How Prostate Cancer Growth Patterns Impact Detection and Interreader Agreement on Multiparametric MRI

Area of Science:

• Radiology
• Pathology
• Oncology

Background:

• Multiparametric MRI (mpMRI) enhances prostate cancer (PCa) detection over systematic biopsy.
• Aggressive PCa forms, particularly those with infiltrative growth (INF), can be missed by mpMRI.
• INF lesions exhibit less restricted diffusion compared to other growth patterns.

Purpose of the Study:

• To investigate the impact of prostate carcinoma growth patterns on lesion detection using mpMRI.
• To correlate histopathological growth patterns (expansive vs. infiltrative) with mpMRI findings, including PI-RADS scores and apparent diffusion coefficient (ADC) values.

Main Methods:

• Retrospective analysis of 52 radical prostatectomy patients with preoperative mpMRI.
• Independent classification of lesion growth patterns (expansive/INF) by two pathologists.
• Independent PI-RADS v2.1 classification by two radiologists based on pathological localization.
• Measurement of ADC values correlated with histopathology; interreader agreement assessed using Cohen's Kappa.

Main Results:

• High agreement for both growth pattern classification (κ=0.88) and PI-RADS scoring (κ=0.90).
• PI-RADS scores significantly correlated with expansive (EXP) growth patterns.
• EXP lesions had lower ADC values (0.83×10^-3 mm²/s) than INF lesions (0.97×10^-3 mm²/s).
• Infiltrative lesions were more frequently missed on T2-weighted imaging (29% vs. 4%).

Conclusions:

• Prostate cancer with infiltrative growth patterns is more frequently missed by mpMRI.
• Expansive growth patterns are associated with lower ADC values and higher PI-RADS scores.
• Understanding growth patterns improves the interpretation of mpMRI for aggressive PCa detection.