Page 114 - Read Online

Best et al. Hepatoma Res 2020;6:62 I http://dx.doi.org/10.20517/2394-5079.2020.56 Page 5 of 13

In order to achieve a sensitivity superior to the combination of US and AFP while preserving high
specificity in early HCC detection, a large number of studies examined the suitability of different
biomarkers, such as Glypican-3, AFP-L3, and DCP for HCC surveillance [37-40] . The focus during recent years
has been on validation of AFP, AFP-L3, and DCP which are already available as certified laboratory tests.

AVAILABLE HCC-BIOMARKERS AND DIAGNOSTIC MODELS
AFP and AFP-L3
The overall AFP, which is usually determined in the clinical routine, consists of the three different isoforms:
AFP-L1, AFP-L2, and AFP-L3. AFP is a fetal glycoprotein that can be produced later in life when the
[41]
hepatocytes are in the process of malignant transformation . According to various studies, the sensitivity
[42]
of AFP in HCC detection is between 39% and 65% and the specificity varies between 76% to 94% .
AFP-L3 is a AFP variant that binds to the lectin molecule “Lens culinaris agglutinin” and, in contrast to the
overall AFP, is HCC-specific. While the AFP-L3 fraction is produced exclusively by malignant transformed
hepatocytes, an AFP-L1 elevation, the non-glycosylated AFP main fraction, can also be caused by viral
hepatitis and in this scenario is responsible for an incorrectly increased total AFP level [43,44] . With a cut-off
of 15%, sensitivities between 75%-96.6% and specificities of 90%-92% have been described for AFP-L3 [45,46] .

Des-gamma-carboxy prothrombin
Des-gamma-carboxy prothrombin (DCP), also known as Protein-Induced-by-Vitamin-K-Absence-
or-Antagonist-II (PIVKA II), is a precursor of prothrombin and is formed in the context of
hepatocarcinogenesis due to an impaired vitamin K metabolism. Here, the carboxylation of prothrombin
is so impaired that the serum concentration of the DCP increases. Sensitivities between 48% and 62% and
specificities between 81% and 98% have been described for the DCP, making DCP a more specific marker
than AFP, albeit with lower sensitivity [47-49] .

Combination of AFP, AFP-L3 and DCP
Various clinical trials have clearly demonstrated that there were no correlations among the results of AFP,
AFP-L3, and DCP. Thus, some HCC cases can be positive for only one marker at a time while negative for
the others. In clinical practice, this means that the combination of the above biomarkers leads to a gradual
increase in sensitivity.

Especially when AFP remains the only available marker in clinical routine, the complementary use of
AFP-L3 and DCP represents an additional diagnostic option. In a retrospective Japanese single-center
study with 270 AFP-negative HCC patients, it was demonstrated that the majority of patients with positive
[50]
AFP-L3 and/or DCP findings were correctly recorded .

GALAD model
The aforementioned triple combination of the biomarkers AFP, AFP-L3, and DCP demonstrated superior
detection of HCC compared to their individual utilization with no significant decrease in specificity in an
Asian patient cohort [51,52] . For further optimization, a statistically based model called GALAD score was
developed a few years ago and was extensively validated in several international studies. It is a diagnostic
algorithm based on rigorous statistical analysis. The formula is calculated on the measured absolute values
of the three markers instead of defining cut-off levels. Thus, they are considered as continuous variables
rather than categorical. Gender and age information are also included since older age and male sex
represent independent HCC risk factors [53,54] . A GALAD point value of -0.63 serves as a cut-off value for
optimal sensitivity and specificity regardless of the BCLC stage. Using this model in a British cohort, the
GALAD model achieved an overall AUROC (area under the receiver operating characteristic curve) of
0.97 in detection of all BCLC stages, and early stage HCC (BCLC 0/A) was detected with an AUROC of