OPTIMAL COLORECTAL CANCER SCREENING FOR A HIGHER-RISK POPULATION: THE CASE OF NORTHEASTERN PENNSYLVANIA

Purpose: In northeast Pennsylvania (a six-county area surrounding Scranton and Wilkes-Barre), both colorectal cancer (CRC) incidence and mortality are approximately 25% higher than the corresponding US rates. The high incidence of CRC in the community has resulted in local clinicians recommending repeat colonoscopy (COL) at more frequent intervals, rather than the recommended 10-year interval. Working with a key stakeholder in that region, we sought to determine if more frequent colonoscopy screening is cost-effective for a higher-risk population.

Method: We used two microsimulation models that were developed as part of the Cancer Intervention and Surveillance Modeling Network (CISNET), a consortium of cancer modelers funded by the National Cancer Institute.  We simulated the outcomes of a cohort of 50-year-old hypothetical individuals undergoing COL under different risk assumptions (average risk, higher risk) and screening scenarios (15-yearly COL, 10-yearly COL, 7-yearly COL, all starting at age 50). We calculated life years saved (LYS) and the additional cost to increase the number of lifetime screening colonoscopies by one (recommendation is three screens per lifetime at ages 50, 60, 70). If adenomas are found and removed at screening, the individual is then followed with COL surveillance per guidelines until the age of 85.

Result: For an average-risk cohort, a 10-yearly COL strategy (i.e., COLs at ages 50, 60, 70) resulted in 4-6 additional life years (range reflects two models) and an added $524,000-$671,000 per 1000 screened compared with a 15-yearly COL strategy (i.e., COLs at ages 50, 65), resulting in an incremental cost-effectiveness ratio (ICER) of $98,000-$153,000 per LYS.  Adding an additional lifetime screen (i.e., screening every seven years at ages 50, 57, 64, 71) resulted in an additional 2 years of life and $412,000-$1,399,000 per 1000 screened compared with 10-yearly strategy, resulting in an ICER of $348,000-$579,000 per LYS. Repeating this analysis among a population at an increased risk of CRC (by 25%) reduced the ICERs to $73,000-$123,000 per LYS for 10-yearly vs. 15-yearly COL, and to $295,000-$468,000 per LYS for 7-yearly COL vs. 10-yearly COL.

Conclusion: The ICER for 7-yearly COL in the higher-risk population is much greater than the ICER for 10-yearly COL in the average population. For a population at 25% increased risk of CRC, screening with COL more often than 10-year intervals in not a good use of resources.