New therapies have significantly improved the management of melanoma but brain metastases continue to be a major component of treatment failure. The ability to identify those patients who are at highest risk of developing brain metastases is very challenging with current methods. Successful development of sensitive and specific prognostic methods for prediction of brain metastases risk in earlier stage melanoma patients would enable the development of more effective surveillance strategies and prospective trials for chemoprevention of brain metastasis development in these high-risk patients. Gene expression signatures have been shown to be prognostic in multiple cancer types and prior gene expression profiles (GEP) of metastases have been published for melanoma. However, the sensitivity and specificity as well as the positive and negative predictive value of these assays remains suboptimal and no specific molecular signature has been described for predicting risk of brain metastases. We analyzed multiple independent gene expression datasets and identified a list of genes that robustly predicts the development of brain metastases. To translate this finding to the clinic, we will optimize the most robust genes in formalin-fixed paraffin embedded tissue and integrate the findings with other relevant biomarkers of disease progression. Recent analysis of metastatic lesions from melanoma patients revealed that brain metastases had significantly higher levels of active phosphorylated AKT (pAKT) than extracranial metastases. We validated these findings by expressing activated AKT in our established melanoma mouse model and observed lung and brain metastases in ~80% of the mice. Based on these data, we hypothesize that the combination of an optimized gene expression signature along with other relevant clinical and pathologic variables, will identify those melanoma patients at highest risk for the development of brain metastases. We will use our rich resource of patient samples to test our hypothesis. While therapeutic intervention to prevent brain metastases in melanoma is our long-term goal, in order for this to be implemented most effectively it is crucial that we develop more accurate methods to identify those patients at highest risk for the development of metastatic disease before those metastases occur. The objective of this study is to generate a clinical/molecular predictor for risk of brain metastases in melanoma patients and to validate the sensitivity and specificity of an optimized clinical/molecular prognostic assay for the development of brain metastases in a retrospective study of Stage II/III melanoma patients. With the recent approval of adjuvant therapies that include stage III disease, we have an enormous opportunity to intervene when there is minimal or non-detectable disease and not only prevent melanoma recurrence but also improve overall survival. As these adjuvant therapies are not without risk, successful completion of the aims in this project will have significant translational impact by identifying those patients that are most likely to benefit from adjuvant therapy.

Public Health Relevance

Recently approved therapies have shown unprecedented success in the treatment of advanced stage melanoma but clinical management of brain metastases remains challenging and the prognosis for these patients is grim. This project seeks to identify those patients at highest risk for the development of brain metastases. This will enable initiation of future interventional studies to determine efficacy of these approaches for the prevention of brain metastasis in those patients at highest risk.

National Institute of Health (NIH)
National Cancer Institute (NCI)
Exploratory/Developmental Grants (R21)
Project #
Application #
Study Section
Special Emphasis Panel (ZCA1)
Program Officer
Thurin, Magdalena
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
University of Utah
Schools of Medicine
Salt Lake City
United States
Zip Code