To determine the effectiveness of surveillance colonoscopy (SC) and optimize its use by assessing real-world surgically resected cases of ulcerative colitis (UC)-associated colorectal cancer (CRC) and dysplasia.
Clinicopathological data of 406 (238 CRC and 168 dysplasia) patients who underwent surgical resection in 10 UC specialized institutions were retrospectively reviewed. The overall survival (OS) rates were compared between the SC and non-SC groups. The incidence of and risk factors for early-onset CRC (<8 years after UC onset) were identified. The distribution of CRC lesions was also assessed.
Cancer stages were significantly more advanced in the non-SC group than in the SC group (P < 0.001). The patients in the SC group showed significantly better OS than those in the non-SC group (5-year OS: 89% vs 70%; log-rank test: P = 0.001). Seventeen percent of patients developed CRC within 8 years after UC onset. The age at UC onset was a risk factor and a good predictor of early-onset CRC (<8 years) (P < 0.01; AUC: 0.85). The most common sites of CRC were the rectum (51%) and sigmoid colon (20%). Multiple CRC was identified in 16% of patients.
Surveillance colonoscopy was effective and improved the OS in patients with UC. We recommend that patients with late-onset UC (>40 years) undergo SCs earlier because of the high incidence of CRC within 8 years of UC onset. Moreover, the rectum and sigmoid colon should be more thoroughly examined.