Juvenile pilocytic astrocytoma (JPA), the most common type of brain tumor in children, has a high cure rate only if complete surgical removal of the tumor is possible. Unfortunately, due to the heterogeneity in the location of JPA, complete resection may not always be possible. As a result, the long-term survival of patients with residual tumor is less than 20% when no adjuvant therapy is given. Currently, no diagnostic tools exist to allow clinicians to distinguish JPA that will be progressive or refractory. Thus observation has been recommended for children below 3 to 5 years of age even with residual tumor. It is because the treatments for aggressive brain tumors in children often carry significant long-term consequences; therefore, such treatment is not acceptable as a means to prevent recurrence in JPA that already has a high probability of cure. Ideally, clinicians would be able to distinguish early on between JPA that are likely to be cured and those that are likely to recur or progress. If this is possible, children with more at-risk tumors could receive more aggressive intervention early, while those with more benign tumors could be spared complications of excessive therapy. The proposed studies seek to identify prognostic markers for subtotally resected JPAs, which mostly located at the midline region and have a 50% chance of recurrent or progressive disease. Based on our recent report, JPA can be classified as two subgroups that may differ in aggressiveness. However, our samples are mostly from totally resected JPA located at the cerebellar region. To address the clinically relevant question of progression in JPA, we need to extend our study to subtotally resected JPA located at the midline region. We propose to (1) profile 40 cases of subtotally resected JPAs located at the midline region to test whether midline located JPA also forms two subgroups, and (2) to identify prognostic markers that can predict the likelihood of progression in subtotally resected JPAs. A supervised analysis and the Leave-One-Out Cross Validation approach will be applied to the expression profiles data to be generated. The completion of this study will not only generate a preliminary multi-gene prognostic model for stratifying patients in future clinical trials to assess the role of adjuvant therapy on JPA patients with subtotal resection, but also provide insight into the pathogenetics of this disease. ? ? ?
