MRI after chemoradiotherapy of rectal cancer: a useful tool to select patients for local excision

PURPOSE: If identification of good responders to neoadjuvant chemoradiotherapy in rectal cancer is possible, there might be opportunities for local excision in selected patients. The aim of this study was to determine whether postchemoradiation MRI in rectal cancer can accurately identify ypT0 to 2/ypN0, because both features are essential for identification of good responders. METHODS: Seventy-nine patients (4 hospitals) underwent postchemoradiation MRI, 62 received a lymph node-specific contrast agent (ultrasmall superparamagnetic iron oxide). An expert and general radiologist prospectively predicted whether the tumor penetrated the mesorectal fat and whether nodes were sterilized after chemoradiation. Histology was the reference standard. Sensitivity, specificity, positive predictive value, and negative predictive value were calculated. RESULTS: For prediction of whether a tumor penetrated the bowel wall, there was an negative predictive value of 0.90 and 0.76 for the expert and general radiologist, respectively. The negative predictive value for prediction of nodal status was 0.95 and 0.85 for expert and general radiologist, respectively. CONCLUSION: This prospective multicenter study demonstrates that MRI with a lymph node-specific contrast agent interpreted by an expert radiologist can select ypT0 to 2/ypN0 rectal cancer with low risk of undetected nodal metastases or invasion through the bowel wall. These patients could thus have been selected for local excision. However, future studies will have to prove equal outcome of such a modified surgical approach compared with current practice.