Whole Slide Image vs Microscope Glass Slide Performance for the Evaluation of Programmed Death-Ligand 1 (PD-L1)...

Area of Science:

• Oncology
• Pathology
• Biotechnology

Background:

• Accurate assessment of programmed death-ligand 1 (PD-L1) expression is crucial for guiding immunotherapy in cancer treatment.
• Traditional scoring relies on microscope glass slides (MGS), but digital pathology using whole slide images (WSI) offers potential advantages.

Purpose of the Study:

• To evaluate the concordance between PD-L1 expression scoring using MGS and WSI.
• To demonstrate comparable performance of MGS and WSI scoring for PD-L1 evaluation across multiple tumor types and cutoffs.

Main Methods:

• PD-L1 expression was assessed using the PD-L1 22C3 pharmDx assay on formalin-fixed, paraffin-embedded specimens for 14 tumor types.
• Qualified study readers scored PD-L1 expression using both MGS and WSI (Aperio AT2 scanner) with a washout period.
• Scoring algorithms included Tumor Proportion Score and Combined Positive Score at predefined cutoffs.

Main Results:

• Percent agreement between MGS and WSI scoring was ≥89% across all tumor types and cutoffs.
• Concordance correlation coefficient values were ≥0.80 for all tumor types and cutoffs.
• Analysis revealed high concordance with low variability and bias between MGS and WSI scoring.

Conclusions:

• Digital pathology scoring of PD-L1 expression using WSI demonstrates comparable performance to traditional MGS scoring.
• WSI offers a reliable alternative for PD-L1 evaluation, supporting its use in clinical practice and research.
• High concordance across diverse tumor types validates the robustness of WSI-based PD-L1 assessment.