Association Between Multi‐Laboratory Parameters Pre‐ and Post‐Treatment and Pathological Features and Their Values in Clinical Decision in Primary Hepatic Carcinoma Authors: Xiaowei Chi, Weifang Liu, Zihan Liu, Jie Liu Published: 2026 ABSTRACT ABSTRACTBackgroundPrimary hepatic cancer (PHC) is still a significant burden worldwide. The efficacy of diagnosis, treatment response monitoring, and prognostication using laboratory biomarkers is still a challenge. This study analyzed multi‐laboratory parameters pre‐ and post‐treatment and their associations with clinical and pathological features and clinical decisions in PHC patients to evaluate their clinical usefulness.MethodsThis study analyzed laboratory data from PHC patients by comparing changes between pre‐ and post‐treatment and the efficacy of commonly used tumor markers and their associations with clinical and pathological features. Data from 441 PHC patients were analyzed.ResultsSerum alpha‐fetoprotein (AFP) and alpha‐L‐fucosidase (AFU), alanine aminotransferase (ALT), aspartate aminotransferase (AST), gamma‐glutamyltransferase (γ‐GT), indirect bilirubin (IBIL), and hepatitis B virus DNA (HBV DNA) were reduced after treatment; AFP, carcinoembryonic antigen (CEA), and cancer antigen 19–9 (CA‐199) showed value in assisting the judgment of surgical resection; serum ferritin increased significantly along with stage advanced. AFP, CEA, CA‐199, and ferritin were lower in surgical resection patients than non‐surgical patients. A significant association was displayed between AFP levels and multiple biomarkers and clinical features when AFP cut‐off was adjusted to 140 ng/mL. AFP was increased in PHC patients with chronic hepatitis B infection and liver cirrhosis, while CA‐199 was decreased in both situations.ConclusionsTumor marker AFP, CEA, CA‐199, serum chemistry test panel AFU, ALT, AST, γ‐GT, bilirubin, protein, and HBV DNA are useful indicators in monitoring therapeutic response and disease progression and may be used as indicators in assisting the judgment of surgical resection of PHC. Introduction Primary hepatic carcinoma (PHC) is still a significant issue in China and many other countries around the world [1, 2, 3, 4]. Hepatocellular carcinoma (HCC) is the most common type of PHC, followed by cholangiocarcinoma (CCA) [5, 6, 7]. The 5‐year overall survival rate of PHC remains very low at 6.28% [8]; thus, early diagnosis, treatment response monitoring, and prognosis evaluation are critical for management of the disease. AFP is a commonly used serum biomarker for PHC diagnosis and screening in high‐risk populations together with ultrasound/sonography; however, about 30%–40% of PHC are AFP negative [9, 10], and ultrasound usually lacks specificity and relies heavily on operator experience, making it problematic in diagnosis, treatment response monitoring, recurrence judgment, and prognostication of PHC. Laboratory testing, including prothrombin induced by vitamin K deficiency or antagonist‐II (PIVKA‐II, also known as Des‐gamma‐Carboxy Prothrombin, DCP), CEA, CA‐199, cancer antigen 724 (CA‐724), and chemistry panel such as alpha‐L‐fucosidase (AFU), bilirubin, serum protein, ferritin, hepatitis B surface antigen (HBsAg), hepatitis B virus DNA (HBV DNA) copy numbers, liver enzymes, and coagulation parameters are useful indicators in PHC diagnosis and management; however, the efficacy of these biomarkers in clinical use and the appropriate selection of the tests were still inconclusive [11, 12, 13, 14, 15, 16]. We previously reported the association between clinical pathological features and AFP positive and negative PHC patients, and we found that patients with HCC, larger tumor size, liver cancer with hepatic cirrhosis, portal vein thrombosis, metastasis, high Child–Pugh score, high Barcelona‐Clinic Liver Cancer (BCLC) stage, advanced clinical stage, HBsAg and HBeAg positivity, and high HBV DNA levels had significantly higher serum AFP levels. Also, patients with high serum AFP levels had higher PIVKA‐II, ALT, AST, AFU, γ‐GT, γ‐GT/ALT ratio, direct bilirubin (DBIL), indirect bilirubin (IDBIL), fibrinogen, and D‐dimer levels. Patients who were AFP positive had higher white blood cell (WBC), neutrophil, monocyte, and platelet count as well as higher neutrophil to lymphocyte ratio (NLR) [10]. With new methods emerging for liver cancer treatment, including surgical resection, chemotherapy‐transarterial chemoembolization (TACE), immuno‐therapy, thermal ablation, radiofrequency ablation or microwave ablation, targeted cancer drugs, radiotherapy, and integrative Chinese and Western medicine (ICWM) [17, 18, 19, 20], it will be important to understand the treatment related pathological features of PHC. In this study, we comprehensively analyzed the changes of laboratory parameters associated with clinical and pathological features pre‐ and post‐treatment in PHC patients from Eastern China; we also adjusted the cut‐off value of serum AFP to see its capability in helping the clinical management of PHC. Lastly, we performed further in‐depth analysis of the value of tumor markers in assisting the judgment of surgical resection of HCC. Tumor marker levels were also compared amongst treatment methods pre‐ and post‐treatments to evaluate their usefulness in treatment response monitoring. Patients and Methods Project Design This retrospective study analyzed primary hepatic carcinoma (PHC) patients admitted for treatment at the Affiliated Hospital of Shandong University of Traditional Chinese Medicine from June 2018 to December 2023. This study adhered to the Declaration of Helsinki for study involving human or human data. The Ethics Committee/Institutional Review Board of the Affiliated Hospital of Shandong University of Traditional Chinese Medicine approved this study (No. 2021–114‐KY). Informed consent to participate was waived, as determined by the Ethics Committee, because this study was a retrospective data analysis only. Four hundred and forty‐one (441) PHC patients were analyzed in the study, with 354 males (80.3%) and 87 females (19.7%) aged 41–89 years. Among patients, 143 had pre‐ and post‐treatment clinical and pathological data for comparison, of which 108 were males and 35 were females aged 41–89 years. Inclusion and exclusion criteria: patients with previously untreated PHC (hepatocellular carcinoma and cholangiocarcinoma) at the Affiliated Hospital of Shandong University of Traditional Chinese Medicine, patients diagnosed in other hospitals within 8 weeks and transferred to this hospital for treatment (421 cases), and relapsed patients (20 cases) were included. Two groups of patients were included in this study: one group with pre‐ and post‐treatment, and one group with pre‐treatment only because of lack of treatment monitoring and follow‐ups. Treatment methods included surgical resection, chemotherapy/TACE, immuno‐targeted therapy, and ICWM. Patients who were post treatment or had secondary/metastatic liver cancer, hepatitis without liver cancer, liver cancer with other malignancy, liver hemangioma, liver cirrhosis only, liver cyst, liver cancer with immunodeficiency, and liver cancer with hematologic diseases were excluded. Diagnosis and staging of PHC were based on China Guidelines for Diagnosis and Treatment of Primary Liver Cancer, 2022 Edition. Blood Sample Collection Fasting blood sample was collected on the morning following admission and after 2–6 weeks of treatments (for pre‐ and post‐treatment comparison group). Samples were delivered to the clinical laboratory for analysis within 2 h after collection. Clinical Chemistry Analysis AFU, ferritin, ALT, AST, γ‐GT, TBIL, DBIL, IBIL, total protein (TP), ALB, and homocysteine (Hcy) were analyzed on the AU5800 chemistry analyzer system (Beckman‐Coulter, Brea, CA, USA) following the manufacturer's instructions and clinical laboratory standard operating procedures. Immunoassay of Tumor Markers Hepatitis B surface antigen (HBsAg), AFP and its sub‐type AFP‐L3, CEA, CA‐199, CA‐724, and PIVKA‐II were analyzed on the Roche Cobas 8000 chemiluminescence immunoassay system (Roche, Indianapolis, IN, USA) following the manufacturer's instructions and clinical laboratory standard operating procedures. Blood Coagulation Test The prothrombin time (PT) test was performed on a STAGO‐R Evolution coagulation analyzer (Diagnostica STAGO S.A.S, Asnières sur Seine Cedex, France) following the manufacturer's instructions and clinical laboratory standard operating procedures. Real Time Quantitative Polymer Chain Reaction (qPCR) Detection of Hepatitis B Virus DNA Hepatitis B virus DNA was detected by real‐time qPCR on a Thermo Fisher Scientific ABI 7500 DNA thermo amplifier (Thermo Fisher Scientific, Waltham, MA, USA). Results were represented in copy number of the DNA reported by the analyzer. Pathological Diagnosis Pathological diagnosis of PHC was dependent on the discretion of the oncologists and surgeons; thus, not all cases were pathologically diagnosed. Tissues were sent to pathology for diagnosis upon surgical removal of tumors. Pathological diagnosis followed standard operating procedures and routine clinical practice in pathology. Statistical Analysis Statistical analysis was performed using SPSS Version 27 for Windows; the statistical analysis was conducted from June 2024 to October 2024. Specifically, Wilcoxon Signed Rank Test, McNemar Test, Mann–Whitney Test, and Related Samples and Independent‐Samples Hodges‐Lehmann were used. Data analysis was conducted by third‐party statistical analysis services. A two‐tailed analysis was adopted, and a p value < 0.05 was considered significant in difference. Results Patient Demographics Table S1 shows the patient demographics in this study. Changes of Tumor Marker Levels Pre‐ and Post‐Treatment in PHC Patients Table 1 shows that only AFP and AFU were reduced after treatment (p < 0.05). TABLE 1Changes of tumor marker levels, chemistry, and HBV DNA pre‐ and post‐treatment in PHC patients (Wilcoxon Signed Rank Test).TestNP50 (P25−P75) p‐values (Two‐tailed)Median difference (95% CI)* Pre‐treatPost‐treatAFU13834.0 (26.8–51.0)31.0 (25.0–42.0)0.000−3.0 (−4.5 to −1.5)AFP9111.6 (4.1–522.0)7.4 (3.3–56.6)0.000−36.0 (−84.6 to −1.9)AFP‐L3232.6 (< 0.9–28.3)2.0 (< 0.9–4.9)0.221−2.6 (−12.7 to 0.5)CEA772.7 (1.8–4.9)3.1 (1.9–4.6)0.527−0.1 (−0.4 to 0.2)CA‐1997227.3 (12.1–93.7)29.7 (11.7–76.1)0.722−0.5 (−6.1 to 3.3)CA‐724321.7 (1.5–5.0)1.5 (1.5–3.1)0.465−0.1 (−0.7 to 0.2)Ferritin37350 (123–817)409 (107–730)0.55117 (−53 to 91)ALT14044.5 (27.0–93.0)29.0 (20.0–55.8)0.000−16.5 (−24.5 to −11.5)AST14056.0 (38.0–127.8)38.5 (26.0–74.0)0.000−18.0 (−27.5 to −11.5)γ‐GT140149.0 (56.3–308.8)98.0 (45.0–189.0)0.000−47.5 (−68.5 to −32.5)TBIL14026.5 (16.4–62.0)24.8 (14.6–47.4)0.060−2.5 (−5.8 to 0.1)DBIL1407.9 (4.2–21.8)6.7 (3.4–20.3)0.152−0.7 (−1.9 to 0.3)IBIL14018.1 (12.5–35.1)17.2 (10.9–26.5)0.005−2.2 (−4.0 to −0.5)TP14068.0 (62.4–74.1)66.9 (61.8–72.9)0.015−1.8 (−3.3 to −0.3)ALB14035.5 (31.7–39.1)34.9 (31.4–38.9)0.440−0.4 (−1.3 to 0.6)Hcy5813.4 (10.7–16.5)11.9 (10.4–14.8)0.084−1.0 (−2.1 to 0.1)PT12813.8 (12.9–14.8)14.0 (13.5–14.9)0.0000.4 (0.2 to 0.6)HBsAg20142 (< 0.01 to > 250)> 250 (< 0.01 to > 250)0.4410 (0 to 19)HBV DNA381147 (< 500 to 42,725)< 500 (< 500 to < 500)0.003−11,601 (−25,736 to −138)Abbreviations: AFP, alpha‐fetoprotein; AFP‐L3, lectin‐reactive fraction of alpha‐fetoprotein; AFU, alpha‐L‐fucosidase; ALB, albumin; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CA‐199, cancer antigen 199; CA‐724, cancer antigen 724; CEA, carcinoembryonic antigen; DBIL, direct bilirubin; HBsAg, Hepatitis B surface antigen; HBV DNA, Hepatitis B virus DNA (virus load); Hcy, homocysteine; IBIL, indirect bilirubin; PT, prothrombin time; TBIL, total bilirubin; TP, total protein; γ‐GT, Gamma‐glutamyltransferase.*Related‐samples Hodges‐Lehmann median difference and its 95% confidence interval (95% CI). Changes of tumor marker levels, chemistry, and HBV DNA pre‐ and post‐treatment in PHC patients (Wilcoxon Signed Rank Test). Abbreviations: AFP, alpha‐fetoprotein; AFP‐L3, lectin‐reactive fraction of alpha‐fetoprotein; AFU, alpha‐L‐fucosidase; ALB, albumin; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CA‐199, cancer antigen 199; CA‐724, cancer antigen 724; CEA, carcinoembryonic antigen; DBIL, direct bilirubin; HBsAg, Hepatitis B surface antigen; HBV DNA, Hepatitis B virus DNA (virus load); Hcy, homocysteine; IBIL, indirect bilirubin; PT, prothrombin time; TBIL, total bilirubin; TP, total protein; γ‐GT, Gamma‐glutamyltransferase. Related‐samples Hodges‐Lehmann median difference and its 95% confidence interval (95% CI). Changes of Liver Function Panel and Hepatitis B Virus DNA Pre‐ and Post‐Treatment in PHC Patients Table 1 also shows that serum ALT, AST, γ‐GT, IBIL, TP, and HBV DNA were reduced after treatment while PT was increased. Capabilities of AFP, CEA, and CA‐199 in Assisting the Judgment of Surgical Resection Results in Table 2 indicate that at specific cut‐off values of AFP (cut‐off value at 62.65 ng/mL), CEA (cut‐off value at 3.42 ng/mL), and CA‐199 (cut‐off value at 50.35 U/mL), the number of cases with or without surgical resection was different, with more surgical resection cases at the lower side of the cut‐off value and no surgical resection cases at the higher side of the cut‐off value of the biomarkers. The specificities (Sp) of AFP, CEA, and CA‐199 in assisting the judgment of surgical resection are 0.534, 0.522, and 0.529, respectively, while the sensitivities (Se) for the markers are 0.744, 0.875, and 0.789, respectively. TABLE 2Capabilities of AFP, CEA, and CA‐199 in assisting the judgment of surgical resection.TestSurgical resection χ 2 p‐values (Two‐tailed)NoYesTotalAFPAFP < 62.6541 (46.6%)29 (74.4%)708.420.004AFP ≥ 62.6547 (53.4%)10 (25.6%)57Total (%)88 (100%)39 (100%)127Sp = 0.534Se = 0.744CEACEA < 3.4243 (47.8%)35 (87.5%)7818.20.000CEA ≥ 3.4247 (52.2%)5 (12.5%)52Total (%)90 (100%)40 (100%)130Sp = 0.522Se = 0.875CA‐199CA199 < 50.3540 (47.1%)30 (78.9%)7010.90.001CA199 ≥ 50.3545 (52.9%)8 (21.1%)53Total (%)85 (100%)38 (100%)123Sp = 0.529Se = 0.789Abbreviations: Se, sensitivity; Sp, specificity. Capabilities of AFP, CEA, and CA‐199 in assisting the judgment of surgical resection. Abbreviations: Se, sensitivity; Sp, specificity. Comparison of Tumor Marker Levels in Different Clinical Stages in PHC Patients Table 3 shows that although AFU, AFP, CA‐199, CA‐724, and ferritin all display statistical differences among clinical stages, only ferritin increased significantly as the stage advanced. TABLE 3Comparison of tumor marker levels in different clinical stages in PHC patients before treatment.TestStageNP50 (P25−P75) pre‐treat p‐values (Two‐tailed)** Median difference (95% CI)* AFUI12730.0 (24.0–43.0)0.001II8438.5 (28.3–55.8)III17935.0 (26.0–54.0)IV3047.0 (35.5–73.5)AFPI11718.3 (3.5–628.5)0.005II74472.5 (9.4− > 1210)III16778.9 (4.6− > 1210)IV29783.9 (7.6− > 1210)AFP‐L3I313.5 (< 0.9–48.5)0.442II1961.5 (< 0.9–285.0)III302.9 (0.9–39.6)IV212.3 (0.7–18.3)CEAI1132.8 (2.0–4.8)0.380II732.8 (1.8–4.9)III1673.3 (1.9–5.8)IV282.6 (1.7–4.9)CA‐199I11425.8 (12.4–65.5)0.000II6947.8 (24.7–130.9)III16248.6 (18.2–165.9)IV26122.0 (41.9–446.8)CA‐724I571.5 (1.4–3.1)0.001II281.5 (1.5–1.5)III972.4 (1.5–3.9)IV122.6 (1.5–5.3)FerritinIa 62234 (115–446)a 0.000ControlIIb 34401 (186–636)b −113 (−244 to −15)IIIb,c 105480 (255–902) b,c −197 (−316 to −92)IVc 14702 (300–1468) c −404 (−986 to −126)PIVKA‐III316,224 (10–20,000)0.963II84474 (16–28,607)III7615 (15–13,979)IV Note: a,b,cThere is no statistical significance among same letter labeled (p > 0.05) while there is statistical significance among different letter labeled (p < 0.05) by using the multiple comparisons, the stepwise step‐down under Independent‐Samples Jonckheere‐Terpstra Test.*Independent‐Samples Hodges‐Lehmann median difference and its 95% confidence interval (95% CI).**Kruskal‐Wallis test. Comparison of tumor marker levels in different clinical stages in PHC patients before treatment. Note: a,b,cThere is no statistical significance among same letter labeled (p > 0.05) while there is statistical significance among different letter labeled (p < 0.05) by using the multiple comparisons, the stepwise step‐down under Independent‐Samples Jonckheere‐Terpstra Test. Independent‐Samples Hodges‐Lehmann median difference and its 95% confidence interval (95% CI). Kruskal‐Wallis test. Association Between Ascites and Tumor Marker Levels in PHC Patients Table 4 shows that the presence or absence of ascites has the greatest impact on the P50 (P25−P75) of AFP in PHC patients, and the Median Difference (95% CI) is the largest. The situation of AFP‐L3, CA‐199, and ferritin is similar; there is no statistical difference in other items (p > 0.05), except that the Median Difference of PIVKA‐II (95% CI) is −2493 (−29,385 to 3777), which shows a good effect (although still not statistically significant due to the small sample size) and suggests that further investigation on a larger sample size on PIVKA‐II is needed. TABLE 4Association between ascites and tumor marker levels in PHC patients.TestAscitesNP50 (P25−P75) pre‐treatment p‐values (Two‐tailed)** Median difference (95% CI)* AFUNo19831.0 (25.0 to 43.0)Yes22541.0 (27.5 to 58.0)0.000−7.0 (−11.0 to −4.0)AFPNo18212.0 (3.3 to 414.0)Yes209416 (10 to > 1210)0.000−113 (−242 to −27)AFP‐L3No473.2 (< 0.9 to 18.6)Yes3728.3 (< 0.9 to 318.8)0.025−21.1 (−47.6 to 0.0)CEANo1832.9 (1.8 to 5.2)Yes2013.1 (1.9 to 5.2)0.732−0.1 (−0.5 to 0.3)CA‐199No17633.9 (14.5 to 95.5)Yes19856.8 (21.6 to 162.6)0.015−11.4 (−22.8 to −1.9)CA‐724No1041.5 (1.5 to 3.5)Yes921.7 (1.5 to 3.9)0.6260.0 (−0.2 to 0.1)FerritinNo112326 (155 to 626)Yes106493 (185 to 982)0.022−119 (−238 to −15)PIVKA‐IINo4498 (103 to 15,154)Yes149334 (15 to 25,820)0.574−2493 (−29,385 to 3777)*Independent‐samples Hodges‐Lehmann median difference and its 95% confidence interval (95% CI).**Mann–Whitney U test. Association between ascites and tumor marker levels in PHC patients. Independent‐samples Hodges‐Lehmann median difference and its 95% confidence interval (95% CI). Mann–Whitney U test. Association Between AFP Levels (AFP ≤ 20 vs. AFP > 20 ng/mL) and Clinical Features in HCC Patients Table 5 shows that there were no statistical associations between AFP levels and TNM and clinical stages (≤ T2 stage vs. ≥ T3 stage, stage I, II, III, and IV) (p = 0.069 and 0.052, respectively) at the AFP cut‐off of 20 ng/mL; thus, this cut‐off may not be a suitable value for clinical evaluation. Further analysis was performed; see below. TABLE 5Association between AFP levels (AFP ≤ 20 vs. AFP > 20 ng/ml) and clinical features in HCC patients.CategoryGroupAFP cut‐off (ng/ml) χ 2 p (Two‐tailed)* AFP ≤ 20AFP > 20Total (%)TNM stage≤ T231 (51.7%)29 (48.3%)60 (100%)3.310.069≥ T3103 (38.9%)162 (61.1%)265 (100%)Total134191325Clinical stageI53 (48.6%)56 (51.4%)109 (100%)7.740.052II22 (29.7%)52 (70.3%)74 (100%)III69 (41.8%)96 (58.2%)165 (100%)IV9 (31.0%)20 (69.0%)29 (100%)Total153224377*Fisher's exact test. Association between AFP levels (AFP ≤ 20 vs. AFP > 20 ng/ml) and clinical features in HCC patients. Fisher's exact test. Association Between AFP Levels (AFP ≤ 140 vs. AFP > 140 ng/mL) and Clinical Features in HCC Patients Table 6 shows that after adjusting the AFP cut‐off value to 140 ng/mL, there were significant associations between AFP levels and TNM stage, clinical stage, liver cirrhosis, family history, portal vein thrombus, portal hypertension, and ascites (p = 0.003, 0.027, 0.001, 0.010, 0.000, 0.000, and 0.000, respectively). TABLE 6Association between AFP levels (AFP ≤ 140 vs. AFP > 140 ng/ml) and clinical features in HCC patients.CategoryGroupAFP (ng/ml) χ 2 p (Two‐tailed)AFP ≤ 140AFP > 140Total (%)TNM stage≤ T241 (68.3%)19 (31.7%)60 (100%)9.080.003≥ T3124 (46.8%)141 (53.2%)265 (100%)Total165160325Clinical stageI66 (60.6%)43 (39.4%)109 (100%)9.220.027II30 (40.5%)44 (59.5%)74 (100%)III85 (51.5%)80 (48.5%)165 (100%)IV11 (37.9%)18 (62.1%)29 (100%)Total192185377SexM160 (51.6%)150 (48.4%)310 (100%)0.1250.724F36 (49.3%)37 (50.7%)73 (100%)Total196187383Liver cirrhosisNo100 (61.7%)62 (38.3%)162 (100%)11.70.001Yes95 (44.0%)121 (56.0%)216 (100%)Total195183378CHBNo90 (53.6%)78 (46.4%)168 (100%)1.510.219Yes70 (46.7%)80 (53.3%)150 (100%)Total160158318Family historyNo132 (52.8%)118 (47.2%)250 (100%)6.650.010Yes16 (32.7%)33 (67.3%)49 (100%)Total148151299MetastasisNo114 (51.4%)108 (48.6%)222 (100%)0.0070.935Yes82 (50.9%)79 (49.1%)161 (100%)Total196187383PVTNo157 (65.4%)83 (34.6%)240 (100%)54.50.000Yes37 (26.2%)104 (73.8%)141 (100%)Total194187381PHNo154 (57.2%)115 (42.8%)269 (100%)14.70.000Yes40 (35.7%)72 (64.3%)112 (100%)Total194187381AscitesNo112 (65.1%)60 (34.9%)172 (100%)29.90.000Small amount26 (53.1%)23 (46.9%)49 (100%)Medium36 (33.0%)73 (67.0%)109 (100%)Large amount21 (41.2%)30 (58.8%)51 (100%)Total195186381Number of tumorsSingle108 (55.7%)86 (44.3%)194 (100%)3.790.052Multiple84 (45.7%)100 (54.3%)184 (100%)Total192186378Abbreviations: CHB, chronic hepatitis B; PH, portal hypertension; PVT, portal vein thrombosis. Association between AFP levels (AFP ≤ 140 vs. AFP > 140 ng/ml) and clinical features in HCC patients. Abbreviations: CHB, chronic hepatitis B; PH, portal hypertension; PVT, portal vein thrombosis. Association Between AFP Levels (AFP ≤ 140 vs. AFP > 140 ng/mL) and Chemistry Tests and HBV DNA in HCC Patients Table 7 shows that after adjusting the AFP cut‐off value to 140 ng/mL, there were significant associations between AFP levels and age, ALT, AST, γ‐GT, total bilirubin, direct bilirubin, indirect bilirubin, albumin, AFU, AFP‐L3, PT, HBsAg, and HBV DNA (p < 0.05 for all). There were no significant associations between AFP levels and total protein, Hcy, CEA, CA‐199, CA‐724, ferritin, and PIVKA‐II (p > 0.05 for all). TABLE 7Association between AFP levels (AFP ≤ 140 vs. AFP > 140 ng/ml) and chemistry tests and HBV DNA in HCC patients.TestAFP (ng/ml)NP50 (P25−P75) Z value p value (Two‐tailed)ALTAFP ≤ 14019532.0 (20.0–71.0)−3.500.000AFP > 14018444.0 (28.0–81.3)ASTAFP ≤ 14019642.0 (27.3–80.0)−7.520.000AFP > 14018593.0 (49.0–187.5)γ‐GTAFP ≤ 14019599.0 (44.0–248.0)−5.520.000AFP > 140184197.0 (113.3–323.8)TBILAFP ≤ 14019522.1 (14.7–39.9)−5.230.000AFP > 14018434.7 (21.7–75.1)DBILAFP ≤ 1401956.4 (3.6–11.6)−5.760.000AFP > 14018412.5 (6.5–38.9)IBILAFP ≤ 14019515.0 (11.0–26.8)−4.870.000AFP > 14018422.9 (14.6–39.8)TPAFP ≤ 14019567.4 (61.0–72.7)−0.1550.877AFP > 14018267.1 (60.7–72.6)ALBAFP ≤ 14019535.7 (30.8–39.1)−2.380.017AFP > 14018333.6 (29.6–38.0)AFUAFP ≤ 14019630.0 (24.0–40.8)−7.170.000AFP > 14018144.0 (33.0–65.0)HcyAFP ≤ 14010814.8 (11.7–19.0)−0.9500.342AFP > 1406016.2 (11.4–20.4)AFP‐L3AFP ≤ 140410.9 (0.9–2.9)−6.870.000AFP > 1403983.3 (23.6–677.5)CEAAFP ≤ 1401892.9 (1.9–5.2)−0.0250.980AFP > 1401743.1 (1.9–5.0)CA‐199AFP ≤ 14018536.4 (13.9–146.5)−0.9190.358AFP > 14016846.8 (22.0–126.0)CA‐724AFP ≤ 1401091.8 (1.5–3.8)−1.180.240AFP > 140791.5 (1.5–3.3)FerritinAFP ≤ 140116341 (153–626)−1.170.242AFP > 14092389 (180–958)PIVKA‐IIAFP ≤ 14010293 (14–12,576)−1.370.172AFP > 140716,224 (382–30,000)PTAFP ≤ 14019313.9 (13.0–14.9)−3.860.000AFP > 14018314.6 (13.6–15.9)HBsAgAFP ≤ 140179100 (0− > 250)−3.440.001AFP > 140167251 (9− > 250)HBV DNAAFP ≤ 140120< 500 (< 500–5460)−2.400.016AFP > 1401311500 (< 500–47,600) Association between AFP levels (AFP ≤ 140 vs. AFP > 140 ng/ml) and chemistry tests and HBV DNA in HCC patients. Comparison of Tumor Marker Levels in PHC Patients With‐ or Without‐Surgical Resection Table S2 shows that AFP, CEA, CA‐199, and ferritin in the surgical resection group of PHC patients were significantly lower than in the non‐surgical resection group (p < 0.05 for all). Tables S3 and S4 show that AFP and AFU levels (especially AFP) were reduced after treatment in both the surgical and non‐surgical resection groups (p < 0.05 for all). Association Between Liver Cirrhosis and Tumor Marker Levels Table S5 shows that AFP was increased in PHC patients with liver cirrhosis (p = 0.000) while CA‐199 was decreased (p = 0.001). Association Between Chronic Hepatitis B Infection and Tumor Marker Levels Similar to with liver cirrhosis, AFP was increased in PHC patients with chronic hepatitis B infection (p = 0.011) while CA‐199 was decreased (p = 0.000) (Table S6). Discussion Evaluation of diagnosis, management, and prognostication indicators/parameters for hepatic carcinoma is of great value in patient care and cost reduction. This work further analyzed multi‐laboratory parameters in association with clinical pathological features of PHC patients pre‐ and post‐treatment based on previous works; the study was also a preliminary evaluation of laboratory parameters in PHC management. The findings of this study are discussed below. When comparing pre‐ and post‐treatment, results indicated that only tumor markers AFP and AFU were significantly reduced after treatment. This suggests that AFP and AFU are valuable indicators for monitoring the treatment response, which is associated with overall survival of HCC patients [21, 22]. There was a suggestion that when post‐operative AFP response is incomplete in high‐AFP HCC patients at the first follow‐up, subsequent enhanced recurrence surveillance and treatments against recurrence should actively be considered [23]. AFU is considered a valuable supplementary marker in HCC early detection, especially in combination with AFP [24]. Preoperative serum AFU is also a prognostic predictor of HCC [22]. Our results indicated that serum liver function tests ALT, AST, γ‐GT, IBIL, and HBV DNA were valuable indicators in monitoring the treatment response of PHC, since these parameters were all reduced after treatment. Serum total protein was decreased after treatment, probably due to treatment (hepatic resection, chemotherapy, and immunotherapy) related consequences [25]. We found that prothrombin time was increased. The coagulation system is aberrant in liver cancer; patients with prolonged prothrombin time levels had significantly poorer overall survival and disease‐free survival rates [26, 27]. Our previous work also found that HCC patients with higher serum AFP levels had higher D‐dimer levels [10]. The interpretation for this phenomenon is that since liver parenchymal cells produce prothrombin, damage to hepatic cells altered the production of PT [28]. Post‐treatment elongation of prothrombin time may also be attributed to therapeutic consequences. Results indicated that at specific cut‐off values of AFP, CEA, and CA‐199, the numbers of cases with or without surgical resection were different, with more surgical resection cases at the lower side of the cut‐off value and no surgical resection cases at the higher side of the cut‐off value of the biomarkers; therefore, a cut‐off of biomarker levels could assist in the decision of whether surgical resection should be performed. Taking AFP as an example, a cut‐off value of 62.65 ≈ 63 could be used as a judgment point for surgical resection. One study suggested that pre‐operative AFP level could be used as an indicator of liver resection margin with caution as about 30%–40% of hepatic carcinoma patients are AFP negative; thus, different resection margins could be determined by different serum AFP levels in AFP‐increased patients [29]. Again, further analysis found that AFP, CEA, CA‐199, and ferritin in the surgical resection group of PHC patients were lower than in the non‐surgical resection group; however, AFP and AFU levels (especially AFP) were reduced after treatment in both surgical and non‐surgical resection groups. This further suggested that monitoring serum AFP, AFU, and ferritin is useful in helping the estimation of treatment response of PHC. It also suggests that surgical resection is still the ideal method of treatment whenever possible. Although AFU, AFP, CA‐199, CA‐724, and ferritin all showed statistical differences among clinical stages, only ferritin increased consistently along with the stage advancement, which suggests that ferritin is more closely associated with clinical stage. One study found that combinatory analysis of serum ferritin and AFP could increase the diagnostic power of HCC and staging. Another study showed that serum ferritin analysis could be helpful in detecting HCC without elevated AFP [30]. Yet another report showed that serum ferritin was an independent marker for predicting the survival of patients with advanced hepatobiliary cancers [31]. Results indicated that ascites were significantly associated with AFP levels in PHC patients. Similarly, AFP‐L3, CA‐199, and ferritin also displayed an association with ascites. Again, due to the small sample size in PIVKA‐II analyzed, we were unable to draw a conclusion on PIVKA‐II results; thus, further investigation is necessary. When the cut‐off was set at 20 ng/mL, AFP did not show a significant association with TNM or clinical stages; thus, this cut‐off may not be suitable for clinical evaluation. After adjusting the cut‐off of AFP to 140 ng/mL, there was a significant association between AFP levels and TNM stage, clinical stage, liver cirrhosis, portal vein thrombus (PVT), portal hypertension (PH), ascites, serum chemistry panel (ALT, AST, γ‐GT, TBIL, DBIL, IBIL, ALB, and AFU), AFP‐L3, PT, HBsAg, and HBV DNA. These results were consistent with our previous report in high AFP patients [10]. AFP was increased in PHC patients with liver cirrhosis; this result was also consistent with our previous report [10]. Surprisingly, we found that CA‐199 decreased in liver cirrhosis patients; however, other studies showed that CA‐199 increased in liver cirrhosis patients [32, 33, 34]. The reason for this phenomenon is unknown; thus, further investigation is needed. Some reports mentioned the interference of the immunoassay leading to false increments [35, 36, 37], so it could also be interfered with by other factors. Similar to in liver cirrhosis, AFP was increased in PHC patients with chronic hepatitis B infection; this result was consistent with our previous report [10]; meanwhile, CA‐199 was decreased. Again, we do not know the mechanism at present for the decrease of CA‐199. Conclusions With 441 PHC patients observed, 143 cases were with pre‐ and post‐treatment comparison; only AFP and AFU were observed to be reduced after treatment, thus those can be used as treatment response monitoring indicators. In addition, serum chemistry test panel AFU, ALT, AST, γ‐GT, bilirubin, protein, as well as the HBV DNA are useful indicators in monitoring the treatment response and disease progression. AFP, CEA, and CA‐199 could have potential value in assisting the judgment of surgical resection of PHC. Adjusting the AFP cut‐off value to 140 ng/mL showed significantly increased clinical value in PHC management. Limitations of This Study This study only analyzed PHC from one hospital in eastern China, and the sample size was small; additionally, only 16 cases of cholangiocarcinoma patients were obtained in this analysis. There were only 143 patients with pre‐ and post‐treatment comparisons due to lack of follow‐up data. For tumor marker PIVKA‐II, there were only 18 cases with this test due to the test not being commonly used in the hospital, so we were unable to draw any conclusions on this biomarker. We were unable to analyze the association between clinical features and histological types of hepatic carcinoma because the data of pathological diagnosis were incomplete; pathological diagnosis is not always required under Chinese Guidelines. We were unable to correlate the tumor size by MRI with clinical and pathological features due to the MRI data being incomplete. The alcohol consumption was arbitrarily defined as “Yes” or “No” without a specific amount of alcohol consumed, so the analytical results were omitted. Author Contributions Xiaowei Chi designed the project and performed data collection, Weifang Liu collected data and patient information, Zihan Liu provided literature and data organization, and Jie Liu was project manager and corresponding author. Funding The authors have nothing to report. Ethics Statement This study adhered to the Declaration of Helsinki for study involving human or human data. The Ethics Committee/Institutional Review Board of the Affiliated Hospital of Shandong University of Traditional Chinese Medicine approved this study (No. 2021–114‐KY). Informed consent to participate was waived, as determined by the Ethic Committee, because this study was a retrospective data analysis only. Consent The authors have nothing to report. Conflicts of Interest The authors declare no conflicts of interest. Supporting information Table S1: Patients demographics. Table S2: Comparison of tumor marker levels in PHC patient with‐ or without‐surgical resection (Mann–Whitney Test). Table S3: Changes of AFU, AFP, CEA, and CA‐199 levels pre‐ and post‐surgical resection in PHC patients (Wilcoxon Signed Rank Test). Table S4: Changes of AFP levels in PHC patients pre‐ and post treatment (McNemar Test). Table S5: Association between liver cirrhosis and tumor marker levels. Table S6: Association between chronic hepatitis B and tumor marker levels.