Initial magnetic resonance imaging tumour regression grade (mrTRG) as response evaluation after neoadjuvant treatment predicts sustained complete response in patients with rectal cancer

Reliable predictors of a sustained clinical complete tumour response (cCR) after neoadjuvant therapy in rectal cancer (RC) are lacking. The aim of this study was to determine if the tumour regression grade (TRG) assessed by magnetic resonance imaging (MRI), at the first restaging after neoadjuvant therapy can predict organ preservation, and to estimate the time interval after which surgery should be recommended in patients who remain in near cCR. Eighty-three consecutive patients were assessed by MRI as having a cCR (mrTRG 1) or near cCR (mrTRG 2) after neoadjuvant therapy. Cox proportional hazards regression models and Kaplan-Meier survival analyses were used to determine associations between resection-free survival (RFS) and mrTRG at the first restaging, and in relation to mrTRG with a landmark period. mrTRG and pathological findings were compared in operated patients. mrTRG 2 at the first restaging significantly predicted poorer RFS during follow up. The best prediction of RFS was mrTRG at landmark 16 weeks after termination of radiotherapy; 42 out of 49 patients (86%) evaluated as mrTRG 1 had cCR at one year of follow up. In contrast, 12 out of 15 patients (80%) evaluated as mrTRG 2 had clinical signs of tumour and were recommended surgery. The first mrTRG, and to an even greater extent mrTRG at landmark 16 weeks predicts RFS. Patients who remain mrTRG 2 at 5–6 months after radiotherapy with signs of tumour should be recommended surgery. These findings may help in patient counselling and surgical decision-making.