Serrated polyp detection rates beyond adenoma detection rate for postcolonoscopy colorectal cancer: findings from the SCREESCO trial.

Abstract

Background: The association between serrated polyp detection rates, including proximal serrated polyp detection rate (PSPDR), and postcolonoscopy colorectal cancer (PCCRC) remains unestablished. We investigated PCCRC incidence and characteristics, and explored associations of PSPDR and adenoma detection rate (ADR) with PCCRC in colonoscopy- and fecal immunochemical test (FIT)-based screening within the Screening of Swedish Colons (SCREESCO) trial.

Methods: Data from participants who underwent colonoscopy in the colonoscopy and FIT arms were analyzed. PCCRC, defined as colorectal cancer (CRC) diagnosed >6 months after a colonoscopy that did not detect CRC, was identified through linkage with cancer registries. Associations between PSPDR and PCCRC incidence were assessed using Cox models.

Results: Of 9982 and 5875 individuals in the colonoscopy and FIT arms, PCCRC was identified in 20 and 17 individuals, with 3-year PCCRC rates of 5.9% (3/51) and 3.5% (4/115), respectively. Compared with screening-detected CRC, PCCRC more often involved right-sided location and advanced stage. The 5-year cumulative incidence of PCCRC was 0.16% (95%CI 0.10%-0.27%) and 0.06% (95%CI 0.02%-0.15%) when colonoscopies were performed by endoscopists with low vs. high PSPDR (<10% vs. ≥10%), respectively. Among endoscopists with high ADR, 5-year PCCRC incidence was 0.13% (95%CI 0.04%-0.32%) vs. 0.06% (95%CI 0.02%-0.16%) for those with low vs. high PSPDR, respectively. Within this group, PCCRC incidence over 8 years decreased as PSPDR increased (hazard ratio per percentage-point increase, 0.90; 95%CI 0.82-1.00; P = 0.04).

Conclusions: Despite the limited number of PCCRC events, this study suggests that PSPDR provides complementary information beyond ADR for identifying colonoscopies with higher PCCRC incidence.