Currently, all NAS approved to treat CHB have inhibitory effects on the HBV polymerase/reverse transcriptase, which has poor proofreading and editing ability, resulting in error-prone replication . Additionally, the high replication rate of HBV, continued existence of the replication template covalently closed circular (cccDNA) and complexity of quasispecies [27, 28] increase the likelihood of the development of drug resistance, especially during prolonged therapy. Once viral breakthrough or drug resistance occurs, rescue therapy was been initiated. But, rescue therapy can promote the development of multidrug mutations . Therefore, identification of novel treatment targets remains a major clinical challenge to improve the efficacy of antiviral therapy and prevent drug-resistance.
It has been demonstrated that the main mutations associated with LAM-resistance are rtM204I and/or rtL180M, and these do not confer resistance to ADV [29, 30]. The mutations associated with resistance ADV are mainly rtA181V and/or rtN236T . Recently, it has been reported that a single amino acid change at position rt181 may induce cross-resistance to LAM and ADV . However, this phenomenon has not been widely observed in clinical practice. The lack of cross-resistance has provided a rationale for the combination of LAM and ADV therapy, which may be the main reason why viral breakthough in the combination group was not observed in the study. In contrast, six patients in ETV monotherapy developed viral breakthrough and five patients had documented ETV–resistant mutations. Two cases had LAM-resistant mutations before onset of therapy (rtM204I/V ± rtL180M). Previous studies have confirmed that the presence of rtM204I/V and rtl180 M can reduce the genetic resistance barrier to ETV [33, 34]. Therefore, pretreatment resistance testing may be useful in detecting viral mutations. If ETV-resistance mutations are present, LAM + ADV combination therapy may be a better choice than ETV in terms of avoiding drug resistence.
The study by Zhang et al.  included patients with higher baseline HBV DNA level (HBV DNA load >107 copies/mL) than that in the current study. They found that the HBV DNA load decreased to less than 103 copies/mL in 2.9% in the combination group compared with 14.3% in monotherapy group at 12 weeks (P < 0.05). Yu et al.  also reported that the rate of patients who had HBV DNA load declines to less than 1 copy/ml in was lower in combination group than that in the monotherapy group at 12 weeks (3.7% vs. 18.0%, P = 0.018). In another study, Jayakumar et al.  found that the median decrease in HBV DNA levels in a combination group were 92.73% and 99.57% at 12 and 24 weeks of therapy, respectively (considering the baseline HBV DNA levels to be 100%). However, in an ETV group, the median decrease was 99.74% and 100%, respectively, in the same follow-up periods. The differences were highly significant at both 12 (P =0.0007) and 24 (P = 0.0115) weeks between the two groups. Furthermore, the results of the analysis prove that LAM plus ADV combination therapy produced more rapid and significant reduction in HBV DNA levels at 12 weeks of therapy, compared with ETV monotherapy, even though there was no significant difference observed at 96 weeks post-treatment. Unfortunately, the raw data were presented in the included studies as percentage of patients with undetectable HBV DNA instead of a decline in level of HBV DNA. Therefore a statistical analysis was not performed.
One potential benefit of rapid suppression of HBV replication is to reduce the risk of drug resistance. It has been shown that the subsequent chance of LAM resistance is directly proportional to the HBV DNA levels after 24 weeks of treatment . A study of LDT versus ADV showed that suppression of HBV DNA levels at 24 weeks correlated with efficacy outcomes and viral breakthrough at 1 year . Longer-term studies with ETV have reported that most patients achieve HBV DNA suppression within the first 24 weeks of treatment, and suppression is maintained for up to 4 years  with minimal drug resistance . Thus, the de novo combination LAM and ADV were expected to achieve a better outcome than that in ETV monotherapy in terms of multidrug resistance.
The current review has shown that the combination group had significantly higher rate of HBeAg seroconversion and ALT normalization compared with the ETV group up to 96 weeks. The high rate of HBeAg seroconversion may be related to the rapid reduction of HBV DNA load at week 12 as mentioned above. Other studies have verified that an early reduction of HBV DNA is associated with an increased likelihood of HBeAg seroconversion . He et al.  reported that at 24 weeks, the rates of HBV DNA negativity in the LAM and ADV combination group were higher than that in the LAM or ADV monotherapy group. At week 96, the percentages of patients with undetectable DNA and HBe seroconversion in the combination group (100%, and 51%, respectively) were superior to LAM (66%, 21%, respectively) or ADV alone groups (49%, 33%, respectively). Ghany et al.  also found that combination therapy with LAM and ADV was associated with a high rate of virological and biochemical response both at week 48 (59% vs. 26%, P = 0.06). At week 192, 76% of the combination vs. 36% of the monotherapy groups had loss of HBeAg (P = 0.03). The current treatment guidelines recommend that antiviral therapy be stopped for HBeAg-positive CHB patients (except those with cirrhosis) when either HBeAg is undetectable or when HBeAg seroconversion has taken place, and when HBV DNA is undetectable for an additional 6-12 months after continuous therapy . HBeAg loss or seroconversion may reduce the risk of developing HCC or progressive liver diseases  in HBeAg-positive CHB patients. This suggests that superiority of combination of LAM and ADV over ETV alone in term of HBeAg seroconversion may provide potential benefits in patients with HBeAg-positive CHB.
The potential for an increased risk of toxicity must always be noted when instituting combination LAM and ADV. In this meta-analysis, either combination group or ETV alone was well tolerated. Only one patient presented with an elevated BUN in the combination group. The levels were monitored closely and remained within normal limits during the treatment period. With prolonged durations of the combination treatment, not only effects, but also costs should be evaluated. Wu et al.  observed that the initiation of rescue therapies for LAM-resistant CHB with adding ADV is likely to be more cost-effective than ADV, ETV, TDF monotherapy. In the current study, a cost-effectiveness analysis was not done because costs of medications were not included.
The limitations of this review include the fact that some studies were not RCTs. Publication bias was also possible. Compared to positive studies, negative studies may be less likely to be published and more likely to take longer to be published. This could have affected the validity of the meta-analysis in this review . In addition, many studies were conducted in China resulting in regional and language bias. Finally, there were only a few studies and the sample size was small.
In conclusion, de novo combination of LAM and ADV therapy for naïve treated patients was not superior to the ETV monotherapy in short duration; however, the combination therapy had higher biochemical response and HBeAg seroconversion rates compared with monotherapy when the therapy duration was prolonged up to 96 weeks. The rate of emergence of viral resistance in combination group was less than that in the ETV group. However, given the limited number of studies included in the analysis, caution should be exercised in extrapolation of the conclusion to all patients infected with CHB. More high-quality, well-designed, randomized controlled, multicenter studies are clearly needed to confirm these observations.