Positive Predictive Value of Guardant Shield
The positive predictive value (PPV) of Guardant Shield for advanced colorectal neoplasia is 15.5%, meaning that among individuals who test positive, approximately 1 in 6 will actually have colorectal cancer or advanced precancerous lesions on subsequent colonoscopy. 1
Key Performance Metrics
The Guardant Shield blood test demonstrates the following validated performance characteristics in average-risk screening populations:
- PPV for advanced colorectal neoplasia: 15.5% (95% CI, 14.2%-16.8%), indicating that 378 out of 2,443 positive tests corresponded to advanced neoplasia on colonoscopy. 1
- Sensitivity for colorectal cancer: 79.2% (95% CI, 68.4%-86.9%), detecting 57 of 72 cancers in the validation cohort. 1
- Specificity for advanced neoplasia: 91.5% (95% CI, 91.2%-91.9%), with 22,306 of 24,371 individuals without advanced disease testing negative. 1
- Negative predictive value: 90.8% (95% CI, 90.7%-90.9%), meaning that a negative test reliably excludes advanced disease in 9 out of 10 cases. 1
Clinical Context and Interpretation
The relatively low PPV reflects the low prevalence of colorectal cancer in average-risk screening populations, not poor test performance. In the validation study of 27,010 participants, only 72 had colorectal cancer (0.27% prevalence), and 2,567 had advanced precancerous lesions (9.5% prevalence). 1 When disease prevalence is low, even highly specific tests will generate more false positives than true positives.
A separate validation study (ECLIPSE) reported similar findings:
- PPV cannot be directly calculated from the published data, but the test demonstrated 83.1% sensitivity for colorectal cancer (95% CI, 72.2%-90.3%) and 89.6% specificity for advanced neoplasia (95% CI, 88.8%-90.3%). 2
- Sensitivity for advanced precancerous lesions was only 13.2% (95% CI, 11.3%-15.3%), indicating that the test misses approximately 87% of advanced adenomas. 2
Comparison to Established Screening Modalities
Colonoscopy every 10 years and annual FIT remain the only tier-1 screening tests with proven mortality reduction in randomized trials. 3 The American Cancer Society and U.S. Multi-Society Task Force endorse these as first-line options for average-risk adults beginning at age 45. 4, 3
Blood-based tests like Guardant Shield are not currently included in guideline-endorsed screening algorithms:
- The NCCN explicitly states that the FDA-approved blood test for circulating methylated SEPT9 DNA is inferior to colonoscopy, FIT, or FIT-DNA for detecting colorectal cancer or advanced adenoma and is not endorsed as a primary screening tool. 3
- No major guideline organization (USPSTF, ACS, NCCN, U.S. Multi-Society Task Force) has incorporated blood-based ctDNA tests into their formal recommendations as of 2024-2025. 4, 3, 5, 6
Practical Implications
Every positive Guardant Shield test requires follow-up colonoscopy to complete the screening process. 1 Given the 15.5% PPV, approximately 84.5% of positive tests will lead to colonoscopies that reveal no advanced neoplasia—representing false positives that expose patients to the risks, costs, and inconvenience of an unnecessary invasive procedure.
The test's primary clinical utility may lie in improving adherence among patients who refuse standard screening options:
- In a randomized trial of 2,004 patients who had declined FIT in the prior 3-9 months, offering Guardant Shield increased screening completion by 17.5 percentage points (30.5% vs. 13.0%; OR 2.94,95% CI 2.34-3.70). 7
- In two Appalachian primary care clinics, adding blood-based testing as an option increased overall screening rates substantially, with 134 of 163 screened patients (82%) choosing the blood test over standard-of-care options. 8
Critical Limitations
The test's poor sensitivity for advanced precancerous lesions (12.5%-13.2%) undermines the prevention goal of colorectal cancer screening. 1, 2 Colonoscopy and FIT detect and enable removal of adenomatous polyps before they progress to cancer, whereas Guardant Shield misses approximately 87% of these lesions. 2
The low PPV means that for every true positive, approximately 5-6 patients will undergo unnecessary colonoscopy. 1 This trade-off may be acceptable for patients who would otherwise refuse all screening, but it represents a significant burden compared to colonoscopy (which is both diagnostic and therapeutic) or FIT (which has higher PPV when used annually). 3, 9