Researchers have noticed that two proteins may be key to screening for early-stage liver cancer


Researchers have identified an improved method of screening high-risk patients for one of the most common types of liver cancer, hepatocellular carcinoma.  (HCC), particularly if the patient has a history of hepatitis. By changing the threshold of one commonly used  screening test and adding a second, complementary test, researchers were able to accurately identify more early stage HCC cases. There is currently no standard or routine screening test for HCC, but high-risk patients (i.e. those with a history of hepatitis infection or cirrhosis) are often tested for high levels of Alpha-fetoprotein (AFP), a protein which is usually produced by a fetus, and in adults can indicate the presence of HCC. Blood levels of AFP can be elevated for other reasons, however, such as pregnancy, hepatitis, and other forms of cancer.

A multi-institutional team of researchers from the NCI's Early Detection Research Network (EDRN) compared the efficacy of AFP with two other proteins, des-gamma carboxyprothrombin (DCP) and lectin-bound alpha-fetoprotein (AFP-L3), which are also associated with liver cancer. They found that AFP was the best choice for finding early stage HCC. In addition, they found that screening for the presence of both AFP and DCP together was best when screening for early stage HCC that is associated with hepatitis infection.


A total of 836 patients were enrolled: 417 (50%) were cirrhosis controls and 419 (50%) were HCC cases, of which 208 (49.6%) had early stage HCC (n = 77 very early, n = 131 early). AFP had the best area under the receiver operating characteristic curve (0.80, 95% confidence interval [CI]: 0.77–0.84), followed by DCP (0.72, 95% CI: 0.68–0.77) and AFP-L3% (0.66, 95% CI: 0.62–0.70) for early stage HCC. The optimal AFP cutoff value was 10.9 ng/mL leading to a sensitivity of 66%. When only those with very early HCC were evaluated, the area under the receiver operating characteristic curve for AFP was 0.78 (95% CI: 0.72–0.85) leading to a sensitivity of 65% at the same cutoff.