Magnetic Resonance Tumor Regression Grade and Residual Mucosal Abnormality as Predictors for Pathological Complete Response in Rectal Cancer Postneoadjuvant Chemoradiotherapy.

Background: Pathological complete response after chemoradiotherapy for rectal cancer occurs in 10% to 30% of patients. The best method to identify such patients remains unclear. Clinical assessment of residual mucosal abnormality is considered the most accurate method. In our institution, magnetic resonance tumor regression grade is performed as routine to assess response.

Objective: The purpose of this study was to compare the sensitivity of magnetic tumor regression grade against residual mucosal abnormality in detecting patients with a pathological complete response.

Design: Magnetic tumor regression grade scores from reported posttreatment MRI scans were documented. Magnetic tumor regression grade 1 to 3 was defined as likely to predict complete or near complete response. Gross appearances of the mucosa were derived from histopathology reports and used as a surrogate for clinical assessment (previously validated). Final histopathological staging was used to determine response.

Settings: The study was conducted at Royal Marsden National Health Service Trust, United Kingdom.

Patients: A total of 143 patients with rectal adenocarcinoma, diagnosed between September 1, 2009, and September 1, 2013, who received neoadjuvant chemoradiotherapy before curative surgery were included.

Main Outcome Measures: The sensitivity of magnetic tumor regression grade and residual mucosal abnormality in detecting patients with pathological complete response were measured

Results: : Eighteen patients had a pathological complete response. Seventeen were detected using magnetic resonance tumor regression grade 1 to 3, with sensitivity 94% (95% CI, 0.74-0.99), and 10 were detected using residual mucosal abnormality, with sensitivity 62% (95% CI, 0.38-0.81). There was no statistical difference between the false positive rates for either method. Magnetic tumor regression grade identified 10 times more patients with a pathological complete response (diagnostic OR = 10.2 (95% CI, 1.30-73.73)) compared with clinical assessment with RMA.

Limitations: Residual mucosal abnormality was used as a surrogate marker for endoscopic appearances.

Conclusions: Most patients with rectal cancer who have a pathological complete response do not manifest a complete response at the mucosal level. Magnetic tumor regression grade is able to identify 10 times more patients than clinical assessment, with no significant compromise in the false positive rate.