Value of combined measurement of alpha-fetoprotein and alpha-fetoprotein L3% in the diagnosis of hepatocellular carcinoma

Objective To investigate the value of combined measurement of alpha-fetoprotein ( AFP) and AFP-L3% in the diagnosis of hepatocellular carcinoma ( HCC) . Methods A retrospective analysis was performed for serological parameters and clinical data of 1208 patients with chronic hepatitis, 1967 patients with liver cirrhosis, and 1569 patients with primary HCC who attended The Fifth Medical Center of Chinese PLA General Hospital from January 2010 to December 2017. The distribution of serum AFP and AFP-L3% was analyzed, and AFP alone, AFP-L3% alone, and AFP combined with AFP-L3% were compared in terms of sensitivity, specificity, and area under the receiver operating characteristic curve ( AUC) . The correlation of AFP with AFP-L3 level and AFP-L3% was also analyzed. A one-way analysis of variance or the Kruskal-Wallis H test was used for comparison of continuous data between groups, and the chi-square test was used for comparison of categorical data between groups. The Spearman rank correlation test was used for correlation analysis of continuous data, the Delong test was used for comparison of AUC, and logistic regression was used to establish the model of AFP combined with AFP-L3% for the diagnosis of HCC. Results The HCC group had significantly higher serum levels of AFP and AFP-L3 than the liver cirrhosis group and the chronic hepatitis group ( H = 1107. 3 and 1076. 9, both P < 0. 01) . The HCC group had a significantly higher positive rate of AFP-L3% than the liver cirrhosis group and the chronic hepatitis group ( 30. 9% vs 3. 0%/2. 8%, P < 0. 01) . The HCC group had a significantly higher level of AFP-L3 than the liver cirrhosis group and the chronic hepatitis group ( 3. 3 ( 0-13. 4) % vs 0/0, H = 1034. 1, P <0. 01) . The correlation analysis showed that serum AFP was significantly correlated with AFP-L3 level ( r = 0. 91, P < 0. 01) , and there was no correlation between AFP-L3% and AFP ( r = 0. 04, P > 0. 05) . With specificity remaining the same, AFP combined with AFP-L3% had a significantly higher sensitivity in the diagnosis of HCC than AFP alone ( 26. 6% vs 18. 4%, P < 0. 01) or AFP-L3% alone ( 36. 6% vs 30. 9%, P < 0. 01) . The subgroup analysis showed that for the samples with AFP ≥10 μg/L, AFP combined with AFP-L3%had a significantly higher AUC in the diagnosis of HCC than AFP alone ( 0. 745 vs 0. 701, P < 0. 01) ; for the tumor with a diameter of >2 cm or hepatitis B-related liver cancer, AFP combined with AFP-L3% had a significantly higher sensitivity than AFP alone ( 27. 6% vs18. 0%/27. 7% vs 19. 1%, both P < 0. 01) or AFP-L3% alone ( 37. 0% vs 30. 3%/36. 9% vs 31. 8%, both P < 0. 01) . For the group with a tumor diameter of ≤2 cm and the group without hepatitis B-related liver cancer, there was no significant difference in sensitivity between AFP combined with AFP-L3% and AFP or AFP-L3% alone in the diagnosis of HCC ( all P > 0. 05) . Conclusion AFP combined with AFP-L3% can significantly improve the sensitivity of HCC diagnosis and has a good clinical value in the diagnosis of hepatitis B-related liver cancer and liver cancer with a tumor diameter of > 2 cm.