Radiogenomics in Medulloblastoma: Can the Human Brain Compete with Artificial Intelligence and Machine Learning?

Molecular Biology in Treatment Decision Processes-Neuro-Oncology Edition

Computational approaches including machine learning, deep learning, and artificial intelligence are growing in importance in all medical specialties as large data repositories are increasingly being optimised. Radiation oncology as a discipline is at the forefront of large-scale data acquisition and well positioned towards both the production and analysis of large-scale oncologic data with the potential for clinically driven endpoints and advancement of patient outcomes. Neuro-oncology is comprised of malignancies that often carry poor prognosis and significant neurological sequelae. The analysis of radiation therapy mediated treatment and the potential for computationally mediated analyses may lead to more precise therapy by employing large scale data. We analysed the state of the literature pertaining to large scale data, computational analysis, and the advancement of molecular biomarkers in neuro-oncology with emphasis on radiation oncology. We aimed to connect existing and evolving approaches to realistic avenues for clinical implementation focusing on low grade gliomas (LGG), high grade gliomas (HGG), management of the elderly patient with HGG, rare central nervous system tumors, craniospinal irradiation, and re-irradiation to examine how computational analysis and molecular science may synergistically drive advances in personalised radiation therapy (RT) and optimise patient outcomes.

A pilot radiogenomic study of DIPG reveals distinct subgroups with unique clinical trajectories and therapeutic targets

An adequate understanding of the relationships between radiographic and genomic features in diffuse intrinsic pontine glioma (DIPG) is essential, especially in the absence of universal biopsy, to further characterize the molecular heterogeneity of this disease and determine which patients are most likely to respond to biologically-driven therapies. Here, a radiogenomics analytic approach was applied to a cohort of 28 patients with DIPG. Tumor size and imaging characteristics from all available serial MRIs were evaluated by a neuro-radiologist, and patients were divided into three radiographic response groups (partial response [PR], stable disease [SD], progressive disease [PD]) based on MRI within 2 months of radiotherapy (RT) completion. Whole genome and RNA sequencing were performed on autopsy tumor specimens. We report several key, therapeutically-relevant findings: (1) Certain radiologic features on first and subsequent post-RT MRIs are associated with worse overall survival, including PD following irradiation as well as present, new, and/or increasing peripheral ring enhancement, necrosis, and diffusion restriction. (2) Upregulation of EMT-related genes and distant tumor spread at autopsy are observed in a subset of DIPG patients who exhibit poorer radiographic response to irradiation and/or higher likelihood of harboring H3F3A mutations, suggesting possible benefit of upfront craniospinal irradiation. (3) Additional genetic aberrations were identified, including DYNC1LI1 mutations in a subgroup of patients with PR on post-RT MRI; further investigation into potential roles in DIPG tumorigenesis and/or treatment sensitivity is necessary. (4) Whereas most DIPG tumors have an immunologically “cold” microenvironment, there appears to be a subset which harbor a more inflammatory genomic profile and/or higher mutational burden, with a trend toward improved overall survival and more favorable radiographic response to irradiation, in whom immunotherapy should be considered. This study has begun elucidating relationships between post-RT radiographic response with DIPG molecular profiles, revealing radiogenomically distinct subgroups with unique clinical trajectories and therapeutic targets.