Osteosarcoma is the most common malignant tumor in children and young adults. Despite the use of surgery and multi-drug chemotherapy, the survival rate of osteosarcoma patients has not improved in the past three decades. Personalized medicine holds the promise of improving the treatment of cancer patients; however, it relies on an accurate way to predict clinical risks before treatment initiation. In osteosarcoma, the only prognostic factor at the time of diagnosis is the detection of metastasis by conventional low-sensitive imaging methods. We reason that the development of a molecular prognostic biomarker will facilitate personalized medicine, so that an alternative or a targeted therapy can be used to improve the outcome of the high-risk patients, while the low-risk patients would be spared from unnecessary chemotherapy. A vast amount of literature has shown that the tumor suppressor p27 can be used to predict prognosis in various adult cancers, such as breast carcinomas, acute myelogenous leukemia, pancreatic cancer and ovarian carcinomas. Our laboratory has demonstrated that p27 is mislocalized in metastatic osteosarcoma cell lines and frequently in human osteosarcoma cases. Functional studies have further shown that cytoplasmic p27 can promote the motility and invasiveness of osteosarcoma cells. In addition, we have demonstrated that the p27 autoantibody is elevated in metastatic osteosarcoma patients using high-density protein arrays. Based on these observations, we hypothesize that p27, which is a prognostic biomarker in adult cancers, can be used to prognosticate pediatric osteosarcoma. To test this hypothesis, we propose four Specific Aims. First, we will correlate the protein expression and subcellular localization of p27 with metastasis and survival information in two series of osteosarcoma cases. Second, we will integrate the p27 results with other related prognostic biomarkers to test if we can increase the prognostic significance of the detection. Third, we will develop a novel Luminex-based assay for the p27 autoantibody to validate if it correlates with the outcomes of osteosarcoma patients using pre-treatment serum samples collected from the Children's Oncology Group. Lastly, we will perform genomic and targeted proteomic analyses of a panel of p27 mutants that have been generated in our laboratory to identify novel p27-related biomarkers. These candidate biomarkers will be validated using the data generated from our SPECS and TARGET consortia. All the biomarker results will be compared with the existing prognostic factor, i.e. metastasis, to test the clinicalutility of the biomarkers. The long-term goal of this study is to develop a biomarker approach for the prognostication of osteosarcoma patients at the time of diagnosis that is more accurate than the conventional approaches. It would facilitate the development of personalized medicine and, hence, improve the survival of osteosarcoma patients. Completion of the proposed study is a critical step to achieve this goal. Furthermore, since p27 has been implicated in other cancers, the results obtained from this study will have a larger impact on the field of cancer biomarkers.