Introduction
Patients with Serrated polyps are at increased risk for colorectal cancer (CRC) compared to those without polyps at screening colonoscopy. However, the burden of the serrated pathway by subtypes of serrated polyps (hyperplastic polyps, Sessile serrated lesions/traditional serrated adenomas) to long-term CRC mortality in screening patients is not well understood.
Aims & Methods
We conducted a retrospective cohort study where we included screening participants of the Austrian Quality Certificate for Screening Colonoscopy between 2010-2020. Data on CRC deaths by primary tumor location in the cohort was obtained by database linkage to the Austrian Death registry. We assessed cumulative incidence of CRC deaths after the diagnosis of SSL/TSA, HP and conventional adenomas with deaths from other causes as competing risk. Whether polyp size in HP, SSL/TSA, adjusted for polyp location, sex and age was associated with CRC death was assessed by Cox regression.
Results
321,964 participants were included in the final analysis. 10-year Cumulative incidence of CRC death was 0.13% (95% CI 0.12-0.14%) for participants with HP, 0.16% (95% CI 0.16-0.16%) for participants with SSL/TSA and 0.25% (95% CI 0.23-0.27%) after the diagnosis of conventional adenomas. 1,271 (95% CI 2.71%) HP were ≥10 mm, as compared to 1,354 (95% CI 18.65%) SSL/TSA and 9,576 (95% CI 13.26%) conventional adenomas. When adjusted for polyp location, the association of polyp size ≥10 mm with CRC death was of similar magnitude in participants with conventional adenomas (HR 3.69, 95% CI 2.45 - 5.55), SSL/TSA (HR 3.24, 95% CI 1.35 - 7.79) and HP (HR 5.08, 95% CI 2.42 - 10.65). Participants who had hyperplastic polyps at screening mostly died of cancers located in the distal colon (62.1%, n = 18), while only 10.3% (n = 3) of CRC deaths after the detection of hyperplastic polyps were of proximal tumors. In participants with SSL/TSA, 42.9% (n = 3) of deaths were from proximal tumors, only 14.3% (n = 1) were located in the sigmoid colon or rectum.
Conclusion
Across all histologic types, participants with a polyp sized ≥10 mm have higher hazards for CRC death after screening colonoscopy compared to polyps <10 mm. Although only few participants had large HP detected at screening, polyp size is associated with CRC mortality in participants with HP, when adjusting for of polyp location.