Correlation between low-level viremia and hepatitis B-related hepatocellular carcinoma and recurrence: a retrospective study

Hepatitis B viral (HBV) infection remains a serious global public health problem. HBV infection is the leading cause of hepatocellular carcinoma (HCC) worldwide, accounting for 33% of cases [1]. In China, chronic hepatitis B (CHB) contributes to approximately 84% of HCC [2]. The incidence of HBV-related deaths due to liver cirrhosis and/or HCC dramatically increased between 1990 and 2013 [3], resulting in a large health and economic burden. Therefore, effective control of CHB is crucial. Serum HBV DNA level is essential for informing decision-making regarding antiviral treatment and the subsequent monitoring of disease progression. A prospective cohort study showed that an HBV DNA level more than 10,000 copies/mL was a strong predictor of HCC development [4]. Long-term, maintained virologic response (VR) is associated with a lack of progression, even in those with decompensated liver cirrhosis [5]. Therefore, an important goal of antiviral treatment is to obtain VR.

Low-level viremia (LLV) can be detected in CHB patients including those taking antiviral therapy, generally referring to detectable HBV DNA levels lower than 2000 IU/mL. Recently, an increasing number of studies have shown that LLV plays a crucial role in the diagnosis and prognosis of HCC. It has been reported that LLV is associated with a higher risk of HCC and poorer overall survival, compared with those who maintained VR [6, 7]. LLV patients with a relatively low viral load can still benefit from effective antiviral therapy, showing as significantly decreased risk of HCC [7]. More importantly, antiviral therapy significantly reduced HCC recurrence after R0 hepatic resection [8].

Although the exact incidence of LLV is not known, the potential LLV population is likely considerable. The timing of antiviral therapy for LLV is controversial. Generally, LLV without liver cirrhosis or elevated ALT is considered inactive CHB, and is not an indication for antiviral therapy [2, 9]. However, due to the risk for HCC and potential benefits of reduced risk of HCC recurrence, it may be necessary for LLV patients to receive antiviral therapy at an earlier CHB stage and to maintain a long-term VR with undetectable HBV DNA loads. Continuous monitoring of cirrhosis and HCC is also important during routine follow-up.

Characteristics of CHB have been well described, and studies have revealed common characteristics of LLV. However, an overall understanding of LLV is lacking. Thus, there is still an urgent need to describe LLV characteristics and distinguish LLV patients from other CHB patients. This would help inform suitable management strategies for LLV patients and improve early detection of HCC. In this study, we explored the characteristics of LLV in HCC patients who had undergone curative hepatectomy. Then, prognostic factors were further analyzed in an effort to understand the differential characteristics of LLV patients versus those with high-level viral loads.

In this cohort of CHB patients with curative hepatectomy, most LLV patients achieved complete VR during follow-up and experienced longer recurrence-free survivals. Post-operative serum HBV DNA levels > 100 IU/mL were an independent prognostic factor for HCC recurrence after curative hepatectomy, confirming that complete VR after curative hepatectomy could dramatically improve tumor recurrence-free survivals. Though multivariate analysis showed a lower risk of HCC recurrence in those with post-operative serum HBV DNA levels < 100 IU/mL, there was still a risk of fibrosis progression and HCC [6, 11]. There were 11 patients with preoperative LLV showed virological breakthrough during follow-up. Among them, post-operative HBV DNA loads ≥2000 IU/mL were observed in seven patients (four patients without any record of antiviral treatment), with shorter tumor recurrence-free survivals (less than 6 months in five patients). These results indicate that LLV without effective antiviral treatment is a risk factor for HCC recurrence. Effective antiviral treatment and complete VR are important for LLV patients.

Serum HBsAg levels are an important characteristic of CHB and useful for identifying stage of disease and effects of nucleos(t)ide analogues therapy [12, 13]. More importantly, HBsAg levels are associated with the risk of progression to HCC, especially in LLV patients with negative HBeAg [14]. In this cohort, a total of 167 patients showed serum HBsAg concentrations > 250 IU/mL, possibly reflecting high amount and enhanced transcriptional activity of covalently closed circular DNA inside the hepatocytes [15]. As we know, HBsAg declines very slowly in the most cases through the nature course, even HBV DNA has been undetectable [16]. Therefore, serum HBsAg levels could be an important indicator for monitoring antiviral treatment in those with LLV or even undetectable HBV DNA. Initial therapeutic target may be at a level of 100 IU/mL, because serum HBsAg levels less than 100 IU/mL suggested a high probability of spontaneous HBsAg clearance [15]. Unfortunately, we could not detect the exact quantitative HBsAg levels in these patients, limited to the test method in our laboratory. Further studies are needed to explore the role of serum HBsAg on CHB control and even in HCC development.

The main difference in tumor characteristics between the two groups was tumor diameter, an important prognostic factor in HCC patients [17]. Multivariate analysis verified that a tumor diameter ≥ 5 cm and intrahepatic metastasis were two independent prognostic factors for recurrence of HCC after curative hepatectomy. However, the cases of intrahepatic metastasis were not high in this cohort, probably due to the curative hepatectomy. Thus, tumor diameter seems a much more important characteristic among LLV patients in this cohort. As we showed, there were 67.3% LLV patients with tumor diameter < 5 cm, which might be related with a lower tumor recurrence rate. However, more studies are needed to confirm this. Besides, small tumor size may be related with less serious liver injury, showing as low-level liver enzymes in LLV patients.

The usage of AFP in diagnosing HCC is controversial. A cutoff value of 20 ng/mL shows high sensitivity but low specificity, whereas a high cutoff value at 200 ng/mL lowers sensitivity dramatically [18], leading to a non-negligible risk of misdiagnosis. Dynamic changes of AFP levels could also be observed in CHB patients, showing as elevated AFP levels before antiviral therapy and decreased AFP levels after antiviral therapy in patients with active CHB [19]. In this study, we used a lower cutoff value of 10 ng/mL and found the proportions of pre- and post-operative AFP levels < 10 ng/mL were high in both groups. More importantly, in patients with tumor recurrence, there were 35.7% (15/42) patients in group 1 and 27.1% (16/59) patients in group 2 showed AFP levels < 10 ng/mL, respectively. These results suggest that low AFP levels are not safe enough on excluding HCC and tumor recurrence, especially in LLV patients. The effectiveness of AFP at 10 ng/mL on screening HCC should be more rigorous evaluated in the future.

There were some limitations in this study. First, metabolism-related fatty liver disease is gradually becoming an important risk factor for HCC. Data on body weight and height were missing in some patients, therefore body mass index could not be calculated and compared between groups. Second, the follow-up time was not long enough in some patients. All these factors should be further studied.

LLV patients with HCC were characterized by smaller tumor size, less serious liver injury, and lower AFP levels compared to patients with HBV DNA ≥ 2000 IU/mL. Preoperative LLV was related with a long tumor recurrence interval, and complete VR after curative hepatectomy was associated with a lower risk of hepatocellular carcinoma recurrence. A more suitable management strategy should be established to reduce the incidence of HCC in LLV patients. HBsAg levels could be an important indicator for monitoring the effects of antiviral therapy in patients with LLV or even undetectable HBV DNA.