It remains unclear whether the spontaneous clearance of hepatitis C virus (HCV) RNA from the serum of HCV-infected patients is associated with complete elimination of the HCV virus from hepatic tissue or low levels of viral replication can persist in some patients but are controlled by specific cellular and humoral immune responses. Past studies have reported the persistence of the specific HCV antibody for up to 2 decades after spontaneous clearance of HCV infection, and this suggests that there is continuous antibody stimulation by HCV antigens or mimicking HCV antigens.1, 2 Further support for low-level viral replication in these patients comes from reports by Carreno et al.3 and Dries et al.,4 who demonstrated the presence of HCV RNA within the hepatic tissue of healthy HCV RNA serum–negative patients who had normal aminotransferase levels. These findings all support the hypothesis that chronic low-level hepatic HCV viral replication can occur and is likely to be associated with continuous HCV antigen exposure and antibody response.
On the other hand, it has been shown that the level of HCV RNA is 104-fold higher within liver tissue in comparison with serum.5 Indeed, patients who were treated with interferon and ribavirin therapy and achieved a sustained virologic response had undetectable HCV RNA in the liver corresponding with a negative serum RNA level. In one report of 400 patients who had achieved a sustained virologic response to interferon therapy, 98% were shown to have undetectable liver levels of HCV RNA.6 The conclusions from this study were that hepatic HCV RNA levels were reflective of serum HCV RNA levels and that, in general, patients who clear HCV RNA from their serum have cleared HCV RNA from their livers.
In this issue of HEPATOLOGY, Hoare et al.7 report on a well-defined group of patients who were persistently HCV antibody–positive but who were serum HCV RNA–negative with normal alanine aminotransferase levels over a 5-year period. These patients had abnormal liver histology, which mainly consisted of portal and lobular infiltrates and a lymphocytic pattern (predominantly CD8+ cells) that was phenotypically similar to that of patients who have known persistent HCV RNA infection. The data suggest that there is a persistent immune response related to the presence of HCV RNA in liver tissue causing these histologic changes. The conclusion that the authors reached is that these inflammatory infiltrates are a marker for viral replication and persistent HCV infection. The authors based their conclusions on the following evidence: (1) 5.7% of the study group became RNA-positive during the follow-up; (2) only 7.5% of these patients had completely normal liver histology, whereas 82% had evidence of hepatic fibrosis; and (3) these nonviremic patients had inflammatory infiltrates in portal tracts that had a predominance of CD8-positive lymphocytes in a pattern similar to that described in patients with known HCV infection.
The question is whether these findings truly equate with the concept that there is persistent viral infection within the livers of these patients and, if so, what the clinical implications are. Unfortunately, the study did not determine hepatic levels of HCV RNA, did not use ultracentrifugation of serum samples (which increases the sensitivity of HCV RNA assays8), and did not use in situ hybridization to determine if infected hepatocytes were present in liver biopsy tissue.9 Furthermore, serum titers of HCV antibody were not assessed to determine if titers were increasing or diminishing; the latter has been reported in patients who have cleared HCV viral infections.10 Thus, the authors relied entirely on surrogate markers to come to their conclusion that persistent viral replication existed.
The authors note that, during a prolonged follow-up of these patients, 9 of 72 became HCV RNA–positive. In five cases, the results were positive on only one occasion, and it was suggested that these findings may represent false-positive reactions. However, four patients were persistently positive for HCV RNA, and this suggested that their disease may have reactivated, thus contributing to the evidence that some of these patients had ongoing HCV replication in the liver. It is important to point out that the study was conducted between 1992 and 2000, a time when there was much progress made in the development of assays much more sensitive in determining HCV RNA. It remains possible that HCV RNA positivity in these patients was missed by an insensitive assay and later picked up by a more sensitive assay.
The second point is that in this group of patients, 92% had the presence of inflammatory infiltrates and 82% had evidence of fibrosis on liver biopsy. Because the result of a single liver biopsy was reported in these patients, these findings raise the question of whether these histologic changes were stable, regressing, or progressing. It may be that previous exposure to HCV infection was associated with hepatic damage and fibrosis formation and that these histologic changes were indeed regressing at the time this biopsy was taken. It is well established that serum HCV RNA negativity following treatment with interferon in patients achieving a sustained virologic response is associated with decreasing inflammatory infiltrates and slow but steady regression of fibrosis, which may take a number of years to achieve.11 The dynamics of these histologic changes can be determined only by further follow-up liver biopsies in this group of patients. It is possible that what we are seeing in this study are the remnants of the previous HCV infection that occurred before the immune system cleared or controlled the virus.
Finally, the finding that the inflammatory infiltrates were phenotypically similar to those of patients known to have the presence of anti-HCV and viral replication seems to be more difficult to explain. At least one study has shown that HCV RNA levels correlate inversely with hepatic inflammation and fibrosis.6 It is well known that serum HCV RNA clearance is accompanied by up-regulation of HCV-specific cellular responses.12 Such immune responses are considered to provide protective immunity. Virus-specific CD4 T-cell proliferative responses and HCV-specific CD8 T-cells are detectable in most anti-HCV–positive individuals who have apparently recovered from an HCV infection.13 In fact, it has been demonstrated that infiltration of the liver with memory CD8 T-cells is required for protection from persistent HCV viral infection and that homing and expansion of specific CD8 T-cells in the liver may be important until a threshold is reached, which will then control viral infection within the liver. Furthermore, it has been shown that these cellular immune responses may be time-dependent, and it may take months or years to achieve true eradication of the virus from the livers of such patients. The question is whether these asymptomatic anti-HCV carriers can actually develop further histologic progression to fibrosis/cirrhosis or develop hepatocellular cancer as a result. Another question is whether these patients are able to transmit the virus to other individuals.
Is there other evidence that active viral replication may not be occurring in these patients? Evidence from the report of Markovic et al.15 demonstrated that patients who cleared hepatitis C infection and who were treated with immunosuppressive or cytolytic therapy for malignant lymphoma did not experience reactivation of hepatitis C. In contrast, for patients with hepatitis B, it is well known that hepatitis B reactivates after chemotherapy and/or immunosuppressive therapy.16 Furthermore, evidence comes from liver transplantation: it has been demonstrated by Everson et al.17 that pretransplant patients who clear HCV RNA or have a sustained virologic response rarely experience reactivation of HCV disease post-transplant despite being on high doses of immunosuppressive therapy.17 This is in contrast to patients who are HCV RNA–positive at the time of liver transplantation, in whom there is almost universal reinfection of the graft. One must keep in mind, however, that it may be possible that patients who spontaneously clear the virus may differ from those patients who clear the virus because of immune stimulation related to interferon therapy.
Are there other explanations for the ongoing inflammatory activity described in these patients? Although the authors have gone through great pains to eliminate other causes of liver disease in this group of patients, such as steatohepatitis and autoimmune hepatitis, another report from the liver transplant literature suggests that patients who have cleared HCV RNA from the serum following interferon therapy can indeed progress to cryptogenic cirrhosis and liver failure.18 Whether these histologic changes represent an autoimmune-like response or low-grade rival replication with persistent immune stimulation remains unknown. However, in patients who are on high doses of immunosuppressive therapy following liver transplantation, it seems unlikely that low-grade viral replication in the liver could be maintained. It is well known that patients with hepatitis C have a 100- to 1000-fold increase in HCV RNA serum levels following liver transplantation in comparison with pretransplant levels. As noted previously, it remains a possibility that the ongoing lymphocytic infiltrates represent memory CD8 T-cells that are controlling the HCV infection.13
In summary, the study by Horae et al.7 in patients who are anti-HCV–positive and HCV RNA serum–negative and have normal alanine aminotransferase levels suggests that some of these patients may have continued low-grade replicative HCV disease based on the presence of inflammatory infiltrates within the liver. Unfortunately, the authors did not use state-of-the-art technology to support their claims, such as determining hepatic RNA levels in liver tissue, using ultracentrifugation technology to increase the sensitivity of RNA detection in the serum, or employing situ hybridization to determine if hepatocytes were infected with the HCV virus. However, what is important in this group of patients is the long-term outcome. Will these patients progress histologically and will they have ongoing fibrosis and even possibly develop cirrhosis and hepatocellular cancer, or will these histologic findings regress with time and will the immune response eventually lead to complete eradication of the virus from the liver? Past experience in medicine indicates that there probably is a whole spectrum of responses. The authors already reported that 7.5% of the patients in this group had completely normal histology. In the past, the timeline with respect to eradication of HCV RNA from the serum following treatment with interferon has been studied extensively. A similar assessment and timeline of eradication of the virus from hepatic tissue have not been studied to date. The rarity of reactivation of HCV despite the use of immunosuppressive or cytolytic therapy, particularly in the transplant setting, suggests that viral replication is controlled in HCV antibody–positive, HCV RNA–negative individuals. In conclusion, long-term follow-up in this reported group of patients will be necessary to determine what the long-term impact of these clinical and histologic findings are. A repeat biopsy 3 to 5 years after the index biopsy would give important further information regarding the natural history in these patients.