About 10% to 45% of persons who develop acute hepatitis C recover spontaneously, signaled by improved symptoms, normalized liver-related chemistries, loss of hepatitis C virus (HCV) RNA from serum, and the development of hepatitis C antibody.1 If spontaneous recovery should happen, it is always within 6 months of the acute infection and almost never beyond that time, when the disease is now regarded as chronic hepatitis C. At this point, viral loss occurs only if treatment is successful. Currently being debated is whether spontaneous and treatment-induced conversion from detectable to nondetectable serum HCV RNA establishes cure of the infection and the resulting liver disease.2
The initial hepatitis C registration treatment trials based therapeutic success on the development of a sustained virologic response (SVR), defined as absent HCV RNA in serum using a sensitive test at the end of treatment and again 6 months later.3 The hope was that this arbitrary endpoint would indicate that hepatitis C was now “cured,” but would require proof from long-term follow-up to confirm permanent resolution of virologic, clinical, biochemical, and histological footprints of chronic hepatitis C.
Accordingly, efforts focused first on establishing whether an SVR signaled durable loss of HCV RNA. More than 40 long-term follow-up studies have now been reported aimed at determining whether viral loss is maintained in patients who had developed treatment-induced SVR.2 Evaluations after reaching an SVR were performed at intervals of 1 to more than 10 years later. In an extensive review of these studies, HCV RNA was noted to have remained undetectable in 97% of a combined total of 4,228 patients from 44 studies during their differing follow-up periods.2 Accompanying the SVR in most instances are improved clinical symptoms and quality of life; a reduction in or normalization of the earlier abnormal liver-related chemistries; improvement in liver histology reflected in decreased hepatocellular inflammation and reversal of fibrosis and even of cirrhosis; and a marked reduction in liver-related morbidity, hepatocellular carcinoma (HCC), and mortality.4-8 It is therefore hardly surprising, and indeed not inappropriate, that achieving an SVR has come to be referred to as a “cure,” or, by those who are more cautious, as a “virologic cure.”
However, there is concern that an SVR does not always establish cure because there are a growing number of reports describing HCV RNA reappearance among individuals who had developed a treatment-induced SVR.9-11 A challenging issue is whether this recurrence represents spontaneous relapse of the original infection9-11 or relapse precipitated during a later immunosuppressive event,12, 13 or whether it is an entirely new HCV infection, as has been reported among persons who continue involvement in high-risk behaviors.14, 15 Adding to the uneasiness are reports of HCC developing months to years after having reached an SVR despite the continued absence of detectable virus, many involving individuals whose liver biopsies when the SVR occurred had shown bridging fibrosis or cirrhosis.16, 17 Absent another etiology for the cancer, such as occult hepatitis B virus infection,18 the inference is that the cancer must link to the preceding HCV infection, even though HCV RNA remains undetectable in the serum. If so, where is the virus actually harbored?
The first place to search for the virus was obviously the liver itself and, indeed, numerous publications have reported finding HCV in liver even when undetectable in serum.9, 19, 20 Moreover, in light of its known lymphotropism,21 the virus has been identified also in immune-related cells such as monocytes/macrophages and B cells,22, 23 dendritic cells,24 and, especially, peripheral blood mononuclear cells (PBMCs).9-11, 25 Evidence that the liver and PBMCs are viral reservoirs when HCV RNA cannot be measured in plasma is now referred to as “occult hepatitis C virus infection.”2, 26 Are these sites then the source for the occasional reports of reactivation of HCV infection or induction of HCC27 despite having developed an SVR? Would persons with occult hepatitis C be infectious to others if they donated blood or organs? An answer to these questions must come from showing that occult hepatitis C does represent actively replicating hepatitis C virus. This requires highly sensitive tests capable of identifying both positive- and negative-strand RNA because hepatitis C viral replication within cells is maintained by the production of replicative intermediate molecules. Indeed, positive- and negative-strand RNA have been detected in PBMCs by several9-11, 28, 29 but not all investigators.30 The inconsistency in identifying replicating virus in extrahepatic sites is conceivably the result of varying sensitivity of the assays utilized that, although they have improved, remain difficult to perform. This uncertainty is presumably what impelled the study by Fujiwara et al.31 reported in this issue of HEPATOLOGY.
These investigators studied 126 patients who had developed acute transfusion-associated hepatitis C, 67 having advanced to chronic hepatitis C, and 59 patients from the U.S. and Japan who had recovered from the acute infection, 11 spontaneously and 48 following treatment. They sought HCV RNA in PBMCs from 48 carriers and 16 patients who had recovered: three spontaneously and 13 after treatment. Following meticulous preparation of the PBMCs, including separation of B and T cell subsets, they used a highly sensitive nested RTD (real-time detection) polymerase chain reaction (PCR) to detect HCV RNA, whereas negative-strand HCV RNA was determined using a highly sensitive rTh-based method. HCV RNA was identified in PBMCs of virtually all chronic carriers, the viral load being highest in the B cells, but could not be detected in PBMCs of those who had recovered spontaneously or following treatment. Moreover, HCV RNA was not detected in supernatants of PBMC cultures, measured at intervals, from persons who had recovered, but the virus was identified in most chronic carriers. Similarly, HCV RNA was not detected in the liver and other tissues of two chimpanzees that had developed acute hepatitis C after inoculation with HCV-positive plasma, but had recovered. Among carriers, a moderate correlation was found between the HCV viral load in serum and the PBMCs, being significantly higher in the B-cell subset than in the total PBMCs, T cells, and non-B, non-T cell fractions. Regarding negative-strand RNA, none was detected in any of the PBMCs and their subsets. Finally, they showed that when PBMCs from healthy blood donors were incubated with plasma from chronic HCV carriers, the PBMCs and subsets from healthy donors became HCV RNA-positive, the concentration again being highest in the B-cell subset. They interpreted this as indicating that HCV RNA can passively adsorb onto normal PBMCs, especially the B subsets, findings that could be mistaken for actual infection of PBMCs. Their overall conclusion is that an SVR, whether occurring spontaneously or following treatment, signals full recovery from HCV infection. Furthermore, they indicated that because HCV RNA is cleared from both plasma and other tissues, PBMCs are unlikely to be the source for recurrence of replicating virus.
How does one reconcile these findings with those of other investigators who have reported identifying replicating HCV in PBMCs? After all, theirs was a meticulously performed study with compelling results, supporting the notion that an SVR implies eradication of HCV and thus presumably cure of the infection. They confirmed that HCV RNA can be detected in PBMCs, but as nonreplicating virus that is probably not harmful to the host or to others. Are the discrepancies due to differences in sensitivities of the assays used by the various investigators studying this issue? Might it be because of differences in the population studied or in the timing of follow-up after having reached the SVR? Unfortunately, the basis for these conflicting results remains unclear, even to these investigators who acknowledged the discrepancies, speculating that it might have been the result of the small size of the population they studied. Clearly, support for their conclusions warrants confirmation by others.
So what can be concluded presently regarding the significance of an SVR? Current data suggest that achieving an SVR almost always signals durable loss of virus and improvement of the associated liver disease, and hence indicates apparent cure. But this may not be universal for as yet unknown reasons. Conceivably, occult HCV infection may remain just that until stressed by an immunosuppressive event. What seems important, in addition to seeking reasons for the conflicting data, is to define characteristics of persons likely to relapse or develop HCC, who would then warrant frequent virologic and biochemical screening after reaching an SVR in order to begin appropriate management early. For the rest, it seems appropriate to perform virologic and biochemical screening annually, as suggested by others,2 particularly if, at the time of reaching an SVR, there was histologic evidence of advanced fibrosis or cirrhosis.