Radiological tumour classification across imaging modality and histology

Abstract

Radiomics refers to the high-throughput extraction of quantitative features from radiological scans and is widely used to search for imaging biomarkers for prediction of clinical outcomes. Current radiomic signatures suffer from limited reproducibility and generalizability, because most features are dependent on imaging modality and tumor histology, making them sensitive to variations in scan protocol. Here, we propose novel radiological features that are specially designed to ensure compatibility across diverse tissues and imaging contrast. These features provide systematic characterization of tumor morphology and spatial heterogeneity. In an international multi-institution study of 1,682 patients, we discover and validate four unifying imaging subtypes across three malignancies and two major imaging modalities. These tumor subtypes demonstrate distinct molecular characteristics and prognoses after conventional therapies. In advanced lung cancer treated with immunotherapy, one subtype is associated with improved survival and increased tumor-infiltrating lymphocytes compared with the others. Deep learning enables automatic tumor segmentation and reproducible subtype identification, which can facilitate practical implementation. The unifying radiological tumor classification may inform prognosis and treatment response for precision medicine.

Deep Learning With Radiomics for Disease Diagnosis and Treatment: Challenges and Potential

“…( 121 ) used an extreme learning machine model combining transfer learning and feature fusion to classify voxels from multimodality MRI of patients with brain tumors automatically. Moreover, cross-modality, uni- or multi-modal, and united adversarial learning-based approaches were employed for lesion segmentation in multiple cancers (lung, breast, and brain) ( 122 ), lung cancer ( 88 ), and liver tumors ( 123 ), respectively.…”

“…Wu et al. ( 122 ) designed a deep learning method across two modalities and three cancer types to ensure reproducible automatic tumor segmentation and recognition. Zhao et al.…”

Lack of Benefit of Extending Temozolomide Treatment in Patients with High Vascular Glioblastoma with Methylated MGMT

“…Additionally, a high tumor vascularity could be related with a faster progression, hindering the damaging effect of temozolomide on tumor cells. In this sense, advanced MRI-based methodologies can complement molecular analysis to help in glioblastoma characterization and therapy selection [54][55][56][57].…”

At the Cutting Edge against Cancer: A Perspective on Immunoproteasome and Immune Checkpoints Modulation as a Potential Therapeutic Intervention

“…Thus, it clearly emerges that additional tumor type-specific studies should be performed to unveil the role of this biomarker in the ICKi response. Anyway, the identification of an additional biomarker as well as of non-invasive techniques that monitor the microenvironment before and during the course of the treatment (e.g., imaging-based radiogenomics) are urgently needed for selecting patients who will benefit from immunotherapy [214,215].…”