Integrating Proteomic Analysis and Machine Learning to Predict Prostate Cancer Aggressiveness

Area of Science:

• Oncology
• Molecular Pathology
• Biomarker Discovery

Background:

• Prostate cancer (PCa) diagnosis is challenged by difficulty in identifying aggressive tumors, leading to overtreatment.
• Accurate prediction of PCa aggressiveness is crucial for personalized therapy and avoiding unnecessary interventions.

Purpose of the Study:

• To investigate the utility of retinoblastoma phosphorylated at Serine 249 (Phospho-Rb S249), N-cadherin, β-catenin, and E-cadherin as biomarkers for aggressive PCa.
• To correlate biomarker expression with clinicopathological data for improved diagnostic accuracy.

Main Methods:

• Immunohistochemistry (IHC) was used to assess biomarker expression in PCa tissues.
• Logistic regression and Classification and Regression Tree (CART) models were employed to analyze biomarker correlations with tumor aggressiveness and clinicopathological data.

Main Results:

• E-cadherin and β-catenin showed a negative correlation with aggressive tumor behavior.
• Phospho-Rb S249 and N-cadherin positively correlated with increased PCa aggressiveness.
• CART analysis identified β-catenin, tumor grade, and Gleason grade as key determinants for identifying patients with Gleason scores ≥ 4 + 3.

Conclusions:

• Biomarkers including E-cadherin, β-catenin, Phospho-Rb S249, and N-cadherin show potential for identifying aggressive prostate cancer.
• Classification and Regression Tree (CART) models offer an effective method for clinical utility assessment of these biomarkers.
• Early detection of aggressive PCa through these biomarkers can guide patient monitoring and treatment strategies.