Keloids are aberrant dermal responses to the wound healing process, characterized by the proliferation of fibroblasts and blood vessels and overproduction of collagen. The most common sites are earlobes, the mandible, presternal regions, upper back and shoulders. Plantar keloids are rare; only a few cases have been reported. A transcription study of keloids on predilection sites revealed aberrant connective tissue differentiation towards cartilage and bone.1 However, little attention has been paid to the pathogenesis of keloids on non-predilection sites. Here, we describe a 45-year-old Chinese woman with a huge plantar keloid (Figure S1a). She denied previous trauma or family history of keloids. We performed enucleation excision at the first stage and fractional resection at subsequent stages. We used RNA-sequencing (RNA-seq) of seven continuous resected dermal tissues from the keloid excised at the first stage to reduce the possibility of intra-heterogeneity. Plantar control dermal samples were taken from 2 non-keloid patients. The study was approved via expedited review by Peking University First Hospital Ethics Committee. We identified 1473 upregulated and 1950 downregulated differentially expressed genes (DEGs) between keloid and non-keloid samples (fold change [FC] >2, false discovery rate [FDR] <0.05). Three of the top 5 upregulated genes with the highest FDR are correlated with cartilage, namely, epiphycan (EPYC), collagen type X alpha 1 chain (COL10A1) and collagen type XI alpha 1 chain (COL11A1). EPYC regulates fibrillogenesis by interacting with collagen fibrils and modifies chondrogenesis.2 COL10A1 is a marker of chondrocyte hypertrophy, whereas COL11A1 is mainly expressed by chondrocytes, mesenchymal stem cells (MSCs) and osteoblasts.2 Normal skin dermis features collagen I and III and lacks collagen X and XI. In cartilage tissue, collagen II is the principal component, whereas collagens III, VI, IX, X, XI, XII and XIV all contribute to the extracellular matrix (ECM).3 Collagen X resides at the interface of cartilage and bone, constituting calcified cartilage.3 Collagen XI is an integral part of cartilage fibrils and plays a key role in the core fibrillar network in cartilage development.3 These significantly upregulated genes in collagen deposition demonstrated that the ECM in keloids acquired some potential characteristics of cartilage tissue. To further explore the functions of the DEGs in keloid development, we explored gene annotation of upregulated DEGs by Metascape. Enriched Gene Ontology (GO) terms of skin development and extracellular structure organization were among the top enriched terms. The network of enriched terms showed a close relation among extracellular structure organization (q value = 2.95E-31), collagen fibril organization (q value = 1.29E-10), skeletal system development (q value = 3.39E-9) and ossification (q value = 3.39E-9). Previous transcription study of upper-back keloids revealed that many top upregulated DEGs were involved in bone/cartilage formation,1 which is consistent with our findings for plantar keloids. To determine core transcription factors (TFs) related to cell identities, we screened all TFs from the DEGs and focused on those capable of converting dermal fibroblasts into disease-modified fibroblasts. We found 45 upregulated and 70 downregulated TFs (FC > 3, FDR < 0.005). Upregulated TFs were enriched mainly in stromal cells, fibroblasts, myofibroblasts and osteoblasts (Figure S1b), representing a dysplasia of cutaneous connective tissue towards bone differentiation. We found positive correlation for most upregulated TFs (Figure S1c). A search of ChEA3, a TF enrichment database, revealed that ZNF469 and RUNX2 from our TF list were enriched in transformed fibroblasts and involved in skeletal system development, whereas AEBP1 was enriched in artery and involved in collagen fibril organization. Recently, a single-cell RNA-seq analysis of presternal keloids revealed that AEBP1 and ZNF469 were core TFs associated with fibroblast-to-myofibroblast phenotype transition.4 ZNF469 may function as a TF or extra-nuclear regulator factor for synthesis or organization of collagen fibres. AEBP1 is involved in fibroblast proliferation and MSC differentiation into collagen-producing cells.5 It could also promote polymerization of collagen I in vitro,5 which might explain the excessive collagen I accumulation in keloids. RUNX2 is essential for osteoblastic differentiation and skeletal morphogenesis; phenotype transition in keloids is potentially regulated by overexpression of RUNX2.1 Hence, ZNF469, AEBP1 and RUNX2 combined may act as core TFs in transforming dermal fibroblasts into bone/cartilage-like fibroblasts and promote keloid development. Immunohistochemistry revealed positive staining for ZNF469, RUNX2 and AEBP1 in dermal fibroblasts of plantar keloid tissue but negative staining in control dermis. Moreover, immunofluorescence showed co-localization of ZNF469 and RUNX2 as well as AEBP1 and RUNX2 in keloid tissue (Figure S1d), which suggests that these TFs may act collaboratively as a regulatory complex in the differentiation of MSCs or fibroblasts into keloid-modified osteoblasts or chondrocytes. Taken together, by analysing the transcription data from rare and multi-sliced plantar keloid tissue, we discovered similar core TFs that were previously reported in predilection sites, all suggesting potential bone/cartilage transformation in keloids. Thus, keloids at different sites may have similar pathogenesis. Three potential core TFs, namely ZNF469, AEBP1 and RUNX2, may be key regulators in the cell identity transition of keloid fibroblasts. How and to what extent these TFs function warrants further investigation. We thank the patients for giving us written permission for using their tissue and clinical information. We also thank National Key Research and Development Project for financially supporting the research (2019YFC0840700). None to declare. PL involved in study design, data acquisition, data analysis, data interpretation and manuscript drafting. RP and MZ involved in data collection and analysis. GZ involved in study design and manuscript review. HL involved in study design, manuscript review and study supervision. All authors have read and approved the final manuscript. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.