| |
Editorial: Occult hepatitis B virus infection in patients with hepatocellular carcinoma: Innocent bystander, cofactor, or culprit?
|
| |
| |
Gastroenterology
January 2004, Volume 126, Number 1
Jorge A. Marreroa, Anna S.F. Lok* a
--The Key Question Is: Can Occult HBV Infection Alone Cause HCC?
--Another Critical Question Relates to the Basis for the Failure to Detect Circulating HBsAg in Patients With Occult HBV Infection
"...It is very likely that occult HBV infection is a cofactor in the development of HCC... Several studies found that patients co-infected with HBV and HCV have increased risks of HCC compared with those with mono-infection... studies on occult HBV infection in patients with HCC must provide details on other causes of chronic liver disease... The possibility that transient acute HBV infection many decades earlier may result in chronic low-grade liver injury and HCC is alarming and need to be confirmed in larger studies..."
Occult hepatitis B virus (HBV) infection is generally defined as the detection of HBV-DNA in the serum or liver tissue of patients who test negative for hepatitis B surface antigen (HBsAg). Occult HBV infection was first reported in the early 1980s when hybridization techniques for the detection of HBV DNA became available. These studies showed that HBV DNA could be detected in HBsAg negative patients with hepatocellular carcinoma (HCC). Recent studies using more sensitive techniques confirmed these observations. However, several questions remain unanswered. What is the prevalence of occult HBV infection in patients with HCC? Is occult HBV infection an innocent bystander, a cofactor or the sole culprit in the development of HCC? What is the basis for the failure to detect circulating HBsAg?
Before these questions can be addressed, it is important to establish standardized diagnostic criteria for occult HBV infection. An important observation in patients with occult HBV infection is the low HBV DNA levels: 102--3 copies/mL in the serum and 0.01--0.1 copy per liver cell. The frequency of detection of HBV DNA is higher in liver tissue than in serum, and more often in frozen than in paraffin-embedded liver tissue. Thus, the material used for HBV DNA detection and the sensitivity of the assay can influence the result. Because of the low amounts of HBV DNA, ultra-sensitive polymerase chain reaction (PCR) assays, with detection limits of less than 10 copies of HBV DNA per reaction are increasingly used. It is therefore important to confirm the specificity of the assays and to ensure that all necessary precautions to prevent cross-contamination have been taken. It has been proposed that diagnosis of occult HBV infection be made only when HBV DNA can be detected using at least 2 sets of primers from different areas of the HBV genome in duplicate assays. Appropriate negative controls must be included in each assay and specificity of the amplification reaction confirmed by sequencing of the amplicons. Most investigators describe precautions used during the PCR assay but there is very little mention of the process of sample collection. Cross-contamination can occur when blood samples are processed; this problem is of particular concern in retrospective studies on stored samples that have been used in the past.
Prevalence of occult HBV infection is related to the overall prevalence of HBV infection in that country. Occult HBV infection has been reported in 7%--19% of blood donors in endemic areas where 70% to 90% of the population had been exposed to HBV, and in 0%--9% in Western countries such as the United States where only 5% of the population had prior exposure to HBV.
Prevalence of occult HBV infection is also related to the presence of markers of HBV exposure. Thus, prevalence of occult HBV infection is higher in seropositive persons particularly those who are positive for hepatitis B core antibody (anti-HBc) only, than seronegative persons who are negative for anti-HBc as well as hepatitis B surface antibody (anti-HBs). Several recent studies reported a high prevalence of occult HBV infection among hepatitis C patients. This is likely related to common risk factors. Prevalence of occult HBV infection is also related to the population studied; rates of detection of serum HBV DNA increasing from 5%--50% among patients with chronic hepatitis to 20%--60% among patients with cirrhosis and HCC. The rates of detection of HBV DNA in liver tissue are higher, up to 60%--70% among patients with chronic hepatitis or HCC.
The study by Pollicino et al. in this issue of GASTROENTEROLOGY investigated the prevalence and molecular status of occult HBV infection among Italian patients with HCC. They studied frozen liver specimens from 107 HBsAg-negative cirrhotic patients with HCC and corresponding non-tumor tissue in 72 patients. The cause of the liver disease was attributed to hepatitis C virus (HCV) in 73 (68%), cryptogenic in 29 (27%), and alcohol in 5 (5%). Hepatitis B antibody results were available in 82 patients, 25 were anti-HBc positive (11 also had anti-HBs) and 57 were seronegative. As controls, frozen liver tissues from 192 patients with HBsAg-negative chronic hepatitis were studied, 153 (80%) patients were anti-HCV positive, 32 (17%) had cryptogenic liver disease, and 7 (4%) had alcoholic liver disease. Markers of previous HBV infection were available in 119 patients, of whom 33 (28%) were anti-HBc positive (18 were also positive for anti-HBs). Nested PCR was performed using primers in the HBV preS-S, precore-core, Pol and X regions.
The authors defined occult HBV infection as detection of HBV DNA in the liver tissue in at least two PCR assays using primers from different regions of the HBV genome. Using this strict criterion, occult HBV infection was found in 68 of 107 (64%) patients with HCC and in 63 of 192 (33%) controls with an odds ratio for the association between occult HBV infection and HCC of 3.6 (95% CI, 2.2--5.9). The association between occult HBV infection and HCC was stronger among anti-HCV negative patients, odds ratio of 9.6 (95% CI, 3.2--28.4) than in anti-HCV positive patients, odds ratio of 2.9 (95% CI, 1.6--5.1). The authors found a significant correlation between occult HBV infection and HCC regardless of age. As expected, occult HBV infection was significantly more common among the HCC patients who were anti-HBc positive, 84% vs. 56% of those who were anti-HBc negative. Direct sequencing confirmed that the amplicons were HBV DNA. These data provide convincing evidence that occult HBV infection is present in a high percent of patients with HCC. However, cross contamination during sample collection and processing cannot be ruled out.
PCR analysis using primers that can discriminate between relaxed circular and covalently closed circular (ccc) HBV DNA detected ccc HBV DNA in 20 of 30 cases studied. Furthermore, HBV RNA was detected in all 10 cases with good quality liver RNA. These findings together with the sequencing results showing no gross rearrangements confirmed that HBV DNA in patients with occult HBV infection is replication competent and transcriptionally active. Based on these results, the authors concluded that occult HBV infection can lead to HCC, and the carcinogenic mechanisms are similar to that of overt HBV infection.
The study by Pollicino et al. and other papers on this topic provided strong evidence of an association between occult HBV infection and HCC. However, the role of HBV in this setting is still unclear. Is it an innocent bystander, a cofactor or the sole culprit? Given the strong evidence of an etiological association between chronic HBV infection and HCC, it is unlikely that occult HBV infection is an innocent bystander in patients with HCC. The finding of a higher prevalence of occult HBV infection among patients with HCC compared with those with chronic hepatitis or cirrhosis support its role in the development of HCC. It is very likely that occult HBV infection is a cofactor in the development of HCC. Several studies found that patients co-infected with HBV and HCV have increased risks of HCC compared with those with mono-infection. The two viruses may interact by causing more severe inflammation and accelerated progression to cirrhosis. Alternatively, direct oncogenic effects of these 2 viruses may be additive or synergistic. Studies in woodchucks and humans have demonstrated potential pathways for direct oncogenic effects through cis- or trans-activation of integrated HBV DNA. Whether HCV has direct oncogenic effects in humans is unclear. Similarly, occult HBV infection can be a cofactor in the development of HCC among patients with alcoholic liver disease or cirrhosis due to other etiologies.
The Key Question is: can occult HBV infection alone cause HCC?
To address this question, studies on occult HBV infection in patients with HCC must provide details on other causes of chronic liver disease including nonalcoholic fatty liver disease, which may masquerade as cryptogenic cirrhosis, hemochromatosis, 1-antitrypsin deficiency, and autoimmune liver disease. In the study by Pollicino et al., more than 70% of the HCC patients studied had other causes of liver disease that may on their own lead to HCC. Moreover, some of the patients with cryptogenic cirrhosis may have nonalcoholic fatty liver disease. Therefore, occult HBV infection is the sole "identifiable" cause of HCC in only a small percentage of patients.
Another Critical Question- relates to the basis for the failureto detect circulating HbsAg in patients with occult HBV infection
In the current study, Pollicino et al. demonstrated the presence of ccc HBV DNA as well as HBV RNA indicating that patients with occult HBV infection have the potential to express HBsAg. Additionally, sequencing showed that there was minimal derangement and in no case were mutation(s) in the S gene responsible for the lack of serum HBsAg reactivity. These findings agree with earlier reports that occult HBV infection is in most instances related to low-level HBV infection with subdetectable levels of HBsAg and not infection with HBV variants that cannot express S proteins or variants that express S proteins with aberrant epitopes that cannot be detected in conventional serology assays.5 Thus, many patients with occult HBV infection could have had overt chronic HBV infection for many years with subsequent spontaneous clearance of HbsAg.21,22 Spontaneous clearance of HBsAg in patients with chronic HBV infection has been reported to occur at the rate of 0.5%--1% per year.23 Although the outcome of these patients is improved, HCC has been reported.24 Other patients might have transient acute HBV infection in the past with serological recovery. Numerous studies showed that low level HBV DNA can be detected in the liver of these patients more than 10 years later.25 A recent study from Japan found that inflammation and fibrosis can be detected up to 10 years after serological recovery from acute hepatitis B.26 These data if confirmed suggest that transient acute HBV infection can result in chronic liver injury and possibly HCC, underscoring the importance of HBV vaccination. Studies in woodchucks kept in environments free of carcinogens reported that 7%--22% of woodchucks that had recovered from transient acute woodchuck hepatitis virus infection developed HCC indicating that transient hepadnavirus infection can lead to HCC albeit at much lower rates than chronic hepadnavirus infection.27 In a landmark study by Beasley et al., the incidence of HCC per 100,000 per year was 495 and 5 among HBsAg carriers and anti-HBc and/or anti-HBs positive subjects, respectively, compared with 0 among seronegative persons.28 The finding of a higher prevalence of occult HBV infection among seropositive persons support the notion that occult HBV infection is frequently a late phase of overt chronic HBV infection or serologically recovered acute HBV infection. The explanation for occult HBV infection in seronegative patients is less clear. It is possible that some of these patients had transient acute HBV infection with clearance of HBsAg initially and decline of anti-HBc and anti-HBs titers to undetectable levels over time. It is also possible that some patients are infected with HBV variants that do not express any HBV proteins but retain their oncogenic effects. Further studies are needed to determine how frequently this occurs and to characterize these variants.
In summary, occult HBV infection is present in a substantial proportion of patients with HCC; the exact prevalence varies in different parts of the world and with the technique used for detection of HBV DNA. Future studies should focus on the role of occult HBV infection in the development of HCC: cofactor versus sole culprit, and on the basis for the failure to detect circulating HBsAg. The possibility that transient acute HBV infection many decades earlier may result in chronic low-grade liver injury and HCC is alarming and need to be confirmed in larger studies. Until these critical issues have been addressed, it is premature to advocate testing all HBsAg negative patients with HCC for occult HBV infection.
|
|
| |
| |
|
|
|
