Integrated morphologic analysis for the identification and characterization of disease subtypes

Abstract

Background and objectiveMorphologic variations of disease are often linked to underlying molecular events and patient outcome, suggesting that quantitative morphometric analysis may provide further insight into disease mechanisms. In this paper a methodology for the subclassification of disease is developed using image analysis techniques. Morphologic signatures that represent patient-specific tumor morphology are derived from the analysis of hundreds of millions of cells in digitized whole slide images. Clustering these signatures aggregates tumors into groups with cohesive morphologic characteristics. This methodology is demonstrated with an analysis of glioblastoma, using data from The Cancer Genome Atlas to identify a prognostically significant morphology-driven subclassification, in which clusters are correlated with transcriptional, genetic, and epigenetic events.Materials and methodsMethodology was applied to 162 glioblastomas from The Cancer Genome Atlas to identify morphology-driven clusters and their clinical and molecular correlates. Signatures of patient-specific tumor morphology were generated from analysis of 200 million cells in 462 whole slide images. Morphology-driven clusters were interrogated for associations with patient outcome, response to therapy, molecular classifications, and genetic alterations. An additional layer of deep, genome-wide analysis identified characteristic transcriptional, epigenetic, and copy number variation events.Results and discussionAnalysis of glioblastoma identified three prognostically significant patient clusters (median survival 15.3, 10.7, and 13.0 months, log rank p=1.4e-3). Clustering results were validated in a separate dataset. Clusters were characterized by molecular events in nuclear compartment signaling including developmental and cell cycle checkpoint pathways. This analysis demonstrates the potential of high-throughput morphometrics for the subclassification of disease, establishing an approach that complements genomics.

Quantitative modeling of clinical, cellular, and extracellular matrix variables suggest prognostic indicators in cancer: a model in neuroblastoma

“…We used fast and accurate methods for digitization and objective quantification. Computer-aided evaluation of pathology image analysis to generate risk stratification have been developed for lymphoma (30), glioblastoma (14), breast and prostate cancers (4). A computer-based grading system to support diagnosis for neuroblastoma that uses grades of differentiation and stromal development was published (29,31).…”

“…Tissue scaffolds regulate cell behavior and influence tumor progression (5)(6)(7)12,13). Biomathematical analysis of promising biomarker candidates, such as genomic, transcriptomic, proteomic, and epigenomic changes, at the tumor tissue level will play an important role in developing a more powerful mathematical modeling of tumor-microenvironment interactions (14,15). The information obtained can then be linked to tumor progression in patients either refractory to current therapy, or who relapse or who might benefit from novel therapies…”

Mapping spatial heterogeneity in the tumor microenvironment: a new era for digital pathology

“…56 Spatial architecture and organization of the tumor sections were considered for subtyping, although the microenvironmental structure was not explicitly modeled. In another study, 59 different cell types were also not discriminated for morphometric analysis in glioblastoma. However, glioblastoma subtypes resulting from clustering analysis of morphological data were found to be enriched with specific microenvironmental cells and genomic, methylation and expression patterns.…”

“…Although the morphological features were not cell-type specific, such a method, when extended to incorporate celltype information, is potentially promising for investigating molecular correlates of individual microenvironmental components. Therefore, although the tumor architecture has been considered during morphological analysis and integrated with omics data, [56][57][58][59] the spatial arrangement and distribution of specific microenvironmental components are rarely explicitly modeled.…”

Whole Slide Imaging for Analytical Anatomic Pathology and Telepathology: Practical Applications Today, Promises, and Perils

“…66,67 Earlier studies from our department established the role of Ki-67 quantitation in glial neoplasms. 68,69 Recently, using multimodal, multiscale approaches and machine-based classification, researchers have devised ways to mine scanned histologic data on glioblastoma in The Cancer Genome Atlas Project and other sources; the resulting quantitative morphometric analysis findings have been further integrated with molecular data to provide in silico cancer research [70][71][72][73][74][75][76][77][78][79][80] and with radiologic data to provide clinicopathoradiologic correlation. 81 Insights from this work also include findings on the importance of tumor-infiltrating lymphocytes in glioblastoma, 72 and in silico approaches from these studies have uncovered novel findings regarding the regulation of asymmetric cell division in glioblastoma by such mediators as the human Brat ortholog TRIM3.…”