Aims: The nature of recurrence affecting those with rectal cancer is changing owing to developments in surgical techniques, radiotherapy and chemotherapy. We sought to assess the nature of recurrence in a large regional Australian population with stage I-III rectal adenocarcinoma, and to identify factors associated with recurrence.
Methods: Patients with rectal adenocarcinoma being treated with curative intent were included in this retrospective cohort study. Univariate analysis and multivariate analysis using Cox regression models were used to identify factors associated with recurrence.
Results: 483 patients were included, and recurrence occurred in 117 of 483 patients (24.2%). Recurrence was locoregional in 15 patients (3.1%), distant in 85 patients (17.6%) and both locoregional and distant in 17 patients (3.5%). On univariate analysis, factors associated with recurrence included stage (stage II, HR 4.83, stage III HR 5.45, p<0.001), grade (grade 3 HR 2.41, p<0.001), distal tumour (HR 1.89, p=0.001), number of nodes involved (HR 1.08, p<0.001), radiotherapy course (long course HR 1.73, p=0.005), radiotherapy type (neoadjuvant HR 1.72, p=0.01; adjuvant HR 1.87, p=0.03), chemotherapy (single agent HR 1.65, p=0.01) and surgery (anterior resection HR 1.69, p=0.01). On multivariate analysis, stage (stage II HR 6.19, p <0.001, stage III HR 6.99, p<0.001), grade (grade 3 HR 2.27, p=0.004), distal tumour (HR 1.86, p=0.02) and nodal involvement (HR 1.06, p=0.02) remained independently associated with any recurrence. Only stage (stage II HR 7.33, p<0.001, stage III HR 8.54, p<0.001) and grade (grade 3 HR 2.59, p=0.001) remained significantly associated with distant recurrence on multivariate analysis.
Conclusions: Increasing stage, grade, distal tumour location and increasing number of involved nodes was independently associated with rectal cancer recurrence.