Locally advanced rectal cancer is commonly treated by neoadjuvant therapy and the resultant tumour response can be quantified histologically. This therapy may also induce radiation colitis, which also can be graded. The aim of this study was to assess the grading of tumour regression and of radiation colitis and their relationship to other prognostic parameters. Between 2000 and 2006, 75 patients (23 women; median duration of follow up, 58 months) with rectal cancer were evaluated. Sixty-three had short-course radiotherapy and 12 had long-course radiotherapy. Tumour regression was graded histologically using the three-point Ryan system: patients with grades 1 and 2 were considered as responders and patients with grade 3 were considered as nonresponders. Radiation colitis was graded histologically as mild, moderate or severe, as described previously (J Pathol 2006; 210: P25). Twenty-nine patients were classified as responders and 46 as nonresponders. The former were less likely to be lymph node positive compared with the latter (P=0.001). Tumour response did not correlate with local recurrence. Responders showed a disease-free survival (not overall survival) advantage at 2 and 5 years over nonresponders. Responders showed a higher rate of postoperative abdominal complications. Histological evidence of regression was demonstrated in patients treated with short-course radiotherapy. There was no relationship between radiation colitis grade and abdominal complications. Radiation colitis grade does not correlate with postoperative complications. More abdominal complications occurred in patients receiving long-course radiotherapy.