Unveiling tumor invasiveness: enhancing cancer diagnosis with phase-contrast microtomography for 3D virtual histology

Abstract

Malignant diseases are characterized by a critical trait known as invasiveness, where tumor cells tend to spread from the primary tissue layer into surrounding healthy tissues and distant organs. Presently, histopathology offers essential insights for diagnosing, classifying, predicting outcomes, and guiding patient-specific treatments. However, histology offers two-dimensional data from chosen cutting planes. Although 3D histological volumes can be generated through serial sectioning or whole slide imaging, this method is laborious, may introduce processing artefacts, and lacks isotropic spatial resolution. These limitations pose a considerable challenge to accurate diagnoses, particularly when dealing with micro-infiltrating carcinomas. These lesions, characterized by minute infiltrations, demand a three-dimensional representation for comprehensive visualization, essential for precise identification and assessment. Emerging X-ray-based virtual histology technology offers three-dimensional visualization of soft-tissue specimens, enabling virtual slicing in any direction or at any point. This approach can assist in guiding tissue sectioning for optimal representation of tumor cross sections during histological analysis. Micro-infiltrating carcinomas from the breast, cervix, and thyroid were imaged using X-ray phase-contrast microtomography (PhC-$$\mu$$ μ CT) at the Elettra synchrotron facility in Trieste, Italy. Comparative assessment of histological and CT slices by pathologists revealed that PhC-$$\mu$$ μ CT aids in classifying lesions by highlighting distinct tissue components and, notably, identifying tissue invasion. Reviewing a volume image allows pathologists to trace the entire lesion, identifying invasion sites that might be overlooked in individual or serial histological sections. Consequently, this proposed method could complement pathologists’ tools, potentially enhancing diagnoses by minimizing under-staging and reducing false negative results.