If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
After hepatitis B surface antigen (HBsAg) seroclearance, the risk of hepatocellular carcinoma (HCC) remains, and the optimal surveillance strategy has yet to be determined. Herein, we aimed to evaluate incidence and risk factors for HCC and establish a novel prediction model for HCC development after HBsAg seroclearance.
We thank Tang and colleagues for their strong interest in our work which develops and validates a risk prediction model for hepatocellular carcinoma (HCC) after hepatitis B surface antigen (HBsAg) seroclearance.1 Tang et al. have raised several considerations about the risk factors we used in our model.2
Currently, the optimal treatment endpoint of chronic hepatitis B (CHB) is seroclearance of HBsAg, which has been well-documented to significantly decrease the risk of developing HCC in patients with CHB.
In previous studies, several risk factors were identified, including cirrhosis or diabetes mellitus, male sex, and old age at HBsAg seroclearance. However, most of these studies were limited by insufficient population size and/or follow-up period. In this study, a total of 831 patients were enrolled, among whom 40 patients developed HCC during follow-up. Through univariate and multivariate Cox regression analyses, age at HBsAg seroclearance, cirrhosis, family history of HCC, and more-than moderate drinking were identified as independent factors associated with HCC development. Based on these independent risk factors, a prediction model for HCC development after HBsAg seroclearance was established. We appreciate that the authors provided a promising model to predict HCC after HBsAg seroclearance. Despite the strengths of this study, several important issues warrant further discussion.
In this study, the authors identified age at HBsAg seroclearance as an independent risk factor of HCC, which may be interpreted as exposure time to HBV in endemic regions as they stated. However, it should be noted that age itself is a potent factor for the development of most cancers, including HCC.
Given the large deviation of patients' age and long-term period of follow-up in this cohort, the onset of HCC is likely to be attributed in large part to advancing age despite the HBsAg seroclearance. Therefore, the age at HCC diagnosis or last follow-up should be categorized (e.g., >60 or not) and included as another potential risk factor in the analysis, which may further justify the effect of age at HBsAg seroclearance (i.e., exposure time to HBV) on HCC development.
In terms of alcohol consumption, in addition to self-reported alcohol use (two and one standard drink of alcohol per day for males and females, respectively), drinkers with serum gamma-glutamyltransferase (GGT) levels over 100 U/L and no evidence of obesity or cholestatic liver disease were also classified as having more-than-moderate alcohol consumption. Accurate estimation of alcohol consumption from questionnaires individually is always difficult, especially with the risk of underestimating heavy consumption. However, as a sensitive enzymatic indicator of liver disease, GGT may be upregulated by numerous pathophysiological conditions (e.g., latent infection, liver damage) with no signs of cholestatic liver disease via biological mechanisms such as oxidative stress, immunity, and metabolism,
making it questionable to use GGT as a surrogate marker to quantify alcohol intake. Consistently, several studies revealed that serum GGT level was poorly correlated with alcohol consumption, even in individuals without CHB.
Based on the health management database of our institution, we obtained a total of 22,338 consecutive patients with CHB after HBsAg seroclearance without evidence of obesity or cholestatic liver disease and found that only 19.7% of those with a GGT ≥100 U/L were heavy drinkers, while 28.9% reported no history of alcohol consumption (Fig. 1F). Furthermore, GGT proved ineffective in identifying heavy drinkers in the univariate regression model, neither in drinkers nor in the whole population (AUC 0.621 and 0.683, respectively, Fig. 1G,H). Thus, the GGT level is not necessarily dependent on alcohol consumption in patients with CHB who achieved HBsAg seroclearance and should be included in the regression analysis as an independent factor along with alcohol consumption. In any case, it would be helpful to know how many patients in this study were considered heavy drinkers based on their GGT scores.
Moreover, patients with CHB, particularly those with cirrhosis, were generally advised to abstain from alcohol to reduce disease-related deaths. This study found that patients with non or moderate alcohol consumption are at relatively low risk of developing HCC, which may lead to a potentially contentious conclusion. Therefore, it is necessary to clarify the effect of moderate alcohol consumption on HCC risk in comparison to those with no alcohol consumption. Even with limited event cases, data on this subgroup of moderate drinkers may benefit future research on the management of patients with CHB and alcohol dependence, allowing for a more practical recommendation.
This work was supported by West China Hospital, Sichuan University (2020HXBH039), and Sichuan University (2020SCU12025).
Y.T. and Y.C. contributed equally as joint first author to this manuscript. Prof. H.C. provided original data, and contributed to supervision and manuscript revision.
The study conformed to the Declaration of Helsinki and was approved by the Ethics Committee of West China Hospital.
Data availability statement
The data used to conduct the research are available from the corresponding author upon request.
Conflict of interest
The authors declare no conflicts of interest that pertain to this work.
Please refer to the accompanying ICMJE disclosure forms for further details.
The following are the supplementary data to this article: