J Hepatol. 2017 Mar 15. pii: S0168-8278(17)30136-8. doi: 10.1016/j.jhep.2017.03.002. [Epub ahead of print]
How Good Is Hepatitis B Surface Antigen Kinetics Informing Strategy to Stop Antiviral Therapy?
Brunetto M1, Cornberg M2, Chan HL3.
Author information
1
Department of Clinical and Experimental Medicine, University of Pisa - Hepatology Unit and Laboratory of Molecular Genetics and Pathology of Hepatitis Viruses - University Hospital of Pisa, Italy. Electronic address: [email protected].
2
Department of Gastroenterology, Hepatology and Endocrinology. Hannover Medical School; German Center for Infection Research (DZIF), partner site Hannover-Braunschweig, Germany. Electronic address: [email protected].
3
Department of Medicine and Therapeutics, Institute of Digestive Disease and State Key Laboratory of Digestive Disease, The Chinese University of Hong Kong, Hong Kong. Electronic address: [email protected].
Reply
We agree with W-J Jeng and Y-F Liaw on the pivotal
role of spontaneous hepatitis B surface
antigen (HBsAg) serum clearance in the clinical his
tory of a hepatitis B virus (HBV) carrier [1] as it
is, the hallmark of the control of HBV replication
and transition from inactive to occult infection
[2]. However, the reported annual rates of HBsAg clearance are highly variable (0.31 to 3.2 x 100 persons/years) because of the heterogeneity of the
studied populations [3–6]: mode of infection, age
and ethnicity are major factors influencing the HBs
Ag loss [5–8]. Furthermore, viral features as
HBV genotype could play a role as suggested by the
lower yearly HBsAg decline in genotype D
infected carriers (0.287 and 0.35 log10 IU/ml during 38 months and 9 years median follow-ups, respectively) as compared to genotype B/C infected
carriers (0.53 and 0.751 log10 IU/ml during the 5 and 3 years prior serum HBsAg loss, respectively) [9–12]. Finally, different HBsAg kinetics
were described according to residual viral replication: a progressive decline till HBsAg clearance
was seen in carriers with undetectable HBV DNA at variance with a sharp HBsAg drop about 2 years prior HBsAg clearance in carriers with detectable viremia [8]. Thus, the identification of a
universal threshold predictive of short term HBsAg
clearance would require the studies of a large
cohort of patients infected with different genotype
s where all the relevant variables are analyzed:
demographic features (age, sex and ethnicity), viral profile (history of HBeAg positivity,
HBeAg/anti-HBe seroconversion time, viral load), overall follow-up duration and yearly HBsAg
clearance rate. Accordingly, the scoring system developed on the HBeAg negative REVEAL cohort
shows that only the combination of several variable
s (age, BMI, HBsAg and HBV DNA serum
levels) can predict with adequate accuracy the probability of HBsAg clearance in the single carrier
[8]. On the other hand, acceptable positive predictive value and diagnostic accuracy are achievable
only at very low HBsAg levels, as 100 IU/ml proposed by several studies in Asia where genotype B
and C HBV are predominant [7,8,13-14]. In fact, when a higher HBsAg threshold (such as 200 IU/ml) is used, HBsAg kinetics have to be added to
improve the diagnostic performance [11-12].
This implies the need to monitor HBsAg levels over
time. Thus, the use of the latter approach
would not modify the current clinical practice since serum HBsAg monitoring would be required
anyway to predict and finally to show HBsAg clearance.
W-J Jeng and Y-F Liaw commented also on the issue of HBsAg quantification in the context of
nucleos(t)ide analogue (NA) therapy. They discussed their data that HBsAg decline during NA
therapy is stronger in patients with higher baseline alanine aminotransferase (ALT; >5x ULN) and
higher alfa-fetoprotein (AFP) levels. AFP reflects
cell turnover and may be another surrogate for
immune responses [15]. The data are in line with our discussion that the immune response but not
the mode of action of the NA is important for HBsAg
decline [16]. However, the value of baseline
parameter (ALT, AFP or IP-10) and on-treatment HBsAg level for the decision when to stop NA
treatment is still limited in our view. There are different stop NA approaches in different regions.
The APASL guideline suggests stopping NA therapy in
HBeAg negative patients after treatment for
at least 2 years with undetectable HBV DNA documented on three separate occasions, 6 months
apart [17]. This is in contrast to the EASL and AASLD guidelines that give a strong
recommendation treating all HBeAg negative patients
until HBsAg loss [2,18]. Demands for a cut-off identifying patients who should maintain on treatment or may stop earlier may be different in
different regions. As we have discussed in our review, HBsAg levels <100 IU/mL may be a good
predictor for maintained HBV DNA suppression after
stopping NA therapy [16]. Higher cut-offs of <150 IU/ml [19] or <200 IU/ml [20] have been discussed by other investigators. Other factors such
as age [19], previous therapies, kinetics of ALT levels or duration of consolidation therapy could
influence the risk of relapse after stopping NA treatment [21]. In our view, the current evidence
does not support a strong recommendation for stopping NA therapy at a certain cut-off at this stage.
To prevent clinical relapse, we suggested the lower
HBsAg cut-off of <100 IU/ml and a
consolidation therapy for at least 3 years in our review [16]. However, another strategy to stop NA
treatment may be to induce flares and increase HBsAg rates. Hadziyannis et al., have stopped
adefovir and observed ALT flares in 76% and subsequently 39% of patients lost HBsAg in the
long-term follow-up [22]. Höner zu Siederdissen et al. showed 20% HBsAg loss after stopping
long-term NA therapy in a small cohort of HBeAg negative patients. Interestingly, not HBsAg
levels at stop of therapy but the peak level of HBV
DNA and peak ALT during the relapse were the
best predictors for later HBsAg decline. The study
also showed that IP-10, IL-12, TNF and IL-10
were significantly induced after stopping NA therapy [23]. As discussed above, the key to achieve
HBsAg loss is the modulation of immune responses. An important task is now to analyze which
part of the immune response is fundamental for these effects. In conclusion, the topic to stop long-term NA therapy seems to be very interesting but still controversial. 作者: StephenW 时间: 2017-3-21 19:45