Background: Rectal cancers have long been treated as a single-entity disease; however, whether the prognosis of high rectal cancer (inferior margin located 10.1 to 15.0 cm from the anal verge) differs from that of mid/low rectal cancer (0 to 10.0 cm) remains disputed.
Methods: Patients with stages I-III rectal adenocarcinomas undergoing curative-intent surgery were enrolled between 2007 and 2013 in this retrospective analysis. Exclusion criteria were neoadjuvant therapy or concurrent cancers. Propensity score matching and Cox regression analysis were performed to compare a 5-year overall and cancer-specific survival between patients with high and mid/low rectal cancer.
Results: Of 613 patients who met the inclusion criteria, 199 (32.5%) and 414 (67.5%) had high and mid/low rectal cancer, respectively. After propensity score matching (187 cases for each group), the high group showed a better overall survival (70.9 vs. 56.9%, p = 0.042) and cancer-specific survival (77.4 vs. 60.3%, p = 0.028) at 5 years compared with the mid/low group with stage III disease. However, high rectal cancer did not demonstrate prognostic superiority in stages I-II disease. Multivariate analysis identified high tumor location as an independent prognostic factor for cancer-specific survival (hazards ratio = 0.422, 95% confidence interval 0.226-0.786, p = 0.007) and overall survival (hazards ratio = 0.613, 95% confidence interval 0.379-0.991, p = 0.046).
Conclusions: Patients with stage III high rectal adenocarcinoma demonstrated better overall and cancer-specific survival than those with mid/low type, and tumor location was an independent prognostic factor for patients with rectal carcinomas.
Keywords: High rectal cancer; Mid/low rectal cancer; Propensity score matching; Survival.