Osteosarcoma represents one of the most aggressive tumors of bone among adolescents and young adults. Despite improvements in treatment, osteosarcoma has a grave prognosis. The identification of prognostic factors is still in its infancy. Weighted gene correlation network analysis (WGCNA) was conducted on mRNA-sequencing and clinical information (gender, survival and metastasis) of osteosarcoma patients from the TARGET database to obtain genes in modules associated with metastasis of osteosarcoma. The Cox regression analysis was then performed on the gene expression profile from TARGET to screen genes associated with patients’ survival. Known genes related to osteosarcoma were obtained by intersecting osteosarcoma-related genes from DisGeNET and DiGSeE, followed by the construction of PPI network of osteosarcoma-related genes and survival-related genes in modules. The screened key genes were subject to multi-factor Cox proportional hazards model, and osteosarcoma patients were classified into high- and low- risk groups according to the risk score to evaluate the potential of key genes to predict the survival of osteosarcoma patients. The WGCNA showed that 4 genes in tan and 19 genes in pink modules were related to the survival of osteosarcoma patients. Osteosarcoma-related known genes (9) were obtained in intersection of DisGeNET and DiGSeE. PPI network identified 4 key genes (KRT5, HIPK2, MAP3K5 and CD5) closely associated with survival of osteosarcoma patients. HIPK2, MAP3K5 and CD5 expression was inversely correlated with survival risk, while KRT5 expression was positively correlated with survival risk. These results show KRT5, HIPK2, MAP3K5 and CD5 serve as prognostic factors of osteosarcoma patients.

骨肉瘤代表了青少年和年轻人中最具侵略性的骨肿瘤之一。尽管治疗有所改善，但骨肉瘤的预后却很严重。预后因素的鉴定仍处于起步阶段。从TARGET数据库对骨肉瘤患者的mRNA测序和临床信息（性别，存活和转移）进行了加权基因相关网络分析（WGCNA），以获取与骨肉瘤转移相关的模块中的基因。然后对TARGET的基因表达谱进行Cox回归分析，以筛选与患者生存相关的基因。与骨肉瘤相关的已知基因是通过与DisGeNET和DiGSeE的骨肉瘤相关基因交叉而获得的，其次是在模块中构建骨肉瘤相关基因和生存相关基因的PPI网络。对筛选出的关键基因进行多因素Cox比例风险模型分析，根据风险评分将骨肉瘤患者分为高危和低危组，以评估关键基因预测骨肉瘤患者生存的潜力。WGCNA显示，棕褐色中的4个基因和粉红色模块中的19个基因与骨肉瘤患者的生存有关。在DisGeNET和DiGSeE的交叉点获得了与骨肉瘤相关的已知基因（9）。PPI网络确定了4个与骨肉瘤患者生存密切相关的关键基因（KRT5，HIPK2，MAP3K5和CD5）。HIPK2，MAP3K5和CD5的表达与生存风险呈负相关，而KRT5的表达与生存风险呈正相关。这些结果表明，KRT5，HIPK2，MAP3K5和CD5是骨肉瘤患者的预后因素。