Abstract
AbstractBackgroundThe current WHO classification integrates histological and molecular features of brain tumors. The aim of this study was to identify generalizable topological patterns with the potential to add an anatomical dimension to the classification of brain tumors.MethodsWe applied non-negative matrix factorization as an unsupervised pattern discovery strategy to the fine-grained topographic tumor profiles of 936 patients with primary and secondary brain tumors. From the anatomical features alone, this machine learning algorithm enabled the extraction of latent topological tumor patterns, termed meta-topologies. The optimal parts-based representation was automatically determined in 10,000 split-half iterations. We further characterized each meta-topology’s unique histopathologic profile and survival probability, thus linking important biological and clinical information to the underlying anatomical patternsResultsIn primary brain tumors, six meta-topologies were extracted, each detailing a transpallial pattern with distinct parenchymal and ventricular compositions. We identified one infratentorial, one allopallial, three neopallial (parieto-occipital, frontal, temporal) and one unisegmental meta-topology. Each meta-topology mapped to distinct histopathologic and molecular profiles. The unisegmental meta-topology showed the strongest anatomical-clinical link demonstrating a survival advantage in histologically identical tumors. Brain metastases separated to an infra- and supratentorial meta-topology with anatomical patterns highlighting their affinity to the cortico-subcortical boundary of arterial watershed areas.ConclusionsUsing a novel data-driven approach, we identified generalizable topological patterns in both primary and secondary brain tumors Differences in the histopathologic profiles and prognosis of these anatomical tumor classes provide insights into the heterogeneity of tumor biology and might add to personalized clinical decision making.
Publisher
Cold Spring Harbor Laboratory