Clinical significance and molecular annotation of cellular morphometric subtypes in lower-grade gliomas discovered by machine learning

Abstract

BACKGROUND Lower grade gliomas (LGG) are heterogenous diseases by clinical, histological, and molecular criteria. We aimed to personalize the diagnosis and therapy of LGG patients by developing and validating robust cellular morphometric subtypes (CMS) and to uncover the molecular signatures underlying these subtypes. METHODS Cellular morphometric biomarkers (CMBs) were identified with artificial intelligence technique from TCGA-LGG cohort. Consensus clustering was used to define CMS. Survival analysis was performed to assess the clinical impact of CMBs and CMS. A nomogram was constructed to predict 3- and 5- year overall survival (OS) of LGG patients. Tumor mutational burden (TMB), and immune cell infiltration between subtypes were analyzed using the Mann-Whitney test. The double-blinded validation for important immunotherapy-related biomarkers were executed using immunohistochemistry (IHC). RESULTS We developed a machine learning pipeline to extract CMBs from whole slide images of tissue histology; identifying and externally validating robust CMS of LGGs in multi-center cohorts. The subtypes had independent predicted OS across all three independent cohorts. In the TCGA-LGG cohort, patients within the poor-prognosis subtype responded poorly to primary and follow-up therapies. LGGs within the poor-prognosis subtype were characterized by high mutational burden, high frequencies of copy number alterations, and high levels of tumor-infiltrating lymphocytes and immune checkpoint genes. Higher levels of PD-1/PD-L1/CTLA-4 were confirmed by immunohistochemical staining. In addition, the subtypes learned from LGG demonstrates translational impact on glioblastoma (GBM). CONCLUSIONS We developed and validated a framework (CMS-ML) for CMS discovery in LGG associated with specific molecular alterations, immune micro-environment, prognosis, and treatment response.

iCEMIGE: Integration of CEll-morphometrics, MIcrobiome, and GEne biomarker signatures for risk stratification in breast cancers

“…Over 300 million cellular objects from 1085 diagnostic slides of 1017 TCGA-BRCA patients were recognized and delineated by an unsupervised feature learning pipeline based on SPSD[ 24 ]. Each cellular object was represented with 15 morphometric properties as described in our previous work[ 10 ].…”

“…In the past, we followed the same path to define the 12-gene expression prognosis score (GEPS)[ 21 ] and the 15-microbe abundance prognosis score (MAPS)[ 19 ] in BC. Here, we developed the 39-CMB prognosis score (CMPS) using an AI-driven CMB detection technique[ 10 ]. We found that CMPS, MAPS, and GEPS had an independent prognostic value.…”

“…Following our previous work[ 10 ], we deployed an unsupervised feature learning pipeline, which was based on the stacked predictive sparse decomposition (SPSD)[ 24 , 25 ], for unsupervised discovery of underlying cellular morphometric characteristics from 15 cellular morphological features that were extracted from the diagnostic slides from the TCGA-BRCA cohort. 256 cellular morphometric biomarkers (CMB) were defined for cellular object representation.…”

“…A complete and accurate pathological cancer classification is still crucial to deciding on the best treatment plan for patients. Recently, we developed a framework powered by artificial intelligence (AI) technique for identifying cellular morphometric biomarkers (CMBs) and cellular morphometric subtypes (CMSs) from the whole slide images (WSI) of Hematoxylin and Eosin (H&E)-stained tissue histology[ 10 , 11 ]. We demonstrated that CMSs were significantly associated with specific molecular alterations, immune microenvironment, and prognosis in lower-grade gliomas[ 10 ].…”

Editorial comment: “Diffuse glioma, not otherwise specified: imaging-based risk stratification achieves histomolecular-level prognostication”

“…More accurate characterization of tumors may provide insight into novel imaging phenotypes, given the unprecedented demand for genetic properties. With the availability of datasets such as The Cancer Genome Atlas Low Grade Glioma (TCGA-LGG), it is now possible to perform large-scale analyses that are not restricted to one institution [7]. While machine learning methods such as radiomics and deep learning can be expected to provide more accurate diagnostic outcomes than conventional analyses that rely on the human eye, it is still challenging to translate these methods into indicators that can be easily applied in clinical practice.…”