Sage Journals: Discover world-class research

Abstract

Background:

Low-level hepatitis B viremia is recognized as a potential driver of hepatocarcinogenesis; however, its prognostic impact on patients with advanced hepatocellular carcinoma (HCC) receiving systemic therapy remains poorly defined.

Objective:

To determine the independent prognostic impact of low-level viremia (LLV; HBV DNA 20–2000 IU/mL) compared with maintained virologic response (MVR; HBV DNA < 20 IU/mL) in patients with intermediate-to-advanced HBV-related HCC undergoing systemic therapy, and to evaluate whether on-treatment viral status serves as a superior predictor to baseline HBV DNA levels.

Design:

This was a multicenter retrospective cohort study.

Methods:

A total of 1814 patients treated at three centers between 2020 and 2024 were retrospectively enrolled. After ⩾4 months of follow-up, patients were classified into LLV (n = 860) or maintained virologic response (MVR; HBV DNA < 20 IU/mL, n = 954) groups. Overall survival (OS) and progression-free survival (PFS) were the primary endpoints. Logistic regression identified risk factors for LLV; Cox proportional hazards models assessed independent prognostic factors. Subgroup analyses were performed by treatment regimen and baseline HBV DNA load.

Results:

Multifactorial analysis identified HBeAg positivity (OR 1.971, 95% CI 1.53–2.541, p < 0.001), AST > 40 U/L (OR 1.437, 95% CI 1.126–1.832, p = 0.004), extrahepatic metastasis (OR 1.640, 95% CI 1.187–2.266, p = 0.003), and detectable baseline HBV DNA (OR 2.482, 95% CI 2.002–3.077, p < 0.001) as independent risk factors for LLV. Compared with the MVR group, patients in the LLV group had significantly reduced PFS (8.3 vs 14.3 months, p < 0.001) and OS (25.5 vs 37.8 months, p < 0.001). Multivariate analysis identified LLV as an independent predictor of worse OS (HR 1.506, 95% CI 1.299–1.746). Baseline HBV DNA load was not associated with survival.

Conclusion:

LLV is a robust, independent indicator of poor prognosis in systemic-treated intermediate-to-advanced HBV-HCC and should be preferred to baseline HBV DNA load for risk stratification and to guide intensified antitumor therapy.

Plain language summary

Low-level viremia shortens survival in systemic-treated advanced HBV-HCC

Introduction: Low-level hepatitis B viremia viremia is recognized as a potential driver of hepatocarcinogenesis; however, its prognostic impact on patients with advanced hepatocellular carcinoma (HCC) receiving systemic therapy remains poorly defined. Aim: We wanted to see if LLV (20–2,000 virus copies per mL) detected during treatment predicts shorter survival in patients with intermediate or advanced hepatitis B-related liver cancer who are receiving medicines such as lenvatinib or immunotherapy. Methods: A total of 1,814 patients treated at three centers between 2020 and 2024 were retrospectively enrolled. After ⩾4 months of follow-up, patients were classified into LLV (n = 860) or maintained virologic response (MVR; HBV DNA < 20 IU/mL, n = 954) groups. Overall survival (OS) and progression-free survival (PFS) were the primary endpoints. Logistic regression identified risk factors for LLV; Cox proportional hazards models assessed independent prognostic factors. Subgroup analyses were performed by treatment regimen and baseline HBV DNA load. Results: Multifactorial analysis identified HBeAg positivity (OR 1.971, 95% CI 1.53–2.541, p < 0.001), AST > 40 U/L (OR 1.437, 95% CI 1.126–1.832, p = 0.004), extrahepatic metastasis (OR 1.640, 95% CI 1.187–2.266, p = 0.003), and detectable baseline HBV DNA (OR 2.482, 95% CI 2.002–3.077, p < 0.001) as independent risk factors for LLV. Compared to the MVR group, patients in the LLV group had significantly reduced PFS (8.3 vs. 14.3 months, p < 0.001) and OS (25.5 vs. 37.8 months, p < 0.001). Multivariate analysis identified LLV as an independent predictor of worse OS (HR 1.506, 95% CI 1.299–1.746). Baseline HBV DNA load was not associated with survival.

Keywords

hepatitis B virus hepatocellular carcinoma low-level viremia maintained virologic response systemic antitumor therapy

Introduction

Globally, primary hepatic malignancies are the sixth most prevalent cancer type, with HBV infection implicated in over half of all cases of HCC.^1,2 The contribution of HBV has increased substantially in China and is now associated with 69.9% of all HCC cases.³ Owing to the insidious onset and highly aggressive nature of HCC, most patients are diagnosed when the disease is at an intermediate or advanced stage. Current first-line therapeutic strategies for intermediate or advanced HCC predominantly involve tyrosine kinase inhibitor (TKI) monotherapy, for example, lenvatinib or sorafenib, or immune checkpoint inhibitor-based combination regimens such as atezolizumab plus bevacizumab.⁴ The median overall survival (OS) of patients with intermediate or advanced HCC has improved from 10.7 months in 2018 to 19.2 months in 2022.^5,6 The optimization of the management of etiological factors, including antiviral therapy for HBV/HCV and metabolic dysfunction-associated steatohepatitis control, over the same period has contributed to this improvement in patient prognosis.⁴ Current clinical guidelines recommend that patients with HBV-associated HCC undergoing systemic antitumor therapy receive concomitant prophylactic antiviral therapy to suppress HBV replication.⁷ However, existing nucleos(t)ide analogs are unable to eradicate covalently closed circular DNA, the persistent transcriptional template for HBV in infected hepatocytes.⁸

The association between HBV DNA levels and HCC risk has been well studied, with a consensus that HBV DNA levels > 2000 IU/mL are significantly associated with hepatic inflammatory activity, fibrosis progression, and HCC risk. Therefore, comprehensive virologic surveillance, especially quantitative HBV DNA monitoring, is essential for patient risk stratification. Standardized management protocols, including viral surveillance and antiviral therapy, have been developed for this high-risk group.⁹ The pathological significance of low-level viremia (LLV⁹; defined as serum HBV DNA levels of 20–2000 IU/mL) in driving hepatic fibrosis and hepatocarcinogenesis remains controversial. A study by Sinn et al. involving patients with compensated cirrhosis demonstrated that patients with LLV were at a significantly higher risk of HCC development than those with undetectable HBV DNA.¹⁰ By contrast, Lee et al. reported no statistically significant association between LLV and HCC incidence in this population.¹¹

A limited number of studies have evaluated the prognostic implications of LLV in patients with HCC. Although a study by Kim et al. demonstrated no significant association between LLV and clinical outcomes, these findings may have been affected by substantial selection bias, specifically the underrepresentation of intermediate or advanced -stage patients, with only 31.3% of patients categorized as Barcelona Clinic Liver Cancer (BCLC) stage B/C/D, and the minimal use of systemic therapies, with only 2.5% of patients receiving sorafenib treatment.¹² As systemic treatment is now a mainstay of care, the present study investigated the prognostic impact of LLV and its contributing factors in patients with intermediate or advanced stage HBV-related HCC undergoing systemic therapy. Furthermore, we comprehensively evaluated the association between baseline HBV DNA levels and patient prognosis across the virological strata.

Methods

Study design and participants

This multicenter retrospective study was conducted at the Fifth Medical Center of the General Hospital of the Chinese People’s Liberation Army, Youan Hospital of Capital Medical University, and the Affiliated Hospital of Guizhou Medical University. All patients with HCC who attended the three study centers between January 2020 and January 2024 were screened for inclusion in the study. The inclusion criteria were as follows: (1) diagnosis of intermediate-to-advanced HCC, (2) acceptance of first-line systemic therapy, (3) seropositive for HBsAg, and (4) patients with BCLC stage C or BCLC stage B. The exclusion criteria were as follows: (1) co-infection with other viruses, (2) incomplete follow-up data, (3) follow-up interval of more than 2 months, (4) regular use of antiviral drugs prescribed during antitumor therapy, (5) HBV DNA ⩾2000 IU/mL after 4 months of follow-up, (6) OS of less than 4 months, and (7) received regional antitumor therapy at baseline. The patients were categorized into two groups according to their serum HBV DNA levels after at least 4 months of follow-up: LLV group, with serum HBV DNA levels of 20–2000 IU/mL on at least one occasion; and maintained virologic response (MVR) group, with serum HBV DNA persistently undetectable (<20 IU/mL). The reporting of this study conforms to Strengthening the reporting of observational studies in epidemiology (STROBE) (Supplemental Table 1).¹³

Related definitions and study endpoints

OS was measured from protocol-specified treatment commencement until all-cause mortality or censoring at the final evaluable time point (cutoff, January 2025). Progression-free survival (PFS) was defined as the time interval from baseline imaging to the earliest occurrence of radiologically confirmed disease progression (centrally blinded) or death from any cause according to RECIST 1.1. Antiviral history was defined as the receipt of regular antiviral therapy for at least 6 months prior to inclusion in the study. Baseline HBV DNA was defined as the level measured at enrollment, that is, immediately before the initiation of systemic antitumor therapy.

Data collection

Baseline data collected included information on sex, age, other population characteristics, serologic markers such as alpha-fetoprotein (AFP) and bilirubin, HBV DNA levels, radiography, antiviral regimens, and antitumor regimens. Thereafter, HBV DNA levels were measured, and abdominal imaging was performed every 1–2 months during follow-up.

Statistical analysis

Categorical and continuous variables are expressed as frequencies and mean ± SD or median and IQR (as applicable), respectively. Categorical variables were analyzed using the chi-squared test. Continuous variables were analyzed using the Student’s t-test or the Wilcoxon test, as appropriate. Prognosis was assessed using the Kaplan–Meier method. Logistic regression modeling was used to identify risk factors for LLV. In addition, Cox proportional hazards regression analyses were used to estimate adjusted HRs. A two-tailed p value ⩽0.05 was considered to indicate statistical significance. All statistical analyses were performed using SPSS software version 26.0 (IBM Corp., Armonk, NY, USA).

Results

Patient characteristics

Among all screened patients, 1814 met the inclusion criteria, 954 (52.6%) in the MVR group and 860 (47.4%) in the LLV group (Figure 1). Follow-up was discontinued in January 2025, with a median follow-up duration of 28 months (95% CI 26–30 months). The median age of the patients was 57 (IQR, 50–64) years; the majority were male (n = 1569; 86.5%). Of the 1814 patients, 965 (53.2%) had undetectable HBV DNA at baseline, and 849 (46.8%) had detectable HBV DNA at baseline. All patients received nucleos(t)ide analog antiviral therapy, primarily with entecavir (n = 1273; 70.2%).

Figure 1.

Patients flow chart.

Antitumor regimens typically involved either TKI monotherapy (n = 797; 43.9%) or TKI combined with programmed cell death protein 1 (PD-1) inhibitors (n = 1017; 56.1%). Six TKI monotherapies were used: lenvatinib (n = 443; 55.6%), sorafenib (n = 283; 35.5%), donafenib (n = 45; 5.6%), apatinib (n = 11; 1.4%), regorafenib (n = 10; 1.3%), and anlotinib (n = 5; 0.6%). The TKI combined with PD-1 inhibitor regimens included lenvatinib plus sintilimab (n = 409; 40.2%), lenvatinib plus camrelizumab (n = 151; 14.9%), and sorafenib plus sintilimab (n = 98; 9.6%). The baseline characteristics of the two patient groups are presented in Table 1 and Supplemental Figure 1.

Table 1.

Baseline characteristics of the study population.

Variables	MVR (n = 954)	LLV (n = 860)	p Value
Age (>50 years), n (%)	733 (76.8)	664 (77.2)	0.850
Male gender, n (%)	828 (86.8)	741 (86.2)	0.695
Diabetes, n (%)	156 (16.4)	114 (13.3)	0.064
Hypertension, n (%)	315 (33.0)	260 (30.2)	0.203
Smoking history, n (%)	421 (44.1)	379 (44.1)	0.979
Dinking history, n (%)	364 (38.2)	328 (38.1)	0.995
Vascular invasion, n (%)	403 (42.2)	422 (49.1)	0.004
Portal hypertension, n (%)	459 (48.1)	436 (50.7)	0.272
Cirrhosis, n (%)	729 (76.4)	678 (78.8)	0.217
AFP (>400 ng/ml), n (%)	272 (28.5)	317 (36.9)	<0.001
HBeAg (seropositive), n (%)	124 (13.0)	230 (26.7)	<0.001
ALT (>40 U/L), n (%)	289 (30.3)	324 (37.7)	<0.001
AST (>40 U/L), n (%)	402 (42.1)	495 (57.6)	<0.001
Tumor size (>5 cm), n (%)	337 (35.3)	401 (46.6)	<0.001
Extrahepatic organ metastasis			0.003
Yes	225 (23.6)	255 (29.7)
No	729 (76.4)	605 (70.3)
Baseline HBV DNA level			<0.001
Detectable	322 (33.8)	527 (61.3)
Undetectable	632 (66.2)	333 (38.7)
Antiviral therapy type			0.571
Entecavir	675 (70.8)	598 (69.5)
Other	279 (29.2)	262 (30.5)
Antiviral history			0.429
Yes	582 (61.0)	509 (59.2)
No	372 (39.0)	351 (40.8)
Child-Pugh			0.097
Stage A	668 (70.0)	571 (66.4)
Stage B	286 (30.0)	289 (33.6)
BCLC			0.006
Stage B	414 (43.4)	319 (37.1)
Stage C	540 (56.6)	541 (62.9)
Antitumor therapy type			0.787
TKI	422 (44.2)	375 (43.6)
TKI + PD-1 inhibitors	532 (55.8)	485 (56.4)

AFP, alpha-fetoprotein; ALT, alanine aminotransferase; AST, aspartate aminotransferase; BCLC, Barcelona Clinic Liver Cancer; HBeAg, hepatitis B e antigen; LLV, low-level viremia; MVR, maintained virologic response; TKI, tyrosine kinase inhibitor; TKI plus PD-1 inhibitors, tyrosine kinase inhibitor plus programmed cell death protein-1 inhibitors; Tumor size, maximum tumor diameter.

Factors associated with LLV

As shown in Table 1, LLV was positively correlated with AFP > 400 ng/mL, HBeAg seropositivity, ALT > 40 U/L, AST > 40 U/L, tumor size > 5 cm, extrahepatic organ metastasis, detectable baseline HBV DNA, and BCLC stage C. These factors were therefore included in a multifactorial analysis (Supplemental Table 2), which identified HBeAg seropositivity (OR 1.971, 95% CI 1.53–2.541, p < 0.001), AST > 40 U/L (OR 1.437, 95% CI 1.126–1.832, p = 0.004), extrahepatic organ metastasis (OR 1.640, 95% CI 1.187–2.266, p = 0.003), and detectable baseline HBV DNA (OR 2.482, 95% CI 2.002–3.077, p < 0.001) as independent risk factors for LLV.

LLV and prognosis

Univariate and multivariate analyses of PFS and OS were performed using Cox regression analysis (Table 2). Univariate analyses showed that vascular invasion, AFP > 400 ng/mL, AST > 40 U/L, detectable baseline HBV DNA, Child-Pugh Class B, BCLC stage C, and LLV were associated with reduced PFS. Reduced OS was correlated with the same factors, and the additional factor was tumor size >5 cm. Multifactorial analysis revealed that PFS was reduced in patients with AFP > 400 ng/mL (HR 1.173, 95% CI 1.032–1.332, p = 0.015), AST > 40 U/L (HR 1.191, 95% CI 1.049–1.352, p = 0.007), LLV (HR 1.656, 95% CI 1.468–1.868, p < 0.001) and that OS was reduced in those with AFP > 400 ng/mL (HR 1.237, 95% CI 1.062–1.441, p = 0.006), tumor size > 5 cm (HR 1.240, 95% CI 1.060–1.450, p = 0.007), Child-Pugh Class B (HR 1.380, 95% CI 1.186–1.607, p < 0.001), BCLC stage C (HR 1.421, 95% CI 1.058–1.907, p < 0.019), and LLV (HR 1.506, 95% CI 1.299–1.746, p < 0.001). In addition, a comparison of the prognosis of patients in the LLV and MVR groups using Kaplan–Meier curves revealed that the median OS (25.5 vs 37.8 months, p < 0.001) and PFS (8.3 vs 14.3 months, p < 0.001) were significantly shorter in the LLV group than in the MVR group (Figure 2).

Table 2.

Analysis of prognostic risk factors for patients receiving systemic therapy.

Univariate analysis	Progression-free survival		Overall survival
Univariate analysis	HR (95% CI)	p value	HR (95% CI)	p value
Age (>50 years)	0.917 (0.800–1.052)	0.216	0.903 (0.766–1.065)	0.225
Male gender	0.919 (0.775–1.090)	0.333	1.152 (0.944–1.405)	0.165
Diabetes	1.003 (0.854–1.179)	0.969	0.968 (0.798–1.175)	0.744
Hypertension	1.067 (0.944–1.205)	0.300	0.889 (0.765–1.032)	0.123
Smoking history	1.071 (0.955–1.202)	0.241	0.947 (0.823–1.089)	0.443
Dinking history	0.991 (0.881–1.116)	0.887	0.906 (0.784–1.046)	0.178
Vascular invasion	1.261 (1.123–1.415)	<0.001	1.693 (1.473–1.946)	<0.001
Portal hypertension	1.106 (0.986–1.240)	0.086	1.147 (0.999–1.318)	0.052
Cirrhosis	0.991 (0.864–1.137)	0.894	1.103 (0.930–1.307)	0.260
AFP (>400 ng/ml)	1.264 (1.118–1.428)	<0.001	1.523 (1.318–1.759)	<0.001
HBeAg (seropositive)	1.002 (0.865–1.160)	0.982	1.127 (0.950–1.338)	0.171
ALT (>40 U/L)	1.118 (0.990–1.262)	0.071	1.079 (0.931–1.251)	0.312
AST (>40 U/L)	1.317 (1.174–1.478)	<0.001	1.523 (1.325–1.751)	<0.001
Tumor size (>5 cm)	1.120 (0.995–1.261)	0.061	1.646 (1.431–1.894)	<0.001
Extrahepatic organ metastasis	1.067 (0.938–1.213)	0.324	1.450 (1.250–1.681)	<0.001
Baseline HBV DNA level (detectable)	1.215 (1.083–1.365)	0.001	1.610 (1.401–1.851)	<0.001
Antiviral therapy type Entecavir	1.022 (0.902–1.157)	0.735	0.964 (0.829–1.122)	0.636
Child-Pugh (B)	1.138 (1.002–1.292)	0.046	1.527 (1.318–1.769)	<0.001
BCLC (C)	1.191 (1.059–1.340)	0.004	1.859 (1.600–2.161)	<0.001
LLV	1.718 (1.530–1.929)	<0.001	1.743 (1.514–2.005)	<0.001
TKI	1.021 (0.909–1.146)	0.729	0.926 (0.805–1.065)	0.280
Multivariate analysis
Vascular invasion	1.152 (0.964–1.377)	0.120	1.059 (0.826–1.358)	0.651
AFP (>400 ng/ml)	1.173 (1.032–1.332)	0.015	1.237 (1.062–1.441)	0.006
AST (>40 U/L)	1.191 (1.049–1.352)	0.007	1.145 (0.982–1.337)	0.085
Baseline HBV DNA level (detectable)	0.940 (0.825–1.071)	0.350	1.113 (0.948–1.307)	0.192
Child-Pugh (B)	1.044 (0.915–1.191)	0.523	1.380 (1.186–1.607)	<0.001
BCLC (C)	0.988 (0.826–1.182)	0.895	1.421 (1.058–1.907)	0.019
LLV	1.656 (1.468–1.868)	<0.001	1.506 (1.299–1.746)	<0.001
Tumor size (>5 cm)			1.240 (1.060–1.450)	0.007
Extrahepatic organ metastasis			1.136 (0.927–1.391)	0.219

AFP, alpha-fetoprotein; ALT, alanine aminotransferase; AST, aspartate aminotransferase; BCLC, Barcelona Clinic Liver Cancer; CI, confidence interval; HBeAg, hepatitis B e antigen; HR, hazard ratio; LLV, low-level viremia; TKI, tyrosine kinase inhibitor; Tumor size, maximum tumor diameter.

Figure 2.

K-M survival curves were used to analyze OS and PFS in patients in the LLV and MVR groups.

Subgroup analysis of LLV and prognosis

These results indicate that patients with LLV have a poorer prognosis than those with MVR. To exclude the influence of the antitumor regimen on the results, we divided the patients into subgroups according to treatment regimen. The median OS of the LLV group remained significantly shorter than that of the MVR group in patients treated with TKI monotherapy (20.9 vs 37.8 months, p < 0.001), lenvatinib monotherapy (20.0 vs 32.1 months, p = 0.001), a TKI plus a PD-1 inhibitor (23.1 vs 39.6 months, p < 0.001), and lenvatinib plus sintilimab (23.1 vs 44.7 months, p < 0.001). Similarly, the median PFS of the LLV group was significantly shorter than that of the MVR group in all four of these subgroups (Supplemental Figure 2). We further stratified patients by BCLC stage into BCLC-B and BCLC-C cohorts and compared outcomes between the LLV and MVR groups within each subset. In both BCLC-B and BCLC-C, patients with LLV exhibited significantly shorter progression-free survival and overall survival than those with MVR (Supplemental Figure 3).

Patients were also divided into subgroups according to baseline HBV DNA load. Patients in the LLV group had significantly reduced OS compared to those in the MVR group in patients with undetectable baseline HBV DNA (26.3 vs 43.3 months, p < 0.001) and in all patients with detectable baseline HBV DNA (20.1 vs 28.7 months, p < 0.001). Patients with detectable baseline HBV DNA were further divided into four subgroups based on the baseline HBV DNA level (20–500, >500, 20–2000, and ⩾2000 IU/mL). In all four subgroups, the OS of patients in the LLV group was significantly shorter than that of patients in the MVR group (Figure 3). Similarly, the LLV group had a significantly shorter median PFS than the MVR group in all six of these subgroups (Supplemental Figure 4).

Figure 3.

K-M survival curves were used to analyze the effect of LLV and MVR on OS in different subgroups of patients: baseline HBV DNA undetectable (a), baseline HBV DNA detectable (b), baseline HBV DNA = 20–500 IU/mL (c), baseline HBV DNA=20–2000 IU/mL (d), baseline HBV DNA > 500 IU/mL (e), baseline HBV DNA > 2000 IU/mL (f).

Baseline HBV DNA and prognosis

As shown in Table 2, univariate analyses revealed that detectable baseline HBV DNA was associated with shorter OS and PFS; however, in multivariate analyses, this association was offset by other factors. A further multifactorial analysis that did not consider LLV as an influencing factor (Supplemental Table 3) showed that detectable baseline HBV DNA significantly reduced the OS of patients compared with undetectable baseline HBV DNA (HR 1.239, 95% CI 1.061–1.448, p = 0.007). Baseline HBV DNA load subgroup analysis demonstrated that although patients with detectable baseline HBV DNA had a shorter OS and PFS than those of patients with undetectable baseline HBV DNA, among patients with detectable baseline HBV DNA, associations of loads of ⩾500 versus <500 IU/mL and ⩾2000 versus <2000 IU/mL showed no significant association between baseline HBV DNA load and patient prognosis (Figure 4).

Figure 4.

K-M survival curves were used to analyze the effect of baseline HBV DNA levels on patient OS and PFS. Patients were stratified into two cohorts based on HBV DNA detection status (detectable vs undetectable), followed by comparative analysis of OS and PFS between the groups (a, d). In patients with detectable HBV DNA, the effects of high versus low HBV DNA load on OS and PFS were compared in subgroup analyses with different HBV DNA levels (b, e, c, f).

Dynamic changes in HBV DNA load affect the ability of HBV DNA load to predict prognosis. As shown in Supplemental Figure 5, of patients with detectable baseline HBV DNA, HBV DNA loads became undetectable in 377 (44.4%) and 455 (53.6%) at the first and second follow-up visits, respectively. HBV DNA levels were significantly lower during follow-up than at baseline. In addition, as shown in Supplemental Figure 6, the median baseline HBV DNA load was significantly higher in patients without antiviral history than in patients with antiviral history (5030 vs 200 IU/mL). After two follow-up visits, the median HBV DNA level was zero in both groups, and the proportion of patients with undetectable HBV DNA was almost the same in both groups (53.2% vs 54.1%).

Discussion

This study comprehensively evaluated the prognostic impact of LLV in patients with intermediate or advanced HBV-related HCC undergoing systemic therapy. Our findings demonstrated significantly shorter median PFS and OS in patients with LLV compared to those with MVR. These associations remained robust through Cox regression analyses and subgroup analyses stratified by antitumor regimens and baseline HBV DNA levels. Multivariate analyses revealed no association between baseline HBV DNA load and patient prognosis, which was further supported by subgroup analyses according to detectable baseline HBV DNA levels.

LLV is defined only in the guidelines of the American Association for the Study of Liver Diseases (2018), as detectable HBV DNA after 48 weeks (12 months) of antiviral therapy and HBV DNA < 2000 IU/mL.⁹ However, in the present study, we modified this criterion to a minimum follow-up duration of 4 months. This adjustment was necessitated by the unique clinical characteristics of intermediate-to-advanced HCC patients. First, previous prospective studies have reported median PFS of 4.3–8.2 months in this population.^6,14 A longer minimum follow-up period would result in therapeutic modifications in approximately 50% of patients during observation, thereby confounding LLV stratification. Second, the median OS for intermediate-to-advanced HCC ranges from 10.7 to 21.2 months.^5,14 Implementing a 12-month follow-up requirement would exclude nearly half of potential enrollees, substantially limiting the prognostic utility of LLV assessment. Notably, even with this shortened 4-month criterion, 469 patients (25.9%) died or were lost to follow-up within the initial 4-month period according to our data. These considerations collectively justify our adoption of a 4-month rather than 48-week follow-up threshold.

The incidence of LLV in the present study was 47.4%, within the wide range of 32.0%–69.0% reported in previous studies.^11,12,15,16 The limited reference guidelines available have led to inconsistencies in the definition of LLV and, therefore, large differences in reported LLV incidences. Reviewing the literature revealed the main reasons for the variation in these reported incidences of LLV as differences in baseline HBV DNA levels, primary disease status, patient enrollment rules, and antiviral status. First, baseline HBV DNA levels in previous studies were categorized as ⩾2000 and < 2000 IU/mL. In our study, additional groups were used, with varying incidences of LLV: undetectable, 34.5%; detectable, 62.1%; <2000 IU/mL, 42.1%; 20–2000 IU/mL, 56.1%; and ⩾2000 IU/mL, 70.4%. Patients with higher baseline HBV DNA levels had a higher incidence of LLV than those with lower baseline HBV DNA levels (Supplemental Table 4). Second, with the exception of the study by Tae-Se et al., previous studies on LLV have focused on patients with compensated cirrhosis without HCC.¹² The patients included in the study by Tae-Se et al. were predominantly BCLC stage O/A (73.1%), whereas the patients with HCC included in our study were BCLC stage B/C. Third, in terms of patient enrollment rules, the study by Jung et al. had the same enrollment rules as our study: regardless of the detection of baseline HBV DNA, patients could be classified as LLV or MVR during follow-up.¹⁵ However, in most studies, only patients with detectable baseline HBV DNA could be classified as LLV. Fourth, previous studies often varied in the proportion of patients receiving antiviral therapy. Moreover, it is difficult to achieve uniformity in antiviral treatment regimens among patients undergoing antiviral therapy.

In the present study, PFS and OS were significantly shorter in the LLV group than in the MVR group, even after subgroup and Cox regression analyses. However, in the study by Tae-Se et al., although the 5-year survival rate of patients in the LLV group was lower than that of patients in the MVR group (67.3% vs 74.3%), the difference was not statistically significant (p = 0.076).¹² The association between LLV and the risk of HCC in patients with cirrhosis is controversial; Jiwon et al. found that patients in the LLV group had a higher risk of HCC than those in the MVR group (adjusted HR 2.36),¹⁷ whereas Daniel et al. reported a comparable HCC risk between the two groups.¹⁶ Despite this controversy, the conclusion that LLV affects the prognosis of patients with HCC is generally reliable.

The association between the baseline HBV DNA load and oncological prognosis also remains controversial. Some studies have reported that in patients treated with various approaches, such as sorafenib and hepatic resection, those with higher baseline HBV levels have a poorer prognosis than those with lower baseline HBV levels.^18,19 By contrast, some studies have shown that patients with high and low baseline HBV DNA levels have a similar prognosis following systemic antitumor therapy, regardless of whether the threshold was set at 2000 or 500 IU/mL.^20,21 This instability in the association between baseline HBV DNA levels and patient prognosis is mainly due to two reasons. First, HBV DNA levels are highly variable. Baseline levels were higher among patients with detectable baseline HBV DNA and no history of antiviral treatment than those with a history of antiviral therapy (median, 5030 vs 200 IU/mL). However, at the second follow-up visit after concurrent antitumor and antiviral therapy, median HBV DNA levels were similar in both groups, with similar rates of conversion to undetectable HBV DNA status. Second, transiently elevated HBV DNA levels have limited effects on disease progression. The prognosis of patients with high baseline HBV DNA levels can improve after viral reduction, whereas those with low baseline levels may experience disease exacerbation upon viral rebound.^22,23 Our data suggested that patients in the LLV group had worse PFS and OS than those in the MVR group, regardless of baseline HBV DNA levels, and we found no significant differences in the prognosis of patients with high and low baseline HBV DNA levels (Supplemental Figure 6). Therefore, in clinical practice, LLV/MVR should be preferred to baseline HBV DNA levels as a predictive indicator of prognosis in patients with HCC.

This study has several limitations. First, as this is a retrospective study, there may be bias. However, we attempted to overcome this by including a large patient cohort from multiple centers and by performing subgroup analyses. Second, we included only patients with BCLC stage B/C. Although this reduces the prognostic impact of differences in disease severity, data on the prognostic impact of LLV in patients with early-stage HCC are lacking. Third, the imaging follow-up time window was non-uniform, which may have introduced bias into the PFS evaluation results. Fourth, we included patients with a minimum follow-up period of only 4 months, shorter than the period specified in previous similar studies. Although this decision was made based on factors related to PFS and OS of patients with intermediate or advanced HCC, there may be issues associated with the short follow-up period. Given these limitations, further prospective, multidisease, multistratified studies are required to optimize the surveillance, prevention, and management of patients with LLV.

Conclusion

The incidence of LLV is high in patients with intermediate or advanced HBV-related HCC. The main independent risk factors for LLV are HBeAg seropositivity, AST levels > 40 U/L, extrahepatic organ metastasis, and detectable baseline HBV DNA levels. Compared with patients with MVR, those with LLV experienced reduced PFS and OS. Predicting patient prognosis based on viral persistence status is more valuable than relying solely on baseline HBV DNA levels.

Supplemental Material

sj-docx-1-tam-10.1177_17588359261435330 – Supplemental material for Low-level HBV viremia independently predicts poor outcomes in patients with intermediate-to-advanced HBV-related hepatocellular carcinoma receiving systemic therapy: a multicenter retrospective study

Supplemental material, sj-docx-1-tam-10.1177_17588359261435330 for Low-level HBV viremia independently predicts poor outcomes in patients with intermediate-to-advanced HBV-related hepatocellular carcinoma receiving systemic therapy: a multicenter retrospective study by Jin Lei, Huaxing Ma, Xiong Chen, Ling Lin, Zhipeng Liang, Guangling Ou, Jiliang Jin, Hongyuan Dai, Ya Zhang, Mei Liu, Haiyang Li and Yinying Lu in Therapeutic Advances in Medical Oncology

Supplemental Material

sj-docx-2-tam-10.1177_17588359261435330 – Supplemental material for Low-level HBV viremia independently predicts poor outcomes in patients with intermediate-to-advanced HBV-related hepatocellular carcinoma receiving systemic therapy: a multicenter retrospective study

Supplemental material, sj-docx-2-tam-10.1177_17588359261435330 for Low-level HBV viremia independently predicts poor outcomes in patients with intermediate-to-advanced HBV-related hepatocellular carcinoma receiving systemic therapy: a multicenter retrospective study by Jin Lei, Huaxing Ma, Xiong Chen, Ling Lin, Zhipeng Liang, Guangling Ou, Jiliang Jin, Hongyuan Dai, Ya Zhang, Mei Liu, Haiyang Li and Yinying Lu in Therapeutic Advances in Medical Oncology

Supplemental Material

sj-docx-3-tam-10.1177_17588359261435330 – Supplemental material for Low-level HBV viremia independently predicts poor outcomes in patients with intermediate-to-advanced HBV-related hepatocellular carcinoma receiving systemic therapy: a multicenter retrospective study

Supplemental material, sj-docx-3-tam-10.1177_17588359261435330 for Low-level HBV viremia independently predicts poor outcomes in patients with intermediate-to-advanced HBV-related hepatocellular carcinoma receiving systemic therapy: a multicenter retrospective study by Jin Lei, Huaxing Ma, Xiong Chen, Ling Lin, Zhipeng Liang, Guangling Ou, Jiliang Jin, Hongyuan Dai, Ya Zhang, Mei Liu, Haiyang Li and Yinying Lu in Therapeutic Advances in Medical Oncology

Footnotes

Acknowledgements

We thank the patients and their families.

Declarations

ORCID iD

Jin Lei

Supplemental material

Supplemental material for this article is available online.

References

Sung

Ferlay

Siegel

, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2021; 71(3): 209–249.

EASL Clinical Practice Guidelines: management of hepatocellular carcinoma. J Hepatol 2018; 69(1): 182–1236.

Practice guidelines for acute pain management in the perioperative setting. A report by the American Society of Anesthesiologists Task Force on pain management, acute pain section. Anesthesiology 1995; 82(4): 1071–1081.

EASL Clinical Practice Guidelines on the management of hepatocellular carcinoma. J Hepatol 2025; 82(2): 315–374.

Llovet

Ricci

Mazzaferro

, et al. Sorafenib in advanced hepatocellular carcinoma. N Engl J Med 2008; 359(4): 378–390.

Cheng

Qin

Ikeda

, et al. Updated efficacy and safety data from IMbrave150: atezolizumab plus bevacizumab vs. sorafenib for unresectable hepatocellular carcinoma. J Hepatol 2022; 76(4): 862–873.

Hwang

Feld

Hammond

, et al. Hepatitis B virus screening and management for patients with cancer prior to therapy: ASCO provisional clinical opinion update. J Clin Oncol 2020; 38(31): 3698–3715.

Lok

Zoulim

Dusheiko

, et al. Hepatitis B cure: from discovery to regulatory approval. Hepatology (Baltimore, Md) 2017; 66(4): 1296–1313.

Terrault

Lok

ASF

McMahon

, et al. Update on prevention, diagnosis, and treatment of chronic hepatitis B: AASLD 2018 hepatitis B guidance. Hepatology 2018; 67(4): 1560–1599.

10.

Sinn

Lee

Goo

, et al. Hepatocellular carcinoma risk in chronic hepatitis B virus-infected compensated cirrhosis patients with low viral load. Hepatology 2015; 62(3): 694–701.

11.

Lee

Park

Lee

, et al. Episodic detectable viremia does not affect prognosis in untreated compensated cirrhosis with serum hepatitis B virus DNA < 2000 IU/mL. Am J Gastroenterol 2022; 117(2): 288–294.

12.

Kim

Sinn

Kang

, et al. Hepatitis B virus DNA levels and overall survival in hepatitis B-related hepatocellular carcinoma patients with low-level viremia. J Gastroenterol Hepatol. 2019; 34(11): 2028–2035.

13.

von Elm

Altman

Egger

, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet 2007; 370(9596): 1453–1457.

14.

Llovet

Kudo

Merle

, et al. Lenvatinib plus pembrolizumab versus lenvatinib plus placebo for advanced hepatocellular carcinoma (LEAP-002): a randomised, double-blind, phase 3 trial. Lancet Oncol. 2023; 24(12): 1399–1410.

15.

Kim

Sinn

Kang

, et al. Low-level viremia and the increased risk of hepatocellular carcinoma in patients receiving entecavir treatment. Hepatology 2017; 66(2): 335–343.

16.

Huang

Tamaki

Lee

, et al. Outcome of untreated low-level viremia versus antiviral therapy-induced or spontaneous undetectable HBV-DNA in compensated cirrhosis. Hepatology 2023; 77(5): 1746–1756.

17.

Yang

Choi

Shim

, et al. Low level of hepatitis B viremia compared with undetectable viremia increases the risk of hepatocellular carcinoma in patients with untreated compensated cirrhosis. Am J Gastroenterol 2023; 118(6): 1010–1018.

18.

Yang

Wen

, et al. A high baseline HBV load and antiviral therapy affect the survival of patients with advanced HBV-related HCC treated with sorafenib. Liver Int 2015; 35(9): 2147–2154.

19.

Sun

Liu

Wang

Correlation between low-level viremia and hepatitis B-related hepatocellular carcinoma and recurrence: a retrospective study. BMC Cancer 2021; 21(1): 1103.

20.

Sun

Yang

, et al. Baseline HBV loads do not affect the prognosis of patients with hepatocellular carcinoma receiving anti-programmed cell death-1 immunotherapy. J Hepatocell Carcinoma 2020; 7: 337–345.

21.

Wang

Zhang

, et al. Association of hepatitis B virus DNA levels with overall survival for advanced hepatitis B virus-related hepatocellular carcinoma under immune checkpoint inhibitor therapy. Cancer Immunol Immunother 2023; 72(2): 385–395.

22.

Chen

, et al. Entecavir plus Biejia-Ruangan compound reduces the risk of hepatocellular carcinoma in Chinese patients with chronic hepatitis B. J Hepatol 2022; 77(6): 1515–24.

23.

Lei

Yan

Zhang

, et al. Comparison of hepatitis B virus reactivation in hepatocellular carcinoma patients who received tyrosine kinase inhibitor alone or together with programmed cell death protein-1 inhibitors. Hepatol Int 2023; 17(2): 281–290.