What is the positive predictive value of the Guardant Shield blood test for colorectal cancer detection in average‑risk adults aged 45 years or older?

Positive Predictive Value of Guardant Shield

The positive predictive value (PPV) of Guardant Shield for advanced colorectal neoplasia is 15.5% in average-risk adults aged 45-85 years. 1

Performance Characteristics from Clinical Validation

The JAMA clinical validation study (ECLIPSE trial) of Guardant Shield enrolled 27,010 average-risk adults and demonstrated:

• PPV for advanced colorectal neoplasia (cancer or advanced precancerous lesions): 15.5% (95% CI, 14.2%-16.8%) 1
• Negative predictive value for advanced neoplasia: 90.8% (95% CI, 90.7%-90.9%) 1
• Sensitivity for colorectal cancer: 79.2% (95% CI, 68.4%-86.9%) 1
• Specificity for advanced neoplasia: 91.5% (95% CI, 91.2%-91.9%) 1

A separate validation study (ECLIPSE) reported in the New England Journal of Medicine with 7,861 evaluable participants found:

• Sensitivity for colorectal cancer: 83.1% (95% CI, 72.2%-90.3%) 2
• Specificity for advanced neoplasia: 89.6% (95% CI, 88.8%-90.3%) 2
• Sensitivity for advanced precancerous lesions: 13.2% (95% CI, 11.3%-15.3%) 2

Critical Clinical Context

This 15.5% PPV means that approximately 84-85% of positive Guardant Shield results will be false positives requiring unnecessary colonoscopy. 1 This represents a substantial burden of follow-up procedures in patients without advanced neoplasia.

Guideline Position on Blood-Based Tests

No major guideline organization—including the USPSTF, American Cancer Society, NCCN, or U.S. Multi-Society Task Force—has incorporated blood-based ctDNA tests like Guardant Shield into formal colorectal cancer screening recommendations as of 2024-2025. 3

• The American College of Physicians explicitly recommends against using serum screening tests for colorectal cancer due to lack of evidence for mortality benefit. 4
• The U.S. Multi-Society Task Force specifically recommends against the Septin9 serum assay for screening due to insufficient evidence. 4, 5
• Blood-based tests lack the evidence base that supports mortality reduction, which is the primary outcome that matters for screening. 4

Comparison to Guideline-Endorsed Tests

The established tier-1 screening modalities with proven mortality reduction are:

• Colonoscopy every 10 years: Class I recommendation with high-quality evidence 3
• Annual FIT: Class I recommendation with high-quality evidence 3

These tests represent the only screening modalities with demonstrated mortality reduction in randomized trials. 3, 5

Clinical Implications

The low PPV of 15.5% creates a substantial false-positive burden: For every 100 positive Guardant Shield tests, approximately 85 patients will undergo colonoscopy without finding advanced neoplasia. 1

The extremely low sensitivity for advanced precancerous lesions (12.5-13.2%) means the test misses approximately 87% of advanced adenomas, which are the primary targets for cancer prevention through screening. 1, 2

A randomized trial demonstrated that offering blood-based testing increased screening adherence by 17.5 percentage points (30.5% vs 13.0%) in patients who had previously declined FIT. 6 However, this adherence benefit must be weighed against the substantially lower detection of advanced adenomas compared to colonoscopy or FIT. 6

Common Pitfalls to Avoid

• Never use blood-based tests like Shield in symptomatic patients with alarm symptoms (rectal bleeding, narrowed stools, unexplained weight loss, change in bowel habits)—these patients require immediate diagnostic colonoscopy regardless of any screening test results. 4
• All positive blood-based tests require follow-up colonoscopy—failure to complete diagnostic workup renders the screening program ineffective. 4
• Do not use blood-based tests as a substitute for guideline-endorsed screening in patients who can tolerate colonoscopy or FIT, as these tests lack mortality benefit evidence. 4, 3