Preoperative versus postoperative radiotherapy for rectal cancer in a decision analysis and outcome prediction model

Reduction of local recurrences has been achieved by radiotherapy, but also by improved surgical technique (total mesorectal excision). Radiotherapy has adverse effects and cannot exceed local dose limits. Neoadjuvant radiotherapy may result in overtreatment. We aimed to define the minimum local benefit that would have to be postulated for radiotherapy in order to bring a benefit to the overall cohort. We hypothesized that saving radiotherapy as treatment for a subset of patients with high-risk tumors and local recurrences improves the outcome of the overall cohort. We sought to simulate preoperative versus postoperative radiotherapy in theoretical decision analysis model based on published recurrence rates, with overall survival being the primary end point.

Computerized literature search for studies published between 1996 and 2011, supplemented by manual review of the retrieved reference lists.

Postoperative radiotherapy evolved as preferred strategy with cure rates of 65.6 % vs. 63.7 % for postoperative and neoadjuvant radiotherapy, respectively, and a decrease of radiation exposure to 42.9 % of the cohort. The system was sensitive to (1) the fraction of stage I cancers included in the cohort, (2) the difference between local recurrence rates (LRR) for neoadjuvant radiotherapy, adjuvant radiotherapy, or surgery-only approach, and (3) the compliance with the postoperative radiotherapy. If the surgery-only recurrence was set to the published 10 %, 13 %, and 27 %, respectively, adjuvant radiotherapy had to achieve LRR below the threshold values of 6.3 %, 8.5 %, and 18.3 % to reverse the impact of compliance.

Radiotherapy only improves cancer-specific survival of the cohort if there is a large difference in LRR with versus without it. Routine treatment may therefore be inferior to a tailored radiotherapy regimen.