Correlation Between ImageJ and Conventional Manual Scoring Methods for Programmed Death-Ligand 1...

Area of Science:

• Oncology
• Pathology
• Biotechnology

Background:

• Immunohistochemistry (IHC) is a key method for quantifying programmed cell death-ligand 1 (PD-L1) expression in tumors.
• Accurate PD-L1 quantification can predict patient response to cancer immunotherapies like anti-PD1/PD-L1 agents.

Purpose of the Study:

• To evaluate the efficacy of ImageJ software in scoring PD-L1 expression via IHC.
• To compare ImageJ scoring results with the conventional manual method across different tissue types and imaging parameters.

Main Methods:

• ImageJ software was employed to analyze IHC-stained tissue sections for PD-L1 expression.
• ImageJ scoring was performed using different color space options (RGB, HSB, Brightness/Contrast) and magnifications (20x, 40x).
• ImageJ results were compared against scores from the conventional manual method and between camera-captured and whole slide images.

Main Results:

• ImageJ provided comparable scores to manual methods for homogeneous tissues (diffuse large B cell lymphoma) using RGB or Brightness/Contrast settings.
• Significant discrepancies were observed between ImageJ (HSB) and manual scoring for both homogeneous and heterogeneous tissues (head and neck squamous cell carcinoma).
• ImageJ analysis showed variations based on magnification and image acquisition method (camera vs. scanner).

Conclusions:

• ImageJ is a suitable tool for scoring PD-L1 in homogeneous tumor tissues.
• For heterogeneous tumor tissues, the conventional manual scoring method is recommended over ImageJ due to potential inaccuracies.
• Image acquisition parameters significantly influence ImageJ scoring outcomes.