What is the reliability of CA (Cancer Antigen) 19.9 as a tumor marker in patients with suspected pancreatic cancer who are Lewis antigen negative?

CA 19-9 Reliability in Lewis Antigen-Negative Patients

CA 19-9 is unreliable in Lewis antigen-negative patients (5-10% of the population) because they cannot synthesize this sialylated Lewis-a blood group antigen, rendering the test undetectable or falsely negative even in the presence of pancreatic cancer. 1, 2

Why Lewis-Negative Status Matters

• CA 19-9 is a sialylated Lewis-a blood group antigen that requires functional Lewis antigen expression for synthesis and secretion. 1
• Approximately 5-10% of the population is Lewis antigen-negative (genotype Le(a-b-)) and cannot produce CA 19-9, making the test ineffective for cancer detection in these individuals. 2, 3, 4
• The National Comprehensive Cancer Network explicitly states that CA 19-9 will be undetectable in Lewis antigen-negative individuals, which creates false-negative results even when pancreatic cancer is present. 1, 5

Emerging Evidence: Not All Lewis-Negative Patients Are Non-Secretors

• Recent data challenges the traditional understanding: A 2018 study found that only 41.9% of Lewis-negative pancreatic cancer patients had CA 19-9 values ≤2 U/mL, and surprisingly, 27.4% had elevated CA 19-9 (>37 U/mL). 5
• The diagnostic utility in Lewis-negative patients showed an area under the ROC curve of 0.842, approaching the performance in all patients (0.898), suggesting CA 19-9 may retain some utility even in Lewis-negative genotypes. 5
• However, this contradicts established guidelines and requires validation before changing clinical practice. 1, 2

Clinical Implications for Lewis-Negative Patients

When CA 19-9 Cannot Be Relied Upon:

• Do not use CA 19-9 for diagnosis, prognosis, or monitoring in confirmed Lewis-negative patients using traditional interpretation. 2, 3, 4
• Lewis-negative status is an independent poor prognostic factor (HR 1.30,95% CI 1.03-1.64) in pancreatic cancer, separate from CA 19-9 levels. 5

Alternative Approaches:

• Consider alternative tumor markers such as Dupan-2, which shows significantly higher serum levels in Lewis-negative individuals and can detect pancreatic cancer when CA 19-9 fails. 6
• Dupan-2 demonstrated the highest frequency of elevated levels in CA 19-9-negative patients, particularly those with Lewis-negative phenotypes. 6
• CA50 and Span-1 are also Lewis-dependent and will be falsely low in Lewis-negative patients, making them unsuitable alternatives. 6
• Sialyl SSEA-1 levels are independent of Lewis blood group phenotype but have lower sensitivity (51%) for pancreatic cancer. 6

Critical Pitfalls to Avoid

• Never assume a normal CA 19-9 rules out pancreatic cancer without confirming Lewis antigen status, as 5-10% of patients cannot produce this marker. 2, 3, 4
• Do not delay imaging or biopsy based on normal CA 19-9 in symptomatic patients without knowing Lewis status. 2, 7
• CA 19-9 measurements should be performed after biliary decompression when obstruction is present, as false-positive elevations occur in 10-60% of benign biliary obstruction cases. 1, 3, 4
• Different CA 19-9 testing methods cannot be compared directly, so serial monitoring must use the same laboratory methodology. 1, 7

Practical Algorithm for Suspected Pancreatic Cancer

1. Obtain high-quality cross-sectional imaging (contrast-enhanced CT or MRI with MRCP) as the primary diagnostic modality, regardless of CA 19-9 levels. 2, 7
2. Measure CA 19-9 only after biliary decompression if obstruction is present to avoid false-positive results. 1, 2
3. If CA 19-9 is unexpectedly normal in a patient with imaging-confirmed pancreatic cancer, consider Lewis antigen testing (FUT3 genotyping) to identify false-negative results. 5
4. In confirmed Lewis-negative patients with suspected pancreatic cancer, rely on imaging and tissue diagnosis rather than tumor markers, and consider Dupan-2 as an alternative marker. 6