Development and Optimization of a Subtraction-Normalized Immunocyte Profiling Signature for Prostate Cancer Active...

Area of Science:

• Oncology
• Genomics
• Biomarker Discovery

Background:

• Active surveillance (AS) for prostate cancer (PCa) requires repeat biopsies to detect disease progression.
• Many men on AS avoid necessary biopsies, increasing the risk of missing aggressive cancer.
• There is a need for less invasive tools to identify high-risk PCa in men undergoing AS.

Purpose of the Study:

• To develop and validate a blood-based immunocyte transcriptomic signature for identifying occult aggressive PCa.
• To stratify men with PCa for active surveillance suitability.
• To differentiate patients who should not be on AS from those who can safely remain on AS.

Main Methods:

• Blood samples were collected from training (n=1017) and validation (n=1198) cohorts coincident with prostate biopsy.
• RNA was extracted and sequenced from purified CD2+ and CD14+ immune cells.
• A regularized regression model was built to predict PCa aggressiveness based on National Comprehensive Cancer Network guidelines.

Main Results:

• The final model combined an immunocyte transcriptomic profile with PSA density and age, achieving an AUC of 0.73.
• This model significantly outperformed PSA density alone (AUC 0.69).
• The model provides an individualized risk score with 90% negative predictive value and 50% positive predictive value.

Conclusions:

• The immunocyte transcriptomic model is a promising tool for risk stratification in men considered for AS.
• Further validation in an intended-use cohort is necessary.
• This blood-based test could reduce the need for repeat biopsies in active surveillance for prostate cancer.