Management and prognosis of locally recurrent rectal cancer - A national population-based study.

Background: The rate of local recurrence of rectal cancer (LRRC) has decreased but the condition remains a therapeutic challenge. This study aimed to examine treatment and prognosis in patients with LRRC in Sweden. Special focus was directed towards potential differences between geographical regions and time periods.

Method: All patients with LRRC as first event, following primary surgery for rectal cancer performed during the period 1995-2002, were included in this national population-based cohort-study. Data were collected from the Swedish Colorectal Cancer Registry and from medical records. The cohort was divided into three time periods, based on the date of diagnosis of the LRRC.

Results: In total, 426 patients fulfilled the inclusion criteria. Treatment with curative intent was performed in 149 patients (35%), including 121 patients who had a surgical resection of the LRRC. R0-resection was achieved in 64 patients (53%). Patients with a non-centrally located tumour were more likely to have positive resection margins (R1/R2) (OR 5.02, 95% CI:2.25-11.21). Five-year survival for patients resected with curative intent was 43% after R0-resection and 14% after R1-resection. There were no significant differences in treatment intention or R0-resection rate between time periods or regions. The risk of any failure was significantly higher in R1-resected patients compared with R0-resected patients (HR 2.04, 95% CI:1.22-3.40).

Conclusion: A complete resection of the LRRC is essential for potentially curative treatment. Time period and region had no influence on either margin status or prognosis.