Development and applications of computer image analysis algorithms for scoring of PD-L1 immunohistochemistry

Area of Science:

• Oncology
• Pathology
• Immunotherapy

Background:

• Immune checkpoint inhibitors targeting programmed cell death 1 (PD-1) and programmed cell death ligand 1 (PD-L1) are crucial in cancer therapy.
• Programmed cell death ligand 1 (PD-L1) immunohistochemistry (IHC) is used for patient selection, but visual assessment by pathologists presents challenges in reproducibility.
• Variability in PD-L1 assays, detection systems, and scoring algorithms complicates reliable results.

Purpose of the Study:

• To review computer-aided image analysis software for scoring PD-L1 IHC assays.
• To explore the potential of digital pathology and artificial intelligence in improving PD-L1 assessment.
• To drive discussion and technological advancement in precision medicine for cancer treatment.

Main Methods:

• Review of existing literature and description of various image analysis software programs for computer-aided PD-L1 scoring.
• Discussion of the integration of computer technologies into pathology workflows.
• Exploration of artificial intelligence approaches for analyzing cancer cell features and tumor microenvironment interactions.

Main Results:

• Computer-aided scoring offers a potential solution to the subjectivity and variability inherent in visual PD-L1 IHC assessment.
• Digital pathology tools can enhance the reliability and reproducibility of PD-L1 assay results.
• Technological advancements are paving the way for more accurate simultaneous assessment of multiple cancer biomarkers.

Conclusions:

• Computer-aided PD-L1 scoring holds significant promise for improving the clinical utility of IHC assays in oncology.
• The integration of digital pathology and AI can overcome current limitations in PD-L1 assessment.
• Further technological advancement is essential for the evolution of precision medicine and accurate cancer diagnostics.