BPG is committed to discovery and dissemination of knowledge
Case Control Study
Copyright ©The Author(s) 2026.
World J Clin Oncol. Feb 24, 2026; 17(2): 114622
Published online Feb 24, 2026. doi: 10.5306/wjco.v17.i2.114622
Table 1 Baseline demographic and clinical characteristics of study participants, mean ± SD/median (interquartile range)
Parameter
HCC (n = 50)
Cirrhosis (n = 50)
HCV (n = 50)
Controls (n = 50)
Test
P value
Age (years)57.4 ± 8.255.9 ± 7.854.8 ± 8.054.2 ± 7.6ANOVA0.312
Sex (male/female)34/1632/1831/1930/20χ20.894
ALT (U/L)55.0 (48.0-63.0)52.0 (46.0-61.0)50.0 (44.0-58.0)24.0 (20.0-28.0)Kruskal-Wallis< 0.001a
AST (U/L)62.0 (54.0-70.0)58.0 (50.0-66.0)56.0 (49.0-62.0)25.0 (21.0-29.0)Kruskal-Wallis< 0.001a
Albumin (g/dL)3.2 ± 0.43.3 ± 0.43.5 ± 0.34.2 ± 0.3ANOVA< 0.001a
INR1.42 ± 0.121.40 ± 0.111.35 ± 0.101.01 ± 0.05ANOVA< 0.001a
Table 2 Immune checkpoint expression and T-cell profiles across study groups, mean ± SD/median (interquartile range)
Parameter
HCC (n = 50)
Cirrhosis (n = 50)
HCV (n = 50)
Control (n = 50)
8F/Kruskal-Wallis
P value
CD4/PD135.14 ± 0.4320.87 ± 0.912.59 ± 0.8412.0 ± 1.336700.7< 0.001c
Post hoc71P < 0.001c, 2P < 0.001c, 3P = 0.001b, 4P < 0.001c, 5P < 0.001c, 6P < 0.001c
CD8/PD135.57 ± 0.5823.89 ± 0.8115.6 ± 0.8415.0 ± 1.335854.15< 0.001c
Post hoc71P < 0.001c, 2P < 0.001c, 3P = 0.007b, 4P < 0.001c, 5P < 0.001c, 6P < 0.001c
CD4/TIGIT13.77 ± 0.378.76 ± 0.466.18 ± 0.795.6 ± 1.331011.91< 0.001c
Post hoc71P < 0.001c, 2P > 0.999, 3P = 0.003b, 4P < 0.001c, 5P < 0.001c, 6P < 0.001c
CD8/TIGIT17.66 ± 0.386.56 ± 0.467.6 ± 0.777.0 ± 1.322088.86< 0.001c
Post hoc71P < 0.001c, 2P < 0.001c, 3P =0.002b, 4P < 0.001c, 5P < 0.001c, 6P =0.043a
CD4/TIM314.7 ± 0.346.34 ± 0.463.6 ± 0.773.0 ± 1.332169.06< 0.001c
Post hoc71P < 0.001c, 2P > 0.999, 3P = 0.002b, 4P < 0.001c, 5P < 0.001c, 6P < 0.001c
CD8/TIM317.14 ± 0.448.56 ± 0.464.1 ± 0.773.5 ± 1.332868.84< 0.001c
Post hoc71P < 0.001c, 2P < 00.001c, 3P = 0.002b, 4P < 0.001c, 5P < 0.001c, 6P < 0.001c
CD366.12 ± 0.6867.84 ± 0.8371.25 ± 0.8871.15 ± 1.37339.21< 0.001c
Post hoc71P < 0.001c, 2P < 0.001c, 3P = 0.945, 4P < 0.001c, 5P < 0.001c, 6P < 0.001c
CD439.06 ± 0.7543.95 ± 0.9147.25 ± 0.8847.15 ± 1.37733.67< 0.001c
Post hoc71P < 0.001c, 2P < 0.001c, 3P = 0.95, 4P < 0.001c, 5P < 0.001c, 6P < 0.001c
CD821.06 ± 0.6923.98 ± 0.8724.25 ± 0.8824.15 ± 1.37122.07< 0.001c
Post hoc71P < 0.001c, 2P = 0.05, 3P = 0.947, 4P < 0.001c, 5P < 0.001c, 6P = 0.829
CD4/CD81.86 ± 0.031.83 ± 0.031.95 ± 0.031.96 ± 0.05140.79< 0.001c
Post hoc71P = 0.023a, 2P < 0.001c, 3P = 0.852, 4P < 0.001c, 5P < 0.001c, 6P < 0.001c
AFP (ng/mL)13210 (11008.3-14956.8)4 (2-6)6.85 (6.08-7.5)4.3 (3.08-5.53)144.208< 0.001c
Post hoc71P < 0.001c, 2P < 0.001c, 3P < 0.001c, 4P < 0.001c, 5P < 0.001c, 6P = 0.883
Table 3 Liver function scores in hepatocellular carcinoma and cirrhosis groups, mean ± SD/n (%)
Parameter
HCC (n = 50)
Cirrhosis (n = 50)
t/χ2
P value
MELD2.5 ± 1.02.56 ± 1.09-0.2870.774
Child-Pugh class1.7950.18
A9 (18)13 (26)
B9 (18)12 (24)
C32 (64)25 (50)
Table 4 Tumor characteristics and outcomes in hepatocellular carcinoma patients, n (%)
Parameter
n = 50
Fate
Non-survivor17 (34)
Survivor33 (66)
Lesion size
    < 3 cm18 (36)
    > 3 cm32 (64)
BCLC stage
    A9 (18)
    B1 (2)
    C8 (16)
    D32 (64)
Table 5 Correlations with disease severity in hepatocellular carcinoma patients
ParameterMELD
Child-Pugh
AFP
BCLC stage
r1
P value
r1
P value
r1
P value
r1
P value
CD4/PD-1 (%)-0.730< 0.001a0.855< 0.001a0.750< 0.001a0.855< 0.001a
CD8/PD-1 (%)-0.738< 0.001a0.855< 0.001a0.752< 0.001a0.855< 0.001a
CD4/TIGIT (%)-0.764< 0.001a0.857< 0.001a0.758< 0.001a0.856< 0.001a
CD8/TIGIT (%)-0.764< 0.001a0.856< 0.001a0.756< 0.001a0.856< 0.001a
CD4/TIM-3 (%)-0.764< 0.001a0.858< 0.001a0.759< 0.001a0.858< 0.001a
CD8/TIM-3 (%)-0.764< 0.001a0.835< 0.001a0.745< 0.001a0.835< 0.001a
CD3 (%)-0.765< 0.001a0.837< 0.001a0.748< 0.001a0.837< 0.001a
CD4 (%)-0.756< 0.001a0.830< 0.001a0.746< 0.001a0.830< 0.001a
CD8 (%)-0.751< 0.001a0.835< 0.001a0.749< 0.001a0.835< 0.001a
CD4/CD8 ratio0.745< 0.001a-0.844< 0.001a-0.749< 0.001a-0.840< 0.001a
Table 6 Correlations with disease severity in cirrhosis patients
ParameterMELD
Child-Pugh
AFP
r1
P value
r1
P value
r1
P value
CD4/PD-1 (%)0.926< 0.001a0.557< 0.001a0.1420.324
CD8/PD-1 (%)0.910< 0.001a0.566< 0.001a0.1500.296
CD4/TIGIT (%)0.919< 0.001a0.565< 0.001a0.1380.339
CD8/TIGIT (%)0.927< 0.001a0.565< 0.001a0.1420.324
CD4/TIM-3 (%)0.917< 0.001a0.565< 0.001a0.1400.331
CD8/TIM-3 (%)0.883< 0.001a0.565< 0.001a0.1440.315
CD3 (%)0.932< 0.001a0.563< 0.001a0.1460.307
CD4 (%)0.916< 0.001a0.562< 0.001a0.1410.329
CD8 (%)0.914< 0.001a0.562< 0.001a0.1420.324
CD4/CD8 ratio-0.901< 0.001a-0.562< 0.001a-0.1450.309
Table 7 Immune profiles and mortality in hepatocellular carcinoma patients, mean ± SD/median (interquartile range)
Parameter
Survivors (n = 33)
Non-survivors (n = 17)
t/Z1
P value
CD4/PD-1 (%)34.86 ± 0.2535.63 ± 0.21-10.52< 0.001a
CD8/PD-1 (%)36.20 ± 0.2737.28 ± 0.26-13.63< 0.001a
CD4/TIGIT (%)13.54 ± 0.2114.21 ± 0.17-11.32< 0.001a
CD8/TIGIT (%)17.43 ± 0.2118.11 ± 0.17-11.56< 0.001a
CD4/TIM-3 (%)14.50 ± 0.1715.11 ± 0.17-12.12< 0.001a
CD8/TIM-3 (%)16.86 ± 0.1717.67 ± 0.27-12.66< 0.001a
CD3 (%)65.72 ± 0.4166.90 ± 0.32-10.46< 0.001a
CD4 (%)38.60 ± 0.4339.94 ± 0.36-10.95< 0.001a
CD8 (%)20.64 ± 0.3621.88 ± 0.35-11.65< 0.001a
CD4/CD8 ratio1.87 ± 0.011.83 ± 0.0111.82< 0.001a
AFP (ng/mL)12004 (10341-13654.5)14030 (13672.5-14406)-3.85< 0.001a
Table 8 Prognostic modeling and receiver operating characteristic curve performance for hepatocellular carcinoma mortality
Variable
Random forest AUC (95%CI)
Logistic regression OR (95%CI)
P value
ROC AUC (95%CI)
Cut-off
Sensitivity (%)
Specificity (%)
CD4/PD-1 (%)0.92 (0.85-0.98)2.3 (1.4-3.8)< 0.001a0.92 (0.85-0.98)> 35%8885
CD8/TIM3 (%)0.90 (0.83-0.97)2.5 (1.5-4.2)< 0.001a0.90 (0.83-0.97)> 17%8582
AFP (ng/mL)3.2 (1.8-5.7)< 0.001a0.89 (0.81-0.96)> 130009080
AFP + CD8/TIM30.95 (0.90-0.99)< 0.001a0.95 (0.90-0.99)> 13000, > 17%9290
BCLC stage (D vs A-C)3.8 (2.0-7.2)< 0.001a
Clustering: High-expression50% mortality (n = 30)< 0.001aElevated (PD-1, TIGIT, TIM-3)
Clustering: Low-expression10% mortality (n = 20)< 0.001aLow (PD-1, TIGIT, TIM-3)
Table 9 Cox regression analysis for hepatocellular carcinoma mortality
Variable
HR (95%CI)
P value
CD8/TIM3 (high)2.8 (1.5-5.2)0.001a
BCLC stage D3.1 (1.7-5.6)< 0.001b
Age1.2 (0.9-1.6)0.15
MELD score1.1 (0.8-1.5)0.22
Table 10 Mediation analysis of CD4/programmed death 1 in alpha-fetoprotein-mortality relationship
Effect
1β (95%CI)
P value
Direct effect (AFP to mortality)0.30 (0.15-0.45)< 0.001a
Indirect effect (AFP to CD4/PD-1 to mortality)0.45 (0.25-0.65)< 0.001a
Total effect0.75 (0.55-0.95)< 0.001a
Table 11 Immune profiles by Barcelona Clinic Liver Cancer Stage in hepatocellular carcinoma patients, mean ± SD/median (interquartile range)/n (%)
Parameter
Early stage (BCLC A/B, n = 10)
Advanced stage (BCLC C/D, n = 40)
t/Z/χ2
P value
CD4/PD-1 (%)33.50 ± 0.5036.00 ± 0.40t = -15.00< 0.001b
CD8/TIM-3 (%)15.50 ± 0.5017.80 ± 0.40t =-14.50< 0.001b
AFP (ng/mL)10000 (8000-12000)14000 (12000-15000)Z =-3.50< 0.001b
Mortality1 (10)16 (40)χ2 = 4.20 0.040a
Table 12 Univariate and multivariate Cox proportional hazards regression for overall survival in hepatocellular carcinoma patients
Variable
Univariate HR (95%CI)
P value
Multivariate HR (95%CI)
P value
CD4/PD-1 (%)2.5 (1.6-3.9)< 0.001c2.1 (1.3-3.4)0.002b
CD8/PD-1 (%)2.3 (1.5-3.5)< 0.001c
CD4/TIGIT (%)1.8 (1.1-2.9)0.015a
CD8/TIGIT (%)1.9 (1.2-3.0)0.008b
CD4/TIM3 (%)1.7 (1.0-2.8)0.045a
CD8/TIM3 (%)2.7 (1.7-4.3)< 0.001c2.4 (1.4-4.1)0.001b
CD3 (%)1.5 (0.9-2.5)0.12
CD4 (%)1.4 (0.8-2.4)0.23
CD8 (%)1.6 (0.9-2.8)0.09
CD4/CD8 ratio0.4 (0.2-0.7)0.002b
AFP (ng/mL)3.0 (1.9-4.7)< 0.001c2.6 (1.5-4.5)< 0.001c
MELD score1.1 (0.8-1.5)0.53
Child-Pugh class (C vs A/B)1.8 (1.0-3.2)0.048a
BCLC stage (D vs A-C)3.5 (2.2-5.6)< 0.001c3.0 (1.7-5.3)< 0.001c