A total of 152 patients met the inclusion and exclusion criteria. Patients were divided into 4 groups according to HBsAg and HBeAg values at baseline. There were 81 in group A (HBsAg < 1500, HBeAg (-)), 6 in group B (HBsAg > 1500, HBeAg (-)), 42 in group C (HBsAg < 1500, HBeAg (+)) and 23 in group D (HBsAg > 1500, HBeAg (+)). (Fig. 1)
3.1. Baseline Patient Characteristics
Of the 152 patients enrolled, the mean age was 38.4 years, and 100 (65.8%) patients were male. 52(34.2%), 57 (37.5%), 25 (16.4%), and 18 (11.8%) patients were followed up to 12 weeks, 24 weeks, 36 weeks, and 48 weeks respectively. There were no significant differences in gender, age,body mass index, HBV DNA, liver function, blood routine and liver hardness among the four groups (P > 0.05). (Table 1).
Table 1
Baseline characteristics of patients
|
A group(n = 81)
|
B group(n = 6)
|
C group(n = 42)
|
D group(n = 23)
|
P value
|
Basis for grouping
|
HBsAg<1500, HBeAg(-)
|
HBsAg > 1500, HBeAg(-)
|
HBsAg<1500, HBeAg(+)
|
HBsAg > 1500, HBeAg(+)
|
|
age
|
39.1 ± 9.8
|
36.5 ± 5.1
|
37.8 ± 9.1
|
37.4 ± 8.0
|
0.340
|
Sex; Male; N (%)
|
53(65.4)
|
4(66.7)
|
28(66.7)
|
15(65.2)
|
0.542
|
BMI
|
24.5 ± 6.0
|
25.0 ± 6.2
|
25.0 ± 2.8
|
24.0 ± 2.9
|
0.462
|
treatment
|
PRG-IFN
|
5
|
1
|
0
|
0
|
|
PEG-IFN + NAs
|
76
|
5
|
42
|
23
|
|
HBV-DNA < 50;n(%)
|
55(67.9)
|
4(66.7)
|
31(73.8)
|
15(65.2)
|
0.538
|
Current course of treatment; N (%)
|
12W
|
26(32.1)
|
2(33.3)
|
12(28.6)
|
12(52.2)
|
|
24W
|
32(39.5)
|
2(33.3)
|
15(35.7)
|
8(34.8)
|
|
36W
|
10(12.3)
|
2(33.3)
|
11(26.2)
|
2(8.7)
|
|
48W
|
13(16.0)
|
0(0.0)
|
4(9.5)
|
1(4.3)
|
|
WBC(×109/L)
|
5.2 ± 1.3
|
5.5 ± 0.7
|
5.3 ± 1.3
|
5.4 ± 1.7
|
0.433
|
HGB(g/L)
|
151.4 ± 16.5
|
145.0 ± 14.6
|
150.3 ± 17.8
|
151.7 ± 12.7
|
0.653
|
PLT(×109/L)
|
205.3 ± 67.5
|
218.7 ± 33.4
|
194.4 ± 48.1
|
198.2 ± 63.4
|
0.456
|
NEUT%
|
55.2 ± 9.3
|
61.4 ± 10.0
|
57.1 ± 10.8
|
59.4 ± 8.4
|
0.152
|
ALT(U/L)
|
25.0(16.5, 36.5)
|
28.0(16.8, 45.5)
|
26.0(18.8, 30.5)
|
25.2(19.3, 29.1)
|
0.641
|
AST(U/L)
|
24.0(18.1, 27.1)
|
20.5(19.0, 27.5)
|
24.0(19.2, 27.3)
|
21.5(17.0, 24.0)
|
0.365
|
TBIL(mmol/L)
|
14.9(13.0, 17.2)
|
14.7(11.8, 21.4)
|
14.5(12.6, 16.3)
|
15.1(12.7, 18.8)
|
0.737
|
DBIL(mmol/L)
|
5.2(4.0, 6.1)
|
4.6(3.2, 7.6)
|
5.0(3.9, 6.0)
|
5.0(3.6, 6.9)
|
0.563
|
CAP(dB/m)
|
231.4 ± 14.3
|
239.1 ± 26.3
|
234.1 ± 49.4
|
226.2 ± 41.0
|
0.213
|
LSM(kPa)
|
7.4 ± 2.9
|
7.2 ± 1.7
|
8.2 ± 4.1
|
7.3 ± 2.7
|
0.597
|
3.2 comparison of HBsAg (IU/ml) decreased by more than 1log10IU/ml in different courses of treatment or groups
There were 39 (25.7%) patients whose HBsAg declining more than 1log10IU/ml (compared with baseline data) in 152 patients. Among 152 patients, 12-week treatment group, 24-week treatment group, 36-week treatment group and 48-week treatment group had 21.2% (11/52), 19.3% (11/57), 24.0% (6/25) and 61.1% (11/18) decreased by more than 1log10IU/ml respectively. There was significant difference between groups (correctedχ2 = 11.919, P = 0.008). In group A, B, C and D, 38.3% (31/81), 16.7% (1/6), 16.7% (7/42) and 0.0% (0/23) decreased by more than 1log10IU/ml respectively. There was significant difference between groups (correctedχ2 = 22.067, P = 0.000). (Table 2, Fig. 2)
Table 2
Comparison of HBsAg (IU/ml) decreased by more than 1log10IU/ml in different courses/groups
|
A group
|
B group
|
C group
|
D group
|
Overall
|
12W;n(%)
|
10(38.5)
|
0(0.0)
|
1(8.3)
|
0(0.0)
|
11(21.2)
|
24W;n(%)
|
10(31.3)
|
0(0.0)
|
1(6.67)
|
0(0.0)
|
11(19.3)
|
36W;n(%)
|
2(20.0)
|
1(50.0)
|
3(27.3)
|
0(0.0)
|
6(24.0)
|
48W;n(%)
|
9(69.2)
|
0(0.0)
|
2(50.0)
|
0(0.0)
|
11(61.1)
|
Overall
|
31(38.3)
|
1(16.7)
|
7(16.7)
|
0(0.0)
|
39(25.7)
|
3.3 Factors affecting the decline of HBsAg
A total of 63.2% (96/152) patients showed ALT > 54U/L at week 12/24, the highest was 398U/L. In order to evaluate the factors affecting HBsAg decline, the dependent variable was whether HBsAg decreased by more than 1log10IU/ mL (yes = 0, no = 1). A binary multivariate logistic regression analysis was conducted using the significant items in the single factor as independent variables. The results showed that treatment duration, ALT elevation at 12/24 weeks, HBeAg status and baseline HBsAg level were the factors influencing HBsAg decline. HBsAg > 1500IU/ ml at baseline had the highest Odds Ratio (OR) (72.188, 95%CI 4.375-1185.777), and ALT elevation at 12/24 week had the lowest OR (0.091, 95%CI 0.021–0.391). Gender, age, baseline HBV DNA, and liver fibrosis were not statistically significant. (Table 3).
Table 3
Logistic regression analysis of HBsAg decrease by more than 1log10IU/ml
|
single factor analysis
|
multivariate analysis
|
OR
|
95%CI
|
P
|
OR
|
95%CI
|
P
|
gender
|
1.127
|
0.513–2.472
|
0.766
|
1.129
|
0.407–3.131
|
0.816
|
age
|
0.832
|
0.390–1.773
|
0.633
|
0.632
|
0.234–1.707
|
0.366
|
HBV DNA
|
0.714
|
0.314–1.622
|
0.421
|
1.133
|
0.388–3.307
|
0.366
|
E(kPa)
|
0.862
|
0.404–1.840
|
0.702
|
0.413
|
0.147–1.159
|
0.093
|
course
|
0.411
|
0.189–0.895
|
0.025
|
0.250
|
0.084–0.746
|
0.013
|
12/24WALT>54U/L
|
0.094
|
0.027–0.322
|
0.000
|
0.091
|
0.021–0.391
|
0.001
|
HBeAg(-)
|
15.639
|
3.593–68.065
|
0.000
|
11.953
|
0.284–62.549
|
0.003
|
HBsAg Baseline (IU/ml)
|
10−1500
|
9.667
|
1.256–74.379
|
0.029
|
5.267
|
0.514–53.945
|
0.162
|
>1500
|
72.500
|
7.224−727.606
|
0.000
|
72.188
|
4.375−1185.777
|
0.003
|
In patients with HBsAg decline of 1log10IU/ml or more, receiver operating characteristic curve (ROC curve) was used to evaluate the influence of the above statistically significant factors on HBsAg decline. AUROC of treatment course, ALT elevation, HBeAg status and baseline HBsAg level were 0.616, 0.808, 0.701 and 0.793, respectively. (Table 4, Fig. 3).
Table 4
working characteristic curves of patients with hbsag decline of 1log10IU/ml or more
|
AUC
|
SD
|
95%CI
|
P
|
Sensitivity (%)
|
Specificity (%)
|
Cut point
|
course
|
0.616
|
0.056
|
0.505–0.726
|
0.031
|
28.2
|
93.8
|
>36
|
12/24W ALT
|
0.808
|
0.039
|
0.732–0.884
|
0.000
|
82.1
|
73.0
|
>65
|
baseline HBeAg
|
0.701
|
0.044
|
0.615–0.787
|
0.000
|
97.4
|
45.5
|
<0.1
|
baseline HBsAg
|
0.793
|
0.041
|
0.713–0.873
|
0.000
|
73.7
|
78.6
|
<223.16
|
3.4 The time required for HBsAg to decrease by more than 1log10IU/ mL
Among the 39 patients with HBsAg decrease of more than 1log10IU/ml, the time required for HBsAg decrease of 1log10IU/ml was 64.1% (25/39), 17.9% (7/39) and 17.9% (7/39) at 8–12, 12–24 and 24–36 weeks, respectively. (Table 5).
Table 5
Time required for HBsAg to decrease by more than 1log10IU/ ml
|
A group(n = 31)
|
B group(n = 1)
|
C group(n = 7)
|
D group(n = 0)
|
n
|
8−12W;n(%)
|
20(64.5)
|
1(100.0)
|
4(57.2)
|
0(0.0)
|
25(64.1)
|
12−24W;n(%)
|
6(19.4)
|
0(0.0)
|
1(14.3)
|
0(0.0)
|
7(17.9)
|
24−36W;n(%)
|
5(16.1)
|
0(0.0)
|
2(28.6)
|
0(0.0)
|
7(17.9)
|
3.5 comparison of HBsAg (IU/ml) decreased by 1log10IU/ml at different baselines after treatment
According to the baseline HBsAg level, 152 patients were divided into HBsAg value of 0.05-10 IU/ml group, 10-1500 IU /ml group, > 1500IU/ml group. After interferon treatment, HBsAg decreased by more than 1log10IU/ml in 0.05-10 IU/ml group, 10-1500 IU /ml group and > 1500IU/ml group by 71.4% (10/14), 25.9% (28/109) and 3.4% (1/29), respectively. There was significant difference between groups (χ2 = 23.451, P = 0.000). (Table 6).
Table 6
Comparison of HBsAg (IU/ml) baseline values in different ranges that decreased by more than 1log10IU/ml after treatment
|
0.05−10(n = 14)
|
10−1500(n = 109)
|
>1500(n = 29)
|
P
|
n (%)
|
10(71.4)
|
28(25.9)*
|
1(3.4)*+
|
0.000
|
*P < 0.05 vs. HBsAg<10;+P1 < 0.05vs.HBsAg 10-1500 |
3.6 the relationship between baseline HBsAg level and its decline time of more than 1log10IU/ml
Pearson correlation analysis was used to analyze the relationship between baseline HBsAg level and the time required to decrease by 1log10IU/ml in 39 patients, and the two showed extremely significant positive correlation (P = 0.000, r = 0.731). The scatter plot of baseline HBsAg level and the time required to decrease by 1log10IU/ml was drawn, and the scatter plot was visually observed, indicating that there may be a linear correlation between the two. Establishing the linear regression equation between the baseline HBsAg level (IU/ mL) and the time required for the 1log10IU/ mL decrease of HBsAg (W). Taking the baseline HBsAg level (IU/ mL) as the independent variable and the time required for the 1log10IU/ mL decrease of HBsAg (W) as the dependent variable, a linear regression model was established through linear regression: Y = 12.13 + 0.02 X. (Fig. 4)
3.7 Clinical Cure
Among the 152 patients followed up, 10 cases of CHB were clinically cured, all of which were the dominant population of interferon therapy. S-antigen serological conversion occurred in 4 patients (2 of whom received hepatitis B vaccine after interferon treatment); One patient's HBV DNA was 593IU/ ml at the time of enrollment and turned negative at 12 weeks.(Table 7)
Table 7
10 Clinical datas of HBsAg negative conversion patients
|
baseline
|
therapeutic schedule
|
course
|
Whether HBsAg serologic conversion
HBsAb(mIU/mL)/time&
|
HBV DNA
(IU/ml)
|
HBsAg(IU/ml)
|
HBsAb(mIU/mL)
|
HBeAg
(C.O.I)
|
treatment(W)
|
Consoli-date(W)
|
N1
|
<50
|
187.99
|
0.1
|
0
|
NAs+PEG-IFN
|
13
|
0
|
No
|
N2
|
<50
|
219.9
|
0.9
|
0.9
|
NAs+PEG-IFN
|
28
|
8
|
No
|
N3*
|
<50
|
0.08
|
1.1
|
0
|
PEG-IFN
|
12
|
0
|
680.6/28
|
N4*
|
<50
|
223.16
|
0.1
|
0
|
NAs+PEG-IFN
|
15
|
8
|
85/35
|
N5
|
<50
|
110.66
|
0
|
0
|
NAs+PEG-IFN
|
48
|
12
|
25.9/80
|
N6
|
<50
|
5.34
|
1
|
0
|
NAs+PEG-IFN
|
13
|
8
|
No
|
N7#
|
593
|
41.16
|
0.4
|
0
|
NAs+PEG-IFN
|
24
|
10
|
35.4/24
|
N8
|
<50
|
9.9
|
0.3
|
0
|
PEG-IFN
|
13
|
0
|
No
|
N9#
|
<50
|
8.9
|
0.4
|
0
|
PEG-IFN
|
12
|
10
|
No
|
N10
|
<50
|
10
|
0.2
|
0
|
NAs+PEG-IFN
|
14
|
0
|
No
|
* indicates that the patient received hepatitis B vaccine after interferon treatment
# indicates that the patient is in the interferon consolidation phase
& represents the time from the beginning of interferon therapy to the appearance of surface antibodies
3.8 Treatment Of 152 Patients
152 patients were divided into nine groups (a、b、c、d、e、f、g、h and i) according to the treatment course of interferon, whether they had stopped taking the drug, and whether they had achieved clinical cure (Fig. 5). In this study, all patients with a course of more than 48 weeks had stopped interferon therapy.
Patients in group c and group g were classified as poor compliance (86,56.6%). The reasons for drug withdrawal in group c and g included "the patient thought the treatment effect was not ideal", fatigue, muscle pain, hair loss, loss of appetite, insomnia, epidemic reasons, conflict with the time of COVID-19 vaccine, and 1 patient developed herpes during treatment.Most patients stopped taking drugs for multiple reasons.Among the 152 patients followed up, most of them were accompanied by leukocyte and neutrophil decline, and all of them were treated with leukocyte raising drugs.
Since the 10 patients who achieved clinical cure were all in Group A, that is, HBsAg < 1500IU/ ml and HBeAg (-), we studied the relationship between compliance and clinical cure in group A (n = 81), and a total of 63 patients were included after excluding patients who had not stopped taking drugs temporarily (n = 18). Sixty-three patients were divided into two groups according to their compliance. The difference between groups was statistically significant (χ2 = 17.545, P = 0.000). Among patients with baseline HBsAg levels of 0.05-10IU/ml(n = 14), the clinical cure rate was 35.7%, and the percentage of poor compliance was 50%. Due to limited data, statistical analysis was not carried out. (Table 8).
Table 8
Relationship between compliance and clinical cure in group A
|
Poor compliance (%)
|
Clinical cure (%)
|
χ2
|
P
|
r
|
A group
|
63.5%
|
15.9
|
17.545
|
0.000
|
0.461
|