Background: Our goals are to develop a computational histopathology pipeline for characterizing tumor types\r\nthat are being generated by The Cancer Genome Atlas (TCGA) for genomic association. TCGA is a national\r\ncollaborative program where different tumor types are being collected, and each tumor is being characterized\r\nusing a variety of genome-wide platforms. Here, we have developed a tumor-centric analytical pipeline to process\r\ntissue sections stained with hematoxylin and eosin (H&E) for visualization and cell-by-cell quantitative analysis. Thus\r\nfar, analysis is limited to Glioblastoma Multiforme (GBM) and kidney renal clear cell carcinoma tissue sections. The\r\nfinal results are being distributed for subtyping and linking the histology sections to the genomic data.\r\nResults: A computational pipeline has been designed to continuously update a local image database, with limited\r\nclinical information, from an NIH repository. Each image is partitioned into blocks, where each cell in the block is\r\ncharacterized through a multidimensional representation (e.g., nuclear size, cellularity). A subset of morphometric\r\nindices, representing potential underlying biological processes, can then be selected for subtyping and genomic\r\nassociation. Simultaneously, these subtypes can also be predictive of the outcome as a result of clinical treatments.\r\nUsing the cellularity index and nuclear size, the computational pipeline has revealed five subtypes, and one\r\nsubtype, corresponding to the extreme high cellularity, has shown to be a predictor of survival as a result of a\r\nmore aggressive therapeutic regime. Further association of this subtype with the corresponding gene expression\r\ndata has identified enrichment of (i) the immune response and AP-1 signaling pathways, and (ii) IFNG, TGFB1, PKC,\r\nCytokine, and MAPK14 hubs.\r\nConclusion: While subtyping is often performed with genome-wide molecular data, we have shown that it can\r\nalso be applied to categorizing histology sections. Accordingly, we have identified a subtype that is a predictor of\r\nthe outcome as a result of a therapeutic regime. Computed representation has become publicly available through\r\nour Web site.
Loading....