Colorectal cancer (CRC) screening of asymptomatic adults lowers CRC incidence by finding and removing precancerous adenomatous and serrated lesions, and lowers CRC mortality by detecting cancer at an earlier stage.^[1] The U.S. Preventive Services Task Force (USPSTF) recommends screening average-risk adults aged 45–75 years and individualizing the decision at ages 76–85 years. It does not prefer any one recommended test.^[1] A family history of CRC or advanced adenoma, Lynch syndrome, and other high-risk conditions call for earlier, colonoscopy-based screening.^[2] Any positive noninvasive screening test requires follow-up colonoscopy.^[1] Post-polypectomy surveillance and the management of hereditary polyposis syndromes are covered in their own chapters.

Average risk means no prior CRC, adenomatous polyps, or inflammatory bowel disease, and no personal or family history of a genetic syndrome that confers high lifetime CRC risk (e.g., Lynch syndrome, familial adenomatous polyposis).^[1]

The 2021 statement replaced the 2016 USPSTF recommendation, which started screening at age 50.^[3] Two findings drove the lower starting age. First, microsimulation modeling estimated 22–27 additional life-years gained per 1,000 persons when screening starts at 45 instead of 50, with few added procedural complications. Second, the incidence of early-onset CRC has risen.^[1][4] CRC is now the leading cause of cancer death in men younger than 50 years and the second leading cause in women younger than 50 years.^[5]

All noninvasive tests (gFOBT, FIT, mt-sDNA, blood-based assays) need a follow-up diagnostic colonoscopy when the result is positive. Patients who complete any recommended strategy do better than patients who are not screened at all.^[1][9]

Source: Lin et al. and USPSTF 2021.^[9][1] FIT performance depends on the brand and the positivity threshold.

Next-generation mt-sDNA. In the BLUE-C study (about 20,000 asymptomatic adults who underwent colonoscopy), the next-generation mt-sDNA test had a CRC sensitivity of 93.9%, compared with 67.3% for FIT. Its sensitivity for advanced precancerous lesions was 43.4%, compared with 23.3% for FIT (P<0.001 for both comparisons). Detection of sessile serrated lesions was also much higher (45.8% vs. 5.2%). The trade-off was lower specificity for advanced neoplasia (90.6%; P<0.001 vs. FIT), which means more false-positive results that lead to colonoscopy.^[10]

Blood-based cell-free DNA tests. The FDA has approved two blood tests for average-risk adults aged 45 years and older: Shield and SimpleScreen CRC.^[11][12] These tests meet minimum thresholds for detecting CRC. Their sensitivity for advanced precancerous lesions is much lower than that of stool-based tests, so they contribute little to cancer prevention through polyp removal.^[13] In the PREEMPT CRC validation study, CRC sensitivity was 79.2%, and sensitivity for advanced precancerous lesions did not meet its prespecified acceptance criterion.^[14]

Guideline support for blood tests is limited:

• USPSTF 2021 did not include serum tests, and ACP advises against them.^[1][8]
• Neither USPSTF nor ACG currently recommends Shield.^[11]
• The optimal screening interval for blood tests has not been established.^[11]

Blood tests are best reserved for patients who decline every other recommended modality.^[15][13]

For patients at increased risk, colonoscopy is the preferred screening strategy.^[2]

Lynch syndrome. CRC risk and age at onset differ by mismatch repair gene:^[19]

• MLH1 and MSH2 carriers: lifetime CRC risk 58–82%, mean age at diagnosis 44–61 years.
• MSH6 and PMS2 carriers: lifetime risk 10–22%, mean age at diagnosis 55–66 years.

On this basis, some schemes start surveillance later, at about age 30 for MSH6 and about age 35 for PMS2.^[19]

Refer patients with suspected Lynch syndrome for genetic counseling and testing. Suggestive features include early-onset CRC, multiple affected relatives, and synchronous or metachronous tumors.^[17] All patients with early-onset gastrointestinal cancer should have germline genetic testing.^[16]

Benefits:

• Screening reduces CRC incidence through polypectomy and reduces CRC mortality through earlier detection.^[1]
• Modeling shows that starting at age 45 adds life-years with few additional procedural complications.^[1]

Harms:

• Most procedural harm comes from colonoscopy, whether done for primary screening or to follow up a positive noninvasive test.
• Tests with lower specificity, such as mt-sDNA compared with FIT, produce more false-positive results and more follow-up colonoscopies.^[10]

• Ages 45–49: USPSTF (Grade B) and ACS recommend screening, and ACG/USMSTF issue a weak or conditional recommendation. ACP suggests clinicians consider not screening this group because the modeled net benefit is less certain.^[1][7][8]
• Older adults: ACP advises stopping screening after age 75. USPSTF and ACG favor an individualized decision at ages 76–85.^[8][1]
• Comparative effectiveness: No randomized trial compares mortality outcomes of colonoscopy with FIT or mt-sDNA. Conclusions about comparative effectiveness rely mainly on modeling and on accuracy studies done at a single time point.^[9]
• Blood-based tests: The optimal interval and the long-term effect on CRC mortality are unknown.^[13][11]
• Lynch syndrome: Gene-specific surveillance, including later starting ages for MSH6 and PMS2 carriers, has not been adopted uniformly across guidelines.^[19][17]

• A positive FIT, gFOBT, mt-sDNA, or blood-based test is not a diagnosis. Every positive result needs a diagnostic colonoscopy.^[1]
• Use only high-sensitivity gFOBT or FIT. Low-sensitivity guaiac tests are obsolete.^[1]
• Ask about family history of advanced adenoma as well as CRC, and record it at every screening encounter.^[2]
• Do not apply average-risk intervals to a patient with an FDR diagnosed before age 60. These patients need colonoscopy starting at 40, or 10 years before the relative’s age at diagnosis.^[2]
• Do not substitute a blood test for FIT or mt-sDNA in a patient who is willing to do stool-based testing, because blood tests detect far fewer advanced adenomas.^[13][15]
• MMR gene status, not family history alone, determines Lynch syndrome CRC risk and the appropriate surveillance age.^[19]