Relative Survival is an Adequate Estimate of Cancer-Specific Survival: Baseline Mortality-Adjusted 10-Year Survival of 771 Rectal Cancer Patients
Journal Paper/Review - Aug 2, 2013
Tarantino Ignazio, Achermann Philipp, Gueller Ulrich, Ulrich Alexis, Schmied Bruno M, Horber Daniel, Cerny Thomas, Stanga Zeno, Warschkow Rene
The objective of the present investigation is to assess the baseline mortality-adjusted 10-year survival of rectal cancer patients.
Ten-year survival was analyzed in 771 consecutive American Joint Committee on Cancer (AJCC) stage I-IV rectal cancer patients undergoing open resection between 1991 and 2008 using risk-adjusted Cox proportional hazard regression models adjusting for population-based baseline mortality.
The median follow-up of patients alive was 8.8 years. The 10-year relative, overall, and cancer-specific survival were 66.5 % [95 % confidence interval (CI) 61.3-72.1], 48.7 % (95 % CI 44.9-52.8), and 66.4 % (95 % CI 62.5-70.5), respectively. In the entire patient sample (stage I-IV) 47.3 % and in patients with stage I-III 33.6 % of all deaths were related to rectal cancer during the 10-year period. For patients with AJCC stage I rectal cancer, the 10-year overall survival was 96 % and did not significantly differ from an average population after matching for gender, age, and calendar year (p = 0.151). For the more advanced tumor stages, however, survival was significantly impaired (p < 0.001).
Retrospective investigations of survival after rectal cancer resection should adjust for baseline mortality because a large fraction of deaths is not cancer related. Stage I rectal cancer patients, compared to patients with more advanced disease stages, have a relative survival close to 100 % and can thus be considered cured. Using this relative-survival approach, the real public health burden caused by rectal cancer can reliably be analyzed and reported.