Occult hepatitis B virus infection in patients with chronic liver disease of different etiology in a Brazilian referral center: comparison of two different hepatitis B virus deoxyribonucleic acid amplification protocols: a cross-sectional study

ABSTRACT BACKGROUND: Occult hepatitis B virus infection (OBI) is defined as the presence of hepatitis B virus (HBV) deoxyribonucleic acid (DNA) in the liver of individuals with undetectable hepatitis B virus surface antigen (HBsAg) in the serum. The actual prevalence of OBI and its clinical relevance are not yet fully understood. OBJECTIVE: To evaluate the prevalence of HBV DNA in liver biopsies of HBsAg-negative patients with chronic liver disease of different etiologies in a referral center in Brazil and compare two different HBV DNA amplification protocols to detect HBV. DESIGN AND SETTING: This cross-sectional observational study was conducted at the Liver Outpatient Clinic, Hospital das Clínicas, Universidade Federal de Minas Gerais, Belo Horizonte, MG, Brazil, between January 2016 and December 2019. METHODS: HBV DNA was investigated in 104 liver biopsy samples from individuals with chronic liver disease of different etiologies, in whom HBsAg was undetectable in serum by nested-polymerase chain reaction (nested-PCR), using two different protocols. RESULTS: OBI, diagnosed by detecting HBV DNA using both protocols, was detected in 6.7% of the 104 individuals investigated. Both protocols showed a good reliability. CONCLUSION: In addition to the differences in the prevalence of HBV infection in different regions, variations in the polymerase chain reaction technique used for HBV DNA amplification may be responsible for the large variations in the prevalence of OBI identified in different studies. There is a need for better standardization of the diagnostic methods used to diagnose this entity.


INTRODUCTION
Hepatitis B virus (HBV) infection is one of the most prevalent infections worldwide and is an important cause of morbidity and mortality. It often progresses to chronic hepatitis, liver cirrhosis, and hepatocellular carcinoma (HCC) and is responsible for approximately 780,000 deaths annually. [1][2][3][4] HBV infection is usually diagnosed based on the presence of the HBV surface antigen (HBsAg) in the serum. However, the possibility of persistence of the HBV genome in HBsAgnegative individuals has been demonstrated. This entity termed occult hepatitis B virus infection (OBI), is defined by the presence of HBV deoxyribonucleic acid (DNA) in the liver (in some cases, also in the serum) in the absence of circulating HBsAg. 1,5,6 When HBV DNA is detectable in the serum, its levels are usually very low (< 200 IU/mL). It has been hypothesized that OBI is related to strong suppression of viral activity by host immune surveillance.
From a biomolecular perspective, different mechanisms may be involved in OBI development: mutations in the HBsAg gene, epigenetic changes, host immune responses, human immunodeficiency virus (HIV) and hepatitis C virus (HCV) coinfections, metabolic factors, HBV immune complexes, and genomic integration. [7][8][9][10][11][12] Moreover, there is evidence that microRNAs (miRNAs) are differentially expressed in patients with OBI compared to healthy controls. 13 The exact magnitude, pathogenesis, and clinical relevance of OBI are not completely understood. Individuals with this entity can transmit HBV through blood transfusion or organ transplantation. 1,5,6,11,14,15 In the setting of immunosuppression, the suppressed state of viral activity observed in OBI can be discontinued, leading to the development of typical hepatitis B, which often has a severe course. 16,17 Observational data suggest that OBI may favor or accelerate the progression of other chronic liver diseases, such as HCV infection, 18 and HCC development. 1,6,19,20 The diagnosis of OBI has been established using polymerase chain reaction (PCR) to amplify HBV DNA. Modifications to the PCR technique (nested-PCR and real-time PCR) were used to increase the sensitivity of the method. PCR assays vary in sensitivity and specificity, and the factors associated with the biological material in which the DNA is probed may affect HBV detection rate. Thus, the diagnosis of OBI remains challenging because there is no standard method or protocol for the detection of occult HBV DNA. 21 In Brazil, few studies have evaluated the prevalence of OBI using current case definition criteria. [24][25][26][27]

OBJECTIVE
In this context, the present study aimed to investigate the frequency of OBI in patients with chronic liver disease who underwent liver biopsy as part of the investigation of their disease and to compare two different HBV DNA amplification protocols for HBV detection. Patients were grouped according to the etiology of the underlying liver disease as follows: chronic liver disease associated with HCV, nonalcoholic steatohepatitis, autoimmune liver disease (autoimmune hepatitis, primary biliary cholangitis, and primary sclerosing cholangitis), cryptogenic liver disease and hemochromatosis.

Representativeness of liver biopsies
The representativeness of liver biopsies was assessed based on fragment size and number of portal tracts. The size distribution of the fragments was very close to a normal curve, with a mean size of approximately 13 mm.
The distribution of the number of portal tracts, unlike the biopsy size, showed wide variability with a skewed distribution, despite the higher concentration around the eight portal tracts (Figure 1).
The quality of the DNA present in the samples was analyzed using the A260/A280 ratio. Nucleic acids absorb light with a wavelength of 260 nm. Proteins absorb light with a wavelength of 280 nm.
Thus, the A260/A280 ratio provides a parameter for evaluating the quality of nucleic acid preparation. DNA was considered pure when the A260/A280 ratio was between 1.8 and 2. Values lower than 1.8 indicate protein contamination. Figure 2 shows that the DNA was not of good quality.  According to protocol 1, DNA was amplified using nested PCR and the primers employed were complementary to four conserved regions of the viral genome (pre-S/S, pre-core/core, polymerase, and region X), as described in Table 1 Polyacrylamide gel electrophoresis was performed to verify whether the fragment of interest was amplified by PCR.

Statistical analysis
For this study, OBI cases were considered in individuals in whom DNA amplification was obtained using the two protocols.
Categorical variables are presented as numbers and percentages. Continuous variables were expressed as mean ± standard deviation, as they presented a normal distribution according to the Shapiro-Wilk test. Pearson's chi-square or Fisher's exact test was used to analyze the differences between qualitative data when appropriate. The Student's t-test was used to compare quantitative data. The degree of agreement between tests was calculated using the kappa coefficient of agreement. Statistical significance was set at P value < 0.05.

Epidemiological and clinical data
Of the 104 patients investigated, the mean age was 47.8 (range, . The patients' demographic, clinical, and laboratory characteristics are shown in Table 3.

OBI diagnosed by nested-PCR
HBV DNA was amplified in 13 (12.5%) of the 104 patients evaluated using protocol 1 and in nine (8.7%) using protocol 2 ( No difference was found in the mean age (P = 0.244) or sex distribution (P = 0.698) between patients with and without OBI. No association was found between OBI and any underlying liver disease (P = 0.169). Table 5 summarizes the distribution of OBI cases according to the etiology of underlying liver disease.
No association was observed between the occurrence of OBI and presence of anti-HBc antibodies (P = 0.086). However, such an association was observed when HBV DNA cases were iden-

DISCUSSION
OBI was detected in 6.7% of 104 individuals with HBsAg-negative chronic liver disease, considering only those cases in which HBV DNA was detected by both protocols. The presence of HBV in individuals with chronic liver disease varies from 0.7-73% in different countries. 18,28,[30][31][32][33][34] In Brazil, the range is 2% 35 to 19.5%. 23 This variation is probably due to differences in the prevalence of HBV infection in different regions of Brazil and the world and in the methodology used for HBV detection.
In a previous study conducted at the same institution where the current study was developed, the authors found 4.4% of OBI in explanted livers from patients with HBsAg-negative cirrhotic who underwent liver transplantation. 26 In that study, the protocol of Raimondo et al. was employed, 5 and the investigators analyzed only fresh liver tissue removed from the explanted liver, which provides larger fragments for analysis, facilitating HBV DNA detection. 26 Conversely, in the present study, we used fragments obtained by percutaneous liver biopsy stored in paraffin blocks. However, contrary to expectations, considering the nature of the material, the frequency of OBI found in the current study was approximately three times higher using the same protocol.
It is noteworthy that in the study cited above, 26 sequencings were   Furthermore, in studies by Cacciola et al., 18 Sagnelli et al. 37 and Squadrito et al., 34 OBI was also associated with more severe stages of liver fibrosis or cirrhosis. We found no association between the occurrence of OBI and the etiology of the underlying liver disease. The association between OBI and chronic HCV infection has been observed in some investigations. 18,22,37,38 The small number of OBI cases in our study may explain the lack of this finding in the current study. A difference was found between the two protocols in HBV DNA detection, which reinforces the need for better standardization of the method to diagnose OBI. The protocol by Raimondo et al. 5 allowed the identification of HBV DNA in more cases when compared with the protocol by Chapel et al. 28 These authors described a nested-PCR protocol for HBV DNA detection in paraffin-embedded tissue using primers complementary to a conserved region of the S and Pol genes. 28  Of the phases of HBV infection, the least understood phase is OBI. 42 Several aspects need further investigation, such as the possible influence on the course of associated liver disease, the role of genetic polymorphisms in its development, and the diagnostic value of viral markers. In this context, it was observed that genetic variants of HLA-DP and the presence of anti-HBc may be important predictors of OBI. 43 On the other hand, Daef et al. 44 demonstrated that total anti-HBc is an ineffective marker of OBI. The association with HCV infection has been studied by different authors, but the results have been controversial.
In some studies, the absence of an interaction between OBI and chronic hepatitis C was observed, 45,46 while others have identified that some mutations in HBV may favor its occult phenotype in chronic HCV carriers. 47 OBI has been suggested to be associated with hepatocarcinogen-