Clinical performance of AFP-L3% versus AFP as serological biomarkers for HCC (representative examples published since 2007)
| Assay | Sensitivity (%) | Specificity (%) | Other analytical features and advantages | Limitations | Reference |
|---|---|---|---|---|---|
| AFP total concentration quantification | |||||
| Multiple reaction monitoring—MS | 76.0 (optimum cut-off 6.00 ng/mL) | 77.5 (optimum cut-off 6.00 ng/mL) | - | - | [81] |
| LiBA μTAS (Wako i-30 autoanalyzer) | 59.0 (optimum cut-off 5.90 ng/mL) | 89.5 (optimum cut-off 5.90 ng/mL) | Total AFP concentration range from 0.3 ng/mL to 4,000 ng/mL | - | [81] |
| LiBA μTAS (Wako i-30 autoanalyzer) | 33.6 (cut-off 200 ng/mL) | 98.0 (cut-off 200 ng/mL) | - | - | [82] |
| AFP-L3% quantification | |||||
| LiBA μTAS (Wako i-30 autoanalyzer) | 72.4 (cut-off 7.7%) | 90.9 (cut-off 7.7%) | Sensitivity and specificity remained practically constant, independently of the AFP total concentration | - | [83] |
| Liquid chromatography—parallel reaction monitoring—MS | 80.8 (cut-off 18.6%) | 100 (cut-off 18.6%) | Early detection of HCC with higher accuracy than μTAS Analytical performance identical to μTAS conventional system with better sensitivity and accuracy | Sample preparation prior to MS analysis is required More time-consuming than microarray systems | [83] |
| Multiple reaction monitoring—MS | 81.0 (optimum cut-off 0.132%) | 89.5 (optimum cut-off 0.132%) | Early detection of HCC (screening) Lower incidence of false negatives compared to LiBA assay AFP-L3 concentration range from 0.132% to 100%, with a LLOQ of 0.051 ng/mL AFP | - | [81] |
| LiBA μTAS (Wako i-30 autoanalyzer) | 61.5 (optimum cut-off 0.500%) | 90.0 (optimum cut-off 0.500%) | AFP-L3 concentration range from 0.5% to 99.5%, with a LLOQ of 0.3 ng/mL AFP | AFP-L3% cannot be reported if the total AFP concentration is < 0.3 ng/mL, even in cases of high AFP-L3 concentrations | [81] |
| LiBA μTAS (Wako i-30 autoanalyzer) | 41.5 (cut-off 5% and total AFP concentration < 20 ng/mL) 36.2 (cut-off 5% and total AFP concentration < 10 ng/mL) | 85.1 (cut-off 5% and total AFP concentration < 20 ng/mL) 88.5 (cut-off 5% and total AFP concentration < 10 ng/mL) | Better diagnostic performance than the conventional LiBA system assay for low serum total AFP concentrations (< 20 ng/mL) AFP-L3% correlated with poor prognosis and low survival rate LLOQ of 0.3 ng/mL AFP | - | [84] |
| LiBA μTAS (Wako i-30 autoanalyzer) | 60.0 (optimum cut-off 7% and total AFP concentration 200 ng/mL) 41.1 (optimum cut-off 7% and total AFP concentration < 20 ng/mL) | 90.3 (optimum cut-off 7% and total AFP concentration 200 ng/mL) 91.9 (optimum cut-off 7% and total AFP concentration < 20 ng/mL) | More sensitive in discriminating HCC than the corresponding LiBA system and total AFP concentration assays, even for low total AFP concentrations Has predictive value for long-time survival Short-time and automated assay | - | [82] |
| LiBA (Wako LiBASys clinical autoanalyzer) | 71 (cut-off ≥ 10%) 33 (cut-off ≥ 35%) | 63 (cut-off ≥ 10%) 100 (cut-off ≥ 35%) | Total AFP concentration range from 10 ng/mL to 200 ng/mL, with a LLOQ of 0.8 ng/mL AFP and 0.5% AFP-L3 AFP-L3%, in combination with the AFP level, increases the specificity of diagnosis of HCC in individuals with indeterminate increases of total AFP level (10–200 ng/mL) | - | [27] |
-: blank cell; LLOQ: lower limit of quantification; MS: mass spectrometry; LiBA: liquid-phase binding assay; μTAS: micro-total assay system
