Background & Aims
Blood-based biomarker tests have been developed as a non-invasive colorectal cancer (CRC) screening method to triage patients to colonoscopy. We investigated the conditions under which routine screening with a blood test could be as effective and cost-effective compared to no screening, annual fecal immunochemical testing (FIT), or decennial colonoscopy screening.
Methods
We used the three Cancer Intervention and Surveillance Modeling Network (CISNET) CRC microsimulation models to simulate the remaining lifetime of an average risk cohort of US 45-year-olds. Simulated screening regimens included no screening, annual FIT, decennial colonoscopy, or a blood test used every three years. Base-case analyses simulated a $500 blood test meeting the current minimum CMS coverage decision requirements (74% CRC sensitivity and 90% specificity). The primary outcomes were quality-adjusted life-years gained (QALYG) compared to no screening and lifetime costs, both discounted at 3% per year. Sensitivity analyses varied blood test sensitivity to CRC (from 74% to 92%), sensitivity to advanced adenomas (from 10% to 50%), test interval (1, 2, or 3 years), and costs (from $25 to $500), holding specificity constant at 90%.
Results
A blood test that only met the minimum CMS coverage requirements was less effective and more costly than either FIT or decennial colonoscopy, although it was cost-effective relative to no screening. If used every three years, the blood test saved 86-125 QALYG per 1,000 individuals relative to no screening, at a cost of $ 6,850-8,455 per person (Figure 1). Annual FIT screening yielded 107-135 QALYG per 1,000 individuals at a cost of $3,811-5,384, whereas decennial colonoscopy yielded 132-177 QALYG per 1,000 at a cost of $5,375-7,031. Increasing blood tests’ sensitivity to AA from 10% to 20% improved effectiveness more than increasing CRC sensitivity from 74% to 92% (Figure 2), but a combination of higher AA sensitivity and shorter interval was required for blood tests to match the benefit afforded by FIT. A blood test with 92% CRC sensitivity and 50% AA sensitivity yielded 117-162 QALYG per 1,000 individuals if used every three years and 133-173 QALYG if used every year. The three models agreed such a test with 50% AA sensitivity would not be cost-effective if priced above $125 per unit.
Conclusion
Blood tests that only meet minimum CMS coverage requirements should not be recommended to patients who would otherwise undergo screening by colonoscopy or FIT because this would result in a loss of benefit. In addition to meeting CMS coverage requirements, blood tests needed a combination of higher AA sensitivity (above 40% vs 10% base-case assumption) and lower costs (below $125 vs $500 base-case assumption) to become cost-effective.
Keywords: Blood-based Biomarker Tests, Colorectal Cancer Screening, Threshold Analysis, Liquid Biopsy

Figure 1. Cost-effectiveness frontiers with blood tests with varying test performance.
Figure 2. Effectiveness of blood tests as a function of test interval and sensitivity to advanced adenomas and CRC.