EDITORS,In their letter, 1 Chen and colleagues pointed out that the patients in our study cohort were predominantly infected with HBV genotype A and D, whereas genotypes B and C were found less frequently. 2 They concluded that this distribution was most likely due to the European origin of our study population and that our results have to be examined in larger study populations with a higher prevalence of HBV genotypes B and C. We agree with this conclusion and we are excited to learn more on this topic in patients with genotype B and C infection as this group was naturally underrepresented in our study of patients living in Europe. Chen and colleagues presented data of 106 HBeAg-negative patients infected with HBV genotypes B and C "without anti-viral therapy before." The authors observed no differences in the HBsAg levels among these genotypes. As they also found no differences in earlier studies analysing patients with nucleos(t)ide analogues therapy, they concluded that genotypes B and C are not associated with differences in HBsAg levels. However, for meaningful interpretation of the genotype-specific impact on HBsAg levels the analysed patient group has to be described more precisely as HBsAg levels also depend strongly on the disease stage. [3][4][5] In the cited studies, 6,7 patients with presumable active hepatitis were included as they qualified for therapy. Both HBeAg-positive (n = 32 and 61, respectively) and HBeAg-negative patients (n = 16 and 33, respectively) were analysed and the overall patient numbers, especially for HBeAg-negative patients, were small. In their letter, the patient numbers are also limited (62 genotype C and 44 genotype B infected patients) and no demographic data were given. In detail, it is unclear if "without anti-viral therapy before" means that these patients were all inactive carriers with normal transaminases etc.without need for anti-viral treatment as in our study. Therefore, interpretation of these data seems to be difficult at this point.In addition, Chen et al presented data regarding the prevalence of the basal core promotor (BCP) A1762T/G1764A and precore (PC) mutation G1896A and their impact on HBsAg levels. The authors wrote that none of these mutations was significantly associated with HBsAg levels, although as shown in figure 1D of their study, statistically significantly more HBsAg was found in patients with PC mutation (P = 0.006, exact patient numbers are missing). In contrast, in our study the PC mutation was associated with lower HBsAg levels and this was also true after excluding patients with additional preS mutations from the analysis (3.0 vs 3.3 log IU/mL for patients with and without G1896A, respectively; P = 0.005). However, this effect seems to be mild and the clinical impact still needs to be explored.The observation that the BCP mutation has no impact on HBsAg levels is in line with our results.In conclusion, we agree with the authors that the impact of the genotype and frequent mutations on HBsAg levels has to be confirmed in studies with a high...
