal model of keloid disorder is needed. However, keloid disorder is unique to humans, and the development of an animal model of keloid disorder is highly problematic. We developed the patient-derived keloid xenograft (PDKX), which is a humanized mouse model, and compared it to the traditional mouse xenograft model (transplantation of only keloid lesions). To establish the PDKX model, peripheral mononuclear cells (PBMCs) from ten keloid patients or five healthy control subjects were injected into NOD/SCID/IL-2Rγnull mice, and their keloid lesions were grafted onto the back after the engraftment of immune cells (transplantation of keloid lesions and KP PBMCs or HC PBMCs). Four weeks after surgery, the grafted keloid lesion was subjected to histologic evaluation. Compared to the traditional model, neotissue formed along the margin of the grafted skin, and lymphocyte infiltration and collagen synthesis were significantly elevated in the PDKX model. The neotissue sites resembled the margin areas of keloids in several respects. In detail, the levels of human Th17 cells, IL-17, HIF-1a, and chemokines were significantly elevated in the neotissue of the PDKX model. Furthermore, the weight of the keloid lesion was increased significantly in the PDKX model, which was due to the proinflammatory microenvironment of the keloid lesion. We confirmed that our patient-derived keloid xenograft (PDKX) model mimicked keloid disorder by recapitulating the in vivo microenvironment. This model will contribute to the investigation of cellular mechanisms and therapeutic treatments for keloid disorders.