Analytical Performance of an Immunoprofiling Assay Based on RNA Models

Area of Science:

• Oncology
• Immunology
• Bioinformatics

Background:

• Immuno-oncology treatments are increasingly vital for cancer care.
• Accurate quantification of tumor-infiltrating immune cells is crucial for patient selection.
• Current methods like flow cytometry are limited with formalin-fixed, paraffin-embedded (FFPE) tissues.

Purpose of the Study:

• To introduce ImmunoPrism, a novel hybrid-capture RNA sequencing assay.
• To estimate the relative abundance of eight immune cell types in FFPE solid tumors.
• To provide a framework for validating complex RNA-based assays.

Main Methods:

• Development of immune cell expression models using machine learning.
• Identification of discriminative genes for each immune cell type.
• Analytical validation using FFPE and fresh-frozen samples, flow cytometry, and immunohistochemistry.

Main Results:

• ImmunoPrism demonstrated high precision (±2.72%) and low total error (2.75%) compared to flow cytometry (r²=0.81).
• The assay showed strong correlation with immunohistochemistry for CD8+ cells (r²=0.83) in FFPE samples.
• Performance metrics including limit of detection and reproducibility were determined.

Conclusions:

• ImmunoPrism offers a reliable method for quantifying immune cells in FFPE tumor specimens.
• This assay can support patient stratification for immuno-oncology therapies.
• The validation framework can guide future RNA-based assay development.