Diagnostic Performance and Misclassification Patterns of Preoperative MRI in Rectal Cancer: A Real-World Study.

Area of Science:

• Oncology
• Radiology
• Pathology

Background:

• Magnetic resonance imaging (MRI) is the standard for rectal cancer (RC) staging.
• Real-world agreement between MRI and surgical pathology is often limited.
• Accurate staging is crucial for treatment planning and patient outcomes.

Purpose of the Study:

• To assess the agreement and diagnostic performance of preoperative MRI for dichotomized T and N staging in RC.
• To explore MRI misclassification patterns and associated preoperative factors.

Main Methods:

• Retrospective analysis of 152 consecutive rectal cancer patients.
• Two cohorts: no neoadjuvant therapy (non-NAT) and neoadjuvant therapy (NAT).
• Dichotomized T (≤T2 vs. ≥T3) and N (N0 vs. N+) staging compared to final pathology.

Main Results:

• Overall agreement: 72.4% for T stage (kappa=0.452) and 73.0% for N stage (kappa=0.349).
• MRI performance was better for advanced T-stage assessment than nodal staging.
• T-stage errors predominantly involved overstaging; NAT restaging MRI showed improved agreement.

Conclusions:

• Real-world MRI-pathology agreement for dichotomized rectal cancer staging is moderate.
• MRI is a practical tool for multidisciplinary risk stratification.
• Continued monitoring of MRI performance in routine care is essential.