Brain tumors are the most common cause of pediatric cancer-related death in the United States. Glial tumors (astrocytomas, oligodendrogliomas, etc.) are the most common class of pediatric brain tumors with diffuse astrocytomas being of particular interest because they infiltrate normal brain tissue, making surgical control difficult even when they are of low histological grade. Pediatric and adult diffuse astrocytomas share a morphologic classification, but appear to differ in both their clinical behavior and the molecular processes that drive them. Adult low-grade diffuse astrocytomas nearly always transform into deadly, high grade glioblastomas, while pediatric tumors are much less likely to do so. Of low-grade pediatric tumors that do not transform however, 80% will continue to grow and progress, perhaps requiring multiple interventions. Molecular profiling data on pediatric diffuse astrocytomas is still limited mostly to genomic analyses with minimal methylation data and virtually no reported proteomic information. What data do exist have not been matched with clinical information to see whether specific molecular signals are responsible for tumor progression. Furthermore, because tumors in the adolescent/young adult age range are poorly studied, it is still unclear at what age these apparent shifts in tumor behavior take place.
We propose that comprehensive molecular profiling including proteomic analysis and DNA methylation, as well as whole exome sequencing (WES) and transcriptome profiling of intermediate-grade diffuse pediatric astrocytomas combined with clinical data on time to progression will reveal molecular signatures associated with time to progression. To our knowledge, though progression is almost universal in pediatric tumors, there are no studies that have attempted to correlate this outcome with specific molecular signals. Comparing comprehensive molecular profiles of tumors from children, adolescents, and young adults will clarify differences in the mechanisms that drive tumors in these age groups. Current data indicate that pediatric and adult tumors have very different molecular profiles, however patients in the adolescent and young adult spectrum are poorly studied and it is unclear when this difference occurs.
What are the goals of this project?
Using an independent cohort of diffuse astrocytoma patients within the CBTN, we plan to correlate the genomic and transcriptomic profile to previously performed molecular profiling of patients recruited at Nationwide Children’s Hospital/Ohio State University Wexner Medical Center. Comprehensive molecular profiling, including genomic, transcriptomic, proteomic and DNA methylation are in progress with the goal of understanding the molecular basis for progression and informing treatment decisions.
Using an independent cohort of diffuse astrocytoma patients within the CBTN, we will evaluate and compare genomic and transcriptomic profiles according to age group with the goal of understanding how molecular drivers correlate with human development and providing clinicians with information which can be used in determining prognosis and developing focused molecular testing and treatment strategies for their patients.
What is the impact of this project?
In 2016, the WHO incorporated molecular data into their classification of astrocytomas for the first time by including IDH1/2 mutation status as a core component due to its improved accuracy in both disease prognosis and prediction of disease responses to therapy. What is currently known about pediatric diffuse astrocytomas is that they have a lower tumor mutational burden compared to their adult counterparts and that they display a different set of genetic alterations. There is also a sub-class of tumors with mutations in histone-related protein H3 that appears to be unique to pediatric tumors. Despite these gains in molecular understanding however, there are still large gaps in what is known about pediatric tumor formation and growth. By performing comprehensive molecular profiling on this important subset of tumors, our study will provide insight regarding the molecular processes that are driving their formation and progression.
This study will specifically study pediatric patients utilizing a comprehensive set of analytic platforms. Notably missing from published pediatric cohorts are proteomic data to provide functional insight. The data we gather are by their nature exploratory, requiring bioinformatics analysis to determine what molecular markers are significant enough to then pursue as targets of future research studies. We recognize that our institutional cohort of 50 patients is unlikely to produce a complete picture of all of the important molecular signals even within the relatively narrow class of diffuse astrocytomas, but we expect that our approach will yield some important and novel targets for further study and hope that larger, validation cohorts, such as the CBTN cohort, can expand on. Second, we will incorporate clinical data on time to progression in order to address the question of what molecular characteristics drive tumor progression. Most low grade diffuse astrocytomas continue to grow and progress and understanding the molecular mechanisms could allow not only for better clinical prognostication, but also for potential future treatments that could reduce progression and prevent the need for invasive treatments. These data will pave the way for improved age-based risk stratification as well as future studies to understand if how molecular signatures may interact with or be affected by the normal developmental program of the brain.
Why the CBTN request is important to this project?
Proteomic analysis evaluates the functional unit of gene expression. Due to the technically challenging methodology of these analyses, proteomic profiling is under-represented in the pediatric diffuse astrocytoma literature. Through collaboration between Nationwide Children’s Hospital and Ohio State University Wexner Medical Center, we have established an institutional cohort of pediatric and adolescent diffuse astrocytoma patients whose tumors underwent comprehensive molecular profiling, including proteomics and DNA methylation. While this is a substantial cohort, larger validation studies are needed to confirm our studies. Although proteomic data is not yet available within the CBTN dataset, transcriptomics could function as a surrogate for proteomic data. Thus, through evaluation of genomic and transcriptomic data within the CBTN, we hope to independently validate findings from our institutional cohort obtained from a database of pediatric tumors with known clinical outcomes.
The Children's Brain Tumor Network contributed to this project by providing access to the Pediatric Brain Tumor Atlas.
Richard Graham, St. Jude Children's Hospital
Blake Sells, Ohio State University Wexner Medical Center
Jessica Fleming, Ohio State University Wexner Medical Center
Richard K. Wilson, PhD
Dr. Wilson is an internationally recognized expert in molecular genetics and large-scale genomics, and his laboratories have been among the world’s leaders in genome analysis. His teams have sequenced and analyzed billions of bases of DNA from the genomes of bacteria, yeast, plants, invertebrates