Accuracy of pelvic magnetic resonance imaging in local staging for rectal cancer: a single local health district, real world experience.

Abstract

Neoadjuvant chemoradiotherapy plays a key role in reducing local recurrence rates for locally advanced rectal cancer. Pelvic magnetic resonance imaging (pMRI) is the gold standard for local clinical staging which allows clinicians to decide the treatment patients receive. A more advanced tumour or the presence of high-risk features on pMRI mean that neoadjuvant therapy will be offered to these patients. Understanding the accuracy of pMRI in local staging for rectal cancer is therefore crucial. A retrospective cohort analysis of the accuracy of pMRI staging in a subgroup of patients who had primary rectal cancer surgery without neoadjuvant therapy was performed. Specificity and sensitivity for T-staging, N-staging and presence of high-risk features (threatened circumferential resection margin and extramural venous invasion) were calculated. Patients who had previous pelvic surgery, previous pelvic radiotherapy and previous surgery for continence were excluded. A total of 114 patients were included in the analysis. MRI accurately predicts T-stage in 56.6% and N-stage in 55.8%. Prediction of extramural disease was accurate in 51%. A negative circumferential resection margin was accurately predicted in 98.6% of patients. Overall adherence to reporting proforma was 15.8%. Overall, this study provided valuable information about the clinical staging of patients with rectal cancer who are at an early stage within a large regional catchment area in Australia with pMRI. These results allow us to assess the accuracy of our local staging with ramifications to the clinical decisions being made in the context of the more recent trials which questioned the need for neoadjuvant chemo-radiotherapy in all node positive patients.