Patient characteristics
600 HCC cases and 650 cancer-free controls were included in this study initially. Due to unsuccessful genotyping of 14 HCC cases and 3 controls, a total of 1233 participants consisting of 586 HCC patients (491 males and 95 females) and 647 controls (297 males and 350 females) were included in this study at last. The baseline characteristics of HCC cases and controls were summarized in Table 1. There were no significant differences in age (stratified by 65 years) and family history of liver cancer, while gender, hepatitis B, smoking and drinking status were significant statistically different between two groups (P < 0.05). A total of 299 HCC patients underwent hepatectomy treatment and were followed to December 2019, the maximum follow-up time was 30.7 months with a medium follow-up time 18.0 months, and the characteristics and clinical features of HCC patients treated with hepatectomy were summarized in supplementary table S3.
Table 1
Baseline characteristics of HCC patients and controls
Variable
|
HCC
|
controls
|
P-value
|
(n = 586)
|
(n = 647)
|
|
Age [n (%)]
|
|
|
0.092
|
༜65 years
|
444(75.80)
|
516(79.80)
|
|
≥ 65 years
|
142(24.20)
|
131(20.20)
|
|
Gender [n (%)]
Male
Female
|
491(83.80)
95(16.20)
|
297(45.90)
350(54.10)
|
< 0.001*
|
Smoking status [n (%)]
|
|
|
< 0.001*
|
Never
|
275(46.90)
|
506(78.20)
|
|
Ever/current
|
311(53.10)
|
141(21.80)
|
|
Drinking status [n (%)]
|
|
|
< 0.001*
|
Never
|
326(55.60)
|
491(75.90)
|
|
Ever/current
Hepatitis B [n (%)]
With
without
|
260(44.40)
442(75.40)
144(24.60)
|
156(24.10)
143(22.10)
504(77.90)
|
< 0.001*
|
Family history of liver cancer [n (%)]
|
|
|
0.347
|
without
|
560(95.60)
|
625(96.60)
|
|
with
|
26(4.40)
|
22(3.40)
|
|
*P < 0.05, statically significant. |
Genotype frequency and effects of CD35 on HCC risk
The genotype distribution of CD35 in controls followed the predictions of Hardy-Weinberg equilibrium (supplementary table S1, P > 0.05). Among six SNPs of CD35, only the frequencies of GG, CG, and CC genotypes of rs7525160 in HCC and control groups were statistically significant (P < 0.05, Table 2). After adjusting for gender, smoking, Hepatitis B and drinking status, the risk of HCC was 1.46-times higher in individuals with at least one mutant allele C (CG + CC) than in individuals with the GG genotype [adjusted OR = 1.46, 95% CI (1.09–1.95), P = 0.012] (Table 2). No significant differences were seen in the distribution of rs10494885, rs2296160, rs3737002, rs3849266 and rs6691117 genotypes between the HCC and control group (P > 0.05, Table 2). The three genotypes of rs7525160 polymorphism detected by MALDI-TOF and sequencing map for CD35 rs7525160 GG and CG genotypes were summarized in supplementary figure S1.
Table 2
SNPs of CD35 gene in patients with HCC and controls
SNP ID
|
Genotype
|
HCC n (%)
|
Control n (%)
|
Adjusted OR (95% CI) a
|
P value
|
rs10494885
|
AA
AG
GG
GG/AG
|
74(12.60)
251(42.80)
261(44.50)
512(87.40)
|
90(13.90)
275(42.50)
282(43.60)
557(86.10)
|
1.00(Reference)
1.159(0.76, 1.77)
1.131 (0.74, 1.74)
1.156(0.77, 1.73)
|
0.493
0.575
0.479
|
rs2296160
|
AA
AG
GG
GG/AG
|
77(13.10)
240(41.00)
269(50.40)
509(86.90)
|
79(12.20)
302(46.70)
265(41.00)
567(87.80)
|
1.00(Reference)
0.84(0.54, 1.31)
0.98(0.63, 1.52)
0.91(0.60, 1.38)
|
0.443
0.922
0.652
|
rs3737002
|
CC
CT
TT
TT/CT
|
261(44.50)
245(41.80)
80(13.70)
325(55.50)
|
279(43.10)
281(43.40)
87(13.40)
368(56.90)
|
1.00(Reference)
1.09(0.81, 1.47)
1.02(0.66, 1.58)
1.07(0.81, 1.41)
|
0.571
0.917
0.639
|
rs3849266
|
CC
CT
TT
TT/CT
|
262(44.70)
244(41.60)
80(13.70)
324(55.30)
|
283(43.90)
275(42.60)
87(13.50)
362(56.10)
|
1.00(Reference)
1.11(0.82, 1.49)
1.03 (0.67, 1.58)
1.08(0.82, 1.43)
|
0.506
0.902
0.570
|
rs6691117
|
AA
AG
GG
GG/AG
|
288(49.10)
232(39.60)
66(11.30)
298(50.90)
|
342(52.90)
255(39.40)
50(7.70)
305(47.10)
|
1.00(Reference)
0.93(0.69, 1.24)
1.27(0.77, 2.08)
0.99(0.75, 1.30)
|
0.619
0.343
0.933
|
rs7525160
|
GG
CG
CC
CC/CG
|
180(30.70)
286(48.80)
120(20.50)
406(69.30)
|
258(39.90)
296(45.70)
93(14.40)
389(60.10)
|
1.00(Reference)
1.37(1.01, 1.86)
1.69(1.13, 2.54)
1.46(1.09,1.95)
|
0.047*
0.011*
0.012*
|
a adjusted for gender, smoking, Hepatitis B and drinking status. *P < 0.05, statically significant.
Stratified analysis of demographic characteristics and environmental factors
We stratified to analyze the polymorphism of CD35 rs7525160 and HCC risk, based on factors including demographic characteristics (gender, age) and environmental factors (Hepatitis B, family history of liver cancer, drinking and smoking status) (supplementary table S4). After stratified analysis, we found that CD35 rs7525160 CC/CG genotype increased the risk of HCC in younger than 65 years patients [P for interaction = 0.042, adjusted OR = 1.85, 95%CI (1.30–2.62), P = 0.001].
Genotype frequency and effects on tumor clinicopathological types
Further stratified analysis of the clinical characteristics of HCC patients revealed that CD35 rs7525160 CC/CG genotype could increase the risk of different clinicopathological types of HCC, especially among patients with AFP ≥ 400ng/ml [adjusted OR = 1.94, 95%CI (1.24–3.04), P = 0.004], tumor size of > 5cm [adjusted OR = 1.75, 95% CI (1.19–2.56), P = 0.004], TNM stage III/IV [adjusted OR = 1.64, 95%CI (1.07–2.52), P = 0.024], with a background cirrhosis [adjusted OR = 1.50, 95% CI (1.05–2.15), P = 0.026], with portal vein tumor thrombosis [adjusted OR = 2.30, 95%CI (1.24–4.27), P = 0.008] (supplementary table S5).
CD35 genetic variation on postoperative recurrence of HCC and overall survival analysis
As aforementioned, CD35 rs7525160 pertained to increased HCC risk especially in advanced stage or big tumors, which suggested its potential predictive value in prognosis. To evaluate this, we detected the six CD35 SNPs genotype effects on the recurrence rate and mean recurrence-free survival (MRFS) in 299 HCC patients who underwent curative hepatectomy. Patients who were treated with Transcatheter Arterial Chemoembolization (TACE), radiofrequency ablation and drug only were excluded in order to exclude the influence of different treatments on the prognosis of HCC. In the Kaplan-Meier analyses, our results showed that the MRFS was 17.82 months with 95%CI 16.104–19.544 months in CD35 rs7525160 CC/CG individuals, which was shorter than in individuals with the GG genotype (MRFS 22.05 months, 95%CI 19.819–24.289 months, P = 0.0073). (Fig. 1)
Considering that the recurrence of HCC is related to many clinical factors, we firstly conducted univariate survival analysis to find the possible factors affect the recurrence of HCC in hepatectomy patients. Our result showed that HBV DNA level ≥ 102 IU/mL [HR = 1.77, 95%CI (1.26,2.47), P = 0.011], Child-Pugh Class B [HR = 2.55, 95%CI (1.24,5.22), P = 0.011], Child-Pugh Class C [HR = 6.56, 95%CI (1.60,26.84), P = 0.009], with microvascular invasion [HR = 1.68, 95%CI (1.18,2.38), P = 0.004], tumor BCLC stage B/C [HR = 2.17, 95%CI (1.52,3.08), P < 0.001], AFP level ≥ 400 ng/mL [HR = 1.68, 95%CI (1.19,2.38), P = 0.003], tumor size > 5cm [HR = 1.55, 95%CI (1.11,2.17), P = 0.014], multiple tumor number [HR = 1.62, 95%CI (1.10,2.39), P = 0.014], TNM tumor stage III/IV [HR = 2.47, 95%CI (1.73,3.52), P < 0.001] and with portal vein tumor thrombosis [HR = 2.44, 95%CI (1.58,3.76), P < 0.001] were the possible factors associated with the recurrence of HCC (Table 3). Then, cox proportional hazard ratio model was used to analyze the effects of CD35 rs7525160 genotype on HCC recurrence in hepatectomy patients. As shown in Table 4 and Fig. 2, CD35 rs7525160 remained a significant independent risk factor for postoperative recurrence of HCC [adjusted HR = 1.64, 95%CI (1.10–2.45), P = 0.015]. In overall survival analysis, at a median follow-up of 18.0 months and maximum follow-up time of 30.7 months, 27 of 299 patients (9.0%) had died. The univariable analyses of overall survival are summarized in supplementary table S6. No statistical significance was found between the six Tag SNPs of CD35 and the overall survival.
Table 3 Univariate survival analysis of clinical factors associated with HCC recurrence
Variable
|
n
|
Recur [n (%)]
|
HR (95%CI)
|
P-value
|
Age
<65 years
≥65 years
Gender
Female
Male
Hepatitis B
Without
With
Smoking status
Never
Ever/current
Drinking status
Never
Ever/current
HBV DNA level (IU/mL)
<102
≥102
Child-Pugh Class
A
B
C
Microvascular invasion
Without
With
Tumor stage (BCLC)
0/A
B/C
α-fetoprotein level(ng/mL)
<400
≥400
Tumor size (cm)
≤5cm
>5cm
Tumor number
Single
Multiple
Tumor stage (TNM)
I/II
III/IV
Background cirrhosis
Absent
Present
Portal vein tumor thrombosis
No
Yes
Distant metastasis
No
Yes
|
235
64
48
251
61
238
138
161
166
133
168
130
286
11
2
202
92
228
71
208
91
150
149
244
55
233
66
88
211
265
34
297
2
|
112(81.20%)
26(18.8%)
24(17.4%)
114(82.6%)
26(18.8%)
112(81.2%)
62(44.9%)
76(55.1%)
77(55.8%)
61(44.2%)
67(48.6%)
71(51.4%)
128(92.8%)
8(5.8%)
2(1.4%)
87(63.5%)
50(36.5%)
91(65.9%)
47(34.1%)
85(61.6%)
53(38.4%)
60(43.5%)
78(56.5%)
104(75.4%)
34(24.6%)
92(66.7%)
46(33.3%)
33(23.9%)
105(76.1%)
113(81.9%)
25(18.1%)
137(99.3%)
1(0.7%)
|
1.00
0.74(0.49,1.14)
1.00
0.85(0.55,1.32)
1.00
1.13(0.74,1.73)
1.00
1.08(0.77,1.51)
1.00
1.03(0.74,1.44)
1.00
1.77(1.26,2.47)
1.00
2.55(1.24,5.22)
6.56(1.60,26.84)
1.00
1.68(1.18-2.38)
1.00
2.17(1.52,3.08)
1.00
1.68(1.19,2.38)
1.00
1.55(1,11-2.17)
1.00
1.62(1.10,2.39)
1.00
2.47(1.73,3.52)
1.00
1.32(0.89,1.95)
1.00
2.44(1.58,3.76)
1.00
1.16(0.16,8.33)
|
0.174
0.473
0.571
0.650
0.857
0.011*
0.011*
0.009*
0.004*
<0.001*
0.003*
0.011*
0.014*
<0.001*
0.166
<0.001*
0.880
|
*P<0.05, statically significant.
Table 4. Cox analysis of potential factors for recurrence analysis in patients with resected HCC
Variables
|
Crude HR (95%CI)
|
Adjusted HR (95%CI) a
|
P-value (Wald’s test)
|
rs7525160
GG
CG/CC
HBV DNA level (IU/mL)
<102
>102
Child-Pugh Class (C vs. B/A)
A
B
C
Microvascular invasion
without
with
Tumor stage (BCLC)
0/A
B/C
α-fetoprotein level (ng/mL)
<400
≥400
Tumor size (cm)
≤5
>5
Tumor number
single
multiple
Tumor stage (TNM)
I/II
III/IV
Portal vein tumor thrombosis
without
with
|
1.00
1.67 (1.14,2.44)
1.00
1.77 (1.26,2.47)
1.00
2.55 (1.24,5.22)
6.56 (1.6,26.84)
1.00
1.68 (1.18,2.38)
1.00
2.17 (1.52,3.08)
1.00
1.68 (1.19,2.38)
1.00
1.55 (1.11,2.17)
1.00
1.62 (1.1,2.39)
1.00
2.47 (1.73,3.52)
1.00
2.44 (1.58,3.76)
|
1.00
1.64 (1.10,2.45)
1.00
1.45 (1.01,2.08)
1.00
2.49 (1.16,5.36)
7.62 (1.74,33.44)
1.00
1.18 (0.80,1.75)
1.00
0.9 (0.36,2.27)
1.00
1.34 (0.92,1.95)
1.00
1.05 (0.71,1.54)
1.00
1.48 (0.82,2.67)
1.00
1.52 (0.69,3.37)
1.00
1.38 (0.66,2.88)
|
0.015*
0.044*
0.020*
0.007*
0.403
0.825
0.125
0.806
0.195
0.301
0.388
|
*
P<0.05, statically significant.