THE GORDON WILSON LECTURE: THE HEPATITIS C VIRUS: FROM HIPPOCRATES TO CURE

Abstract

The modern age of viral hepatitis began in the early 1960s with the serendipitous discovery of the Australia antigen, a protein that was later shown to represent the envelope of the hepatitis B virus leading to its designation as the hepatitis B surface antigen. This was the first marker for any hepatitis virus and became not only a diagnostic assay, but also a mandatory blood donor screening test and the basis for the first generation hepatitis B vaccine. Prospective studies of transfusion recipients then showed that hepatitis B virus accounted for only 25% of cases of post-transfusion hepatitis (PTH). Discovery of the hepatitis A virus in 1975 revealed that none of the non-B PTH cases were due to hepatitis A virus infection resulting in the designation of this predominant form of PTH as non-A, non-B hepatitis. Using pedigreed patient samples and the chimp model, it was shown even before the advent of molecular biology that the non-A, non-B agent was small and lipid enveloped suggesting it might be a flavivirus. This was confirmed when the agent was cloned by Houghton and colleagues at the Chiron Corporation in 1987 and renamed the hepatitis C virus (HCV). In 1990 a donor screening assay for HCV was introduced nationwide and by 1997 PTH had virtually disappeared with rates now mathematically estimated to be one case per 2 million transfusions, compared to a 30% incidence before 1970. Clinical and molecular studies of HCV have shown that it results in persistent infection in 75% to 85% of cases due to its high replication rate, its existence as a quasispecies with a high mutation rate and the ability of HCV proteins to blunt and ultimately exhaust the HCV immune response. Although HCV infection is clinically and histologically mild in most patients, it can lead to progressive fibrosis over the course of decades in 30% to 40% and culminate in cirrhosis and end-stage liver disease. HCV can also cause hepatocellular carcinoma, generally on the background of cirrhosis with an incidence of 1% to 3% per year in this setting. After 2 decades of difficult and prolonged treatments with interferon-based regimens that resulted in cure rates of less than 50%, successive generations of HCV specific direct-acting antivirals were introduced that now result in cure rates of 95% to 100% across all genotypes after only 8 to 12 weeks of nontoxic oral therapy. This unprecedented efficacy has led to speculation that HCV infection might be eradicated even in the absence of a vaccine, but there are many impediments to global eradication including ascertainment of silent carriers, access to medication, and high drug cost. Nonetheless, we are in a new era of HCV control and optimism abounds.

INTRODUCTION

We have reached the point in the hepatitis C virus (HCV) epoch where greater than 95% of cases can be cured irrespective of the original mode of transmission, the viral genotype and, in most cases, the stage of hepatic fibrosis, including compensated cirrhosis. Recent therapeutic advances have been so remarkable that there is now rational speculation that HCV infection might be globally eradicated, even in the absence of a vaccine. This manuscript will deal with the back-story of this journey that encompasses only 50 years since discovery of the Australia antigen, the first bona fide marker for any hepatitis virus.

FROM HIPPOCRATES TO THE AUSTRALIA ANTIGEN

Hippocrates is credited with the first description of jaundice (“ikteros”) accompanied by the symptoms that we now associate with viral hepatitis. This description in approximately 400 BCE has proven accurate, although the causation has had many iterations. Hippocrates attributed ikteros to an excess of one of the four humors, namely “yellow bile.” Although Hippocrates' “sense of humors” may have been arcane, his observations, including hardening of the liver that he termed kirros, have withstood the test of time. Unfortunately, there were no major advances in understanding the causation of this form of jaundice over the next 2,000 years, but there were many observations of its occurrence and its epidemiology. No substantive insights into the etiology of hepatitis were advanced until the late 1800s when an outbreak of hepatitis among shipyard workers in Bremen, Germany, was traced to the receipt of smallpox vaccine that had been admixed with human serum (1), and then in the 1800s when several outbreaks of jaundice were linked to human plasma–contaminated vaccines for smallpox and serum-laced Salvarson (arsphenamine) for treating syphilis. Clearly, there appeared to be a transmissible, icterogenic agent residing in human plasma/serum.

Notably, every war throughout the centuries was marked by massive outbreaks of jaundice that we now know were predominantly cases of infectious hepatitis. However, many cases were traced to the percutaneous inoculation of products contaminated with human blood, including 42,000 cases of icteric hepatitis and probably 5 times that number of anicteric hepatitis cases occurring in World War II that were traced to contaminated lots of yellow fever vaccine and later shown to be caused by hepatitis B virus (HBV) (2). Extensive studies by the army during and after World War II characterized two forms of hepatitis, one that was primarily fecal-oral in its transmission mode and was called infectious hepatitis or hepatitis A, and one that was predominantly spread by inoculation of contaminated serum and hence termed serum hepatitis or hepatitis B. Epidemiologic studies by the army and others were extensive and precise, but the presumed agent of these cases remained elusive until the late 1960s when they were elucidated by several serendipitous events as described below.

FROM THE AUSTRALIA ANTIGEN TO THE HEPATITIS B VIRUS

In the early 1960s, a National Institutes of Health (NIH) laboratory headed by the late Baruch Blumberg was studying inherited differences in human proteins (polymorphisms) and their disease implications. The methodology used was simple; serum from a multiply transfused, and presumably multiply exposed, individual was placed in the center well of an agar gel plate and encircled by 6 wells of serum derived from diverse ethnic populations. A precipitin line developing between the center well and any surrounding well indicated an immune reaction and implied antibodies to a human protein that differed between the sources of the diffusing sera. Using this method, Blumberg had described polymorphisms among human beta-lipoproteins (3). Characteristic of precipitin lines that defined this polymorphism was their ability to take on a lipid stain and turn blue in the agar. One day, a precipitin line was observed that did not take up the lipid stain, but subsequently stained red when a protein counter-stain, azocarmine, was applied. This red line represented an immunopreciptin formed by antibody in the serum of a multiply transfused patient with hemophilia and the serum of an Australian aborigine, who by chance was among the population du jour for that day's experiments. There was no hint at that time, circa 1962, that this new antigen might be related to an infectious agent and certainly no clue that it might specifically relate to a hepatitis virus. Having been the first to observe this “red antigen,” my next task was to test patient populations at the NIH Clinical Center. These results were very informative, although not specific. The antigen, now named the Australia antigen (Au) for the first person in whom it was found, was detected in only 1 in 1,000 healthy US blood donors as opposed to 10% of an expanded population of Australian aborigines. Au was found in 1% of various NIH patient populations, but strikingly, in 10% of patients with leukemia. Hence, the first publication of this new finding stressed the association of Au with leukemia (4) and even postulated that the antigen might be a component of a leukemia virus or a protein polymorphism that induced susceptibility to leukemia. The link to leukemia later proved to be indirect, representing neither the causative agent or genetic susceptibility, but rather the fact that patients with leukemia were multiply transfused and hence exposed to hepatitis viruses from then untested donor blood and also immunosuppressed, predisposing to a chronic carrier state once infected.

At this point in the story, circa 1964, Blumberg left the NIH to take a position at the Fox Chase Cancer Center in Philadelphia, Pennsylvania, where he doggedly pursued the meaning of the Australia antigen. On the presumption that Au might impart a genetic predisposition to leukemia, he studied patients with a known inherited predisposition to leukemia, namely patients with Down's syndrome. He observed that Down's patients on average had a prevalence of Au similar to that of the Australian aborigines. Rather than assuming that his hypothesis was correct, Blumberg and the late Tom London performed a critically important study of Down's patients living in diverse environmental settings and found that Down's patients residing in large institutions had Au prevalences near 30% whereas those in small institutions had rates closer to 3%; and, importantly, those living at home or those tested at birth had a zero prevalence (5). These findings were inconsistent with a genetic hypothesis and provided the first clue that Au might be related to an infectious disease that was rampant in large institutions with poor hygienic practices. Which specific infectious disease remained a mystery until Barbara Werner, an associate in the Blumberg lab, came to work on a day when she was feeling weak, tired, and nauseated. Werner's serum had always served as the Au-negative control in the lab, and on this symptomatic day she tested her own blood and found herself to be Au-positive. This was a eureka moment because liver enzyme and bilirubin levels clearly indicated that she was developing a classic case of acute hepatitis and that her hepatitis was associated with Au seroconversion. A similar, but less dramatic, instance of Au seroconversion occurred in a Down's patient with sudden alanine transaminase (ALT) elevations and, on average Au-positive Down's patients had higher levels of ALT than their Au-negative counterparts. In subsequent studies, Prince showed that Au was specifically associated with hepatitis B (6); and then Dane in England (7), using immune electron-microscopy, observed the complete HBV virion (Dane particle) and distinguished it from abundant circular and tubular structures that proved to be antigenically identical to the surface coating of the encapsidated Dane particle; hence, Au was renamed the hepatitis B surface antigen (HBsAg).

FROM HBV TO NON-A, NON-B HEPATITIS

At this point in the HCV story, the scene shifted back to the NIH where prospective studies of post-transfusion hepatitis (PTH) were initiated by Robert Purcell, Paul Holland, and Paul Schmidt. In these studies, open-heart surgery patients were sampled every 1 to 2 weeks post-transfusion for 3 months and then monthly for an additional 3 months. Consecutive, weekly ALT elevations to greater than 2 times the upper limit of normal were considered evidence of PTH if other causes were reasonably excluded. Two important observations were made before 1970. First, that the incidence of PTH, when prospectively determined, was inordinately high at approximately 30%; and second, that the primary risk factor for PTH, in the absence of donor hepatitis testing, was the use of paid-donor blood. In a study led by John Walsh (8), it was shown that recipients of at least 1 unit of paid-donor blood had a 51% incidence of PTH compared to only 7% of those who received all-volunteer donor blood. Thus, in 1970, at the NIH Clinical Center blood bank, we adopted an all-volunteer donor program and simultaneously introduced donor screening for HBsAg. These combined measures led to a precipitous decrease in PTH from 30% to 9% and it could be retrospectively estimated that the primary reason for the declining rate was the exclusion of paid donors rather than testing for HBV carriers (9). In 1973, when more sensitive tests for HBsAg were developed and stored samples could be tested, it was shown that only 25% to 30% of PTH was caused by HBV infection. The large burden of non-B hepatitis remained undefined. In 1975, Feinstone, Kapikian, and Purcell at the NIH discovered the hepatitis A virus (HAV) using immune-electron microscopy (10). We immediately tested our non-B hepatitis cases using frozen stored samples and found that not a single one was due to HAV. It was then, as a presumably interim measure, that these etiologically undefined cases were designated non-A, non-B hepatitis (NANBH) (11). We resisted calling it hepatitis C, the next letter in the hepatitis alphabet, because we had not yet proven that it was transmitted by a virus, and if so, how many agents might be involved. The next step was to prove the existence of a blood transmissible agent and to establish an animal model. Stored sera from patients with post-transfusion NANBH and from blood donors implicated in such cases were inoculated into five chimpanzees and all five developed transaminase elevations at characteristic times post-inoculation, although none developed clinical disease (12). Further, this agent could be serially passaged in chimpanzees. Simultaneously, we collected apheresis units from a prospectively followed patient (patient H) during the upswing of his transaminase increase which peaked at 2,012 IU/L and was accompanied by elevations of serum bilirubin. Dr. Purcell's lab then titered the H inoculum in chimpanzees and showed that it had an infectivity titer of 10^6.5 chimp infectious doses/mL. Much later, this same sample was shown to have an HCV RNA titer of 10⁷ copies/mL such that the chimp infectious titer and the genomic titer were almost identical. Having a pedigreed infectious inoculum and the chimp model, it was then possible to perturb the inoculum to further characterize the non-A, non-B (NANB) agent. Steve Feinstone performed chloroform extraction studies of the H inoculum and showed in the chimpanzee model that chloroform treated NANB did not transmit hepatitis whereas sham-treated plasma remained infectious (13). This was an important experiment as it showed that the NANB agent contained a lipid envelope indicating both that the causative agent could be distinguished from non-enveloped agents such as the HAV and that it could be inactivated by lipid solvents, a method later successfully used to inactivate large volumes of plasma and plasma derivatives. Having established NANB as an enveloped agent, its size was then estimated by filtration studies in which H plasma was passed through filters of varying sizes and the filtrate then tested for infectivity in chimpanzees (14). This established its size as somewhere between 30 and 60 nanometers in diameter. Thus, before ever visualizing the agent or having a test to characterize its nucleic acid or gene products, we could deduce that the NANB agent was small and lipid enveloped. These characteristics narrowed the taxonomic possibilities to that of small RNA viruses in the family of alpha or flaviviruses or to a hepatitis B–like agent, but we had much additional evidence that NANB was not in the hepatitis B family. Hence it appeared that the NANB agent would either be a small RNA virus or a totally new class of viral agents. It was Dan Bradley at the Centers for Disease Control and Prevention (CDC) who first postulated that the NANB agent was most likely a flavivirus, as it later proved to be.

Having gone this far in characterizing the NANB agent, we were then stymied because we had not as yet identified a specific viral antigen or antibody, and because in the 1970s and early 1980s molecular biology was in its infancy and the blind detection of viral genomes was very difficult. Between 1975 and 1985, we and many other laboratories in the US, Europe, and Japan worked arduously to identify a viral marker that could be used to identify infected patients, and particularly to screen blood donors. Many candidate assays were proposed but none could break the code of an NIH panel consisting of randomly coded duplicate samples of pedigreed NANB cases and carefully selected negative controls. In the absence of a serologic or virologic assay and with the chimp model being too costly, too controversial, and too unavailable, the NANB field failed to advance except for critical clinical and histologic observations outlined below.

NANBH: EARLY CLINICAL OBSERVATIONS

While identification of the agent continued to be elusive, clinical observations began to document the serious consequences of chronic NANBH. First, it was observed that the vast majority of acute NANBH cases developed chronic anicteric hepatitis characterized by generally low-grade, but persistent or intermittent ALT elevations. The disease was most often so clinically mild that some believed this entity was simply a “transaminitis” of no clinical significance. That notion was dispelled by liver biopsy data and long-term follow-up. In the NIH series where we initially biopsied 39 patients with chronic NANBH, it was found that the majority had histologically mild disease, but that 10% had already developed cirrhosis by the time of biopsy and that re-biopsy in some over an approximate 5-year interval showed progressive fibrosis such that, in the final analysis, 8 of 39 patients (20%) had cirrhosis and 3 had died of end-stage liver disease (15). Thus, the potential severity of chronic NANBH was established and global studies over the decades have confirmed the approximate 20% incidence of cirrhosis in chronic NANBH/HCV, although longer follow-up now places that figure closer to 30%, and has also established an association with hepatocellular carcinoma (HCC) (see below).

FROM NON-A, NON-B TO HEPATITIS C

As evidence of clinical severity emerged in the late 1970s and early 1980s, virology and serology remained in a dormant state. During that period, Michael Houghton and his colleagues at the Chiron Corporation and Dan Bradley at CDC were quietly but tenaciously performing cloning experiments with serum/plasma derived from both humans and chimpanzees infected with NANBH. Their approach was to pellet plasma from highly pedigreed NANBH cases, to extract nucleic acid from the pellet, and to reverse transcribe extracted RNA. The derived cDNA was then cut with restriction enzymes and inserted into a phage GT-11 expression vector that was used to infect Escherichia coli in culture. The selection of this phage vector was critically important because any transfected viral gene product would be expressed by the bacteria into the surrounding medium. Chiron investigators then made another important assumption, namely that persons with chronic NANBH or patients who had spontaneously recovered from this infection would harbor antibodies to the NANB virus although such antibodies had never been identified. They thus lysed the E. coli to release any expressed antigens and overlaid the agar plate with presumed antibody containing sera. From 1981 to 1986 there was a trail of failures in this approach and, anecdotally, it has been said that there were 6 million negative experiments before identifying a single reactive clone. This reactive clone was then subcloned and the expressed antigen was used to develop an assay for detecting antibodies to the now renamed hepatitis C virus (16,17). In 1987, I was called by George Kuo of Chiron requesting the NIH coded NANB panel. I sent the panel with low expectations of success given 19 previous failures by other investigators. However, when the code was broken, it was found that Chiron had properly identified antibody in all chronically infected patients and implicated blood donors and failed to find antibody in any of the highly pedigreed negative controls. They did not find antibody in two patients with acute hepatitis, but later samples from these patients showed their seroconversion to anti-HCV–positive. Overall, the Chiron assay showed perfect correlation between random duplicate samples, identified all chronic carriers, and had no false-positives. With this apparent breakthrough, I had Chiron test sera from 15 of our prospectively followed patients with well-characterized NANBH. All 15 were antibody-negative in their pre-transfusion sample and anti-HCV–positive in their post-transfusion sample in temporal relationship to their biochemical evidence of PTH (18). Further, we found a linked anti-HCV–positive donor in 80% of NANBH cases using a first-generation assay and 88% with a second-generation assay. Hence, it could be predicted that introduction of this assay to blood donor screening might prevent near 90% of residual PTH, as proved to be the case. Chiron investigators using the cloned viral fragment were then able to “walk the genome” and ultimately characterize the full-length genome. Genomic structure substantiated that this small enveloped virus was in the family flaviviridae and distinct from other known members of this family such as the Yellow Fever virus or West Nile virus. Having established serologic assays, molecular assays, and genomic structure, the NANB agent was officially transmuted into the hepatitis C virus (HCV) and the awkward name retired to historical archives.

HCV: PERSISTENCE AND LONG-TERM CONSEQUENCES

With the availability of specific serologic and molecular assays, with delineation of structural and non-structural viral encoded proteins, and with sera from multiple human HCV-infected cohorts, critical biologic, molecular, and clinical data rapidly proliferated from laboratories across the globe. One of the most important findings was that HCV generally persisted in the host leading to chronic infection in 75% to 85% of cases. The primary reasons for this persistence have now been elucidated. First, the virus replicates very rapidly having a half-life of 17 hours and reaching peak titer within days to no more than 2 weeks after initial infection. This early high viral load and the inhibitory effects of several viral proteins can suppress the innate immune response and overwhelm the adaptive response to a state of T-cell exhaustion. Chronically infected patients often have HCV-specific T cell activity no different from persons never exposed (19). A second critical factor is that HCV, similar to HIV, exists as a family of closely related, but antigenically distinct variants that has been termed the viral quasispecies. This is in large part related to the poor proofreading ability of the HCV polymerase that allows for enormous variation of a rapidly replicating virus. A vast number of variants are present simultaneously such that even if a neutralizing antibody response was mounted to the dominant strain, any of many co-existent strains could escape the antibody response and emerge as a new dominant strain. Hence, the host is continually trying to catch up to pre-existing and emerging mutations and almost invariably fails to do so.

Persistent HCV infection leads to chronic hepatitis which, although generally mild for decades, can ultimately evolve to progressive fibrosis, cirrhosis, and end-stage liver disease in at least 30% of those infected. Conversely, the majority of patients may have an indolent course for 4 to 5 decades and possibly never develop serious HCV consequences over their natural life-span. Unfortunately, there are no established early markers that distinguish slow from rapid progressors, although a recent study has shown that rapid progressors have higher levels of MCP-2, a profibrogenic cytokine, and lower levels of interferon and other protective pro-inflammatory cytokines (20).

Several studies of young women infected with contaminated lots of Rh immune globulin (21) and of healthy blood donors (22) have shown that less than 15% develop severe fibrosis in the first 25 years of infection. However, there are many confounders that accelerate fibrosis progression. Among these are AIDS and other immunodeficiency states (23). High alcohol intake (40 g/day in men and 20 g/day in women) more than doubles the risk of cirrhosis with the proportion increasing with increasing duration of alcohol excess (24). Perhaps the greatest confounder at the present time is coexistent obesity and concomitant fatty liver (nonalcoholic fatty liver diseases) and particularly, nonalcoholic steatohepatitis (NASH) that encompasses obesity, insulin resistance, type 2 diabetes, and hypertension (the metabolic syndrome) (25). NASH, in the absence of HCV, can cause hepatic inflammation, fibrosis, and cirrhosis and in the HCV-infected patient can lead to more rapid fibrosis progression. It is predicted that with the cure of most HCV-infected individuals (see below) and subsequent declining rates of virus-induced cirrhosis, NASH will become the leading cause of cirrhosis in countries where the obesity epidemic is accelerating.

There are two major risks for patients who have progressed to cirrhosis: 1) hepatic decompensation and mortality due to end-stage liver disease, and; 2) HCC. HCC is rare in HCV-infected patients before the development of cirrhosis, but once cirrhosis is established, the risk of developing HCC is 1% to 3% per year (26). Further, although that risk is diminished once HCV infection is eradicated with medication, the risk does not decrease to zero and cirrhotic patients must be followed semi-annually with liver ultrasound and determinations of alpha-fetoprotein both before and after HCV cure. In the first decade of the 21^st century the incidence of HCC increased 3-fold in the US, and that increase was all related to chronic HCV infection (27). It was estimated that HCC incidence would continue to increase by 4-fold by 2030 unless there was a means to curb the HCV endemic. Fortunately, that means has come to fruition in the form of oral, direct acting antivirals (DAAs) that have greater than 90% efficacy as described below. At present, HCV is the leading cause of HCC in the US and Europe and together with HBV is the leading cause throughout the world.

THE NEAR ERADICATION OF TRANSFUSION-ASSOCIATED HEPATITIS

As described above, dramatic reductions in PTH were observed after adoption of an all-volunteer donor system and introduction of first-generation assays for HBsAg (9). Almost all residual PTH was due to the agent of NANBH later renamed the HCV. By the early 1980s, prospective NIH studies showed that PTH incidence hovered around 6%. Various interventions in the early 1980s such as donor ALT testing and the screening for HIV had no measurable effect on PTH incidence. In 1985 and 1986, two studies suggested that testing donors for antibody to the hepatitis B core antigen might serve as a surrogate marker for carriers of NANB and reduce the incidence of PTH by 30% to 40% (28,29). This donor screening assay was mandated in the US in 1977 and NIH prospective studies showed a decline in PTH incidence to approximately 4% by 1989. In 1990, the first-generation assay for anti-HCV was introduced and we measured a decline in transfusion hepatitis incidence to 1.1%. A more sensitive anti-HCV assay was introduced in 1992 and by 1997 hepatitis incidence in the NIH study had decreased to virtual zero. This does not mean that transmission cannot occur, but rather that the rate is so low that it cannot be measured in the relatively small prospective study being performed. It has been mathematically estimated that current PTH risk resides between one case in every 1.5 to 2.0 million transfusions, a remarkable decline from the 30% incidence observed in 1970. Based on incidence figures from the NIH cohort, it can be calculated that almost 5 million cases of PTH might have been transmitted in the US between 1970 and 1990. Up to 20% of such cases might eventually have developed cirrhosis if they did not succumb in the interim to the underlying disease for which they were initially transfused or to another intercurrent illness. Conversely, the introduction of anti-HCV testing in 1990 may have prevented an additional 2.4 million cases in the decades from 1990 to 2010.

A NEW TREATMENT PARADIGM

There has been a dramatic progression in treatment efficacy for chronic HCV infection. Early treatments with interferon alone had only a 6% sustained efficacy if administered for 6 months and 16% if administered for a year. This was accompanied by major side effects and diminished quality of life. In 1998, ribavirin was added to interferon and sustained efficacy increased to 42%, but ribavirin added additional complications, particularly anemia. In 2001, a pegylated formulation of interferon was developed that provided slower degradation and thus higher serum levels of interferon that allowed for weekly dosing and together with ribavirin increased efficacy to 55%. Based on sequencing and 3-dimensional structure determinations, oral DAAs were developed that targeted key replicative and enzymatic sites in the HCV virion. The first of these was directed at the viral NS3-NS4 protease and when this protease inhibitor was added to pegylated interferon and ribavirin, the sustained virologic response (SVR) rate increased to 70%. However, the addition of the protease inhibitor to the standard regimen added further side effects and complexities to the treatment. The major breakthrough came with the development of DAAs targeted to the viral polymerase (NS5B) and the replication complex (NS5A). These oral DAAs, combined into a single pill, negated the need for interferon and ribavirin and greatly simplified treatment, as well as minimizing side effects. Efficacy with oral DAAs increased to more than 90% SVR and recent iterations can achieve 98% to 100% efficacy across all HCV genotypes and greater than 90% efficacy even in patients with compensated cirrhosis (30,31). Such profound levels of SVRs, which are tantamount to cure, were inconceivable only a decade ago. It is a tribute to molecular biology, crystallography, advanced drug-screening technology, and investigator persistence. It is also a tribute to collaboration between industry, academia, and the government toward a common goal. At the present time, there is virtually no one with compensated liver disease who cannot be cured of their HCV infection.

THE “END” OF THE STORY

The NANB-HCV saga has reached the “end of the beginning” and raised the possibility that this could be the “beginning of the end.” The concept of eradicating this virus is no longer fantasy, but the path forward will be arduous. It is no longer a question of whether we can cure the infection, but whether we can identify all the silent carriers of HCV and then triage them to treatment. Of the estimated 3.5 million HCV carriers in the US, only half have been identified. Targeting this large number is daunting. Universal HCV testing would be very costly and logistically impractical. Thus, the CDC has recommended targeting the highest-risk populations, particularly drug addicts attending various treatment and needle-distribution centers and “baby boomers” (persons born between 1945 and 1965 who were in young adulthood when the recreational drug epidemic spread in the US during the 1970s and 1980s). Drug use with needle sharing is the single most important transmission route for HCV, especially since 1990 when transfusion was no longer a significant route of spread. It was thought that identifying high-risk individuals simply by age rather than by asking sensitive questions relating to drug use would simplify the process for physicians, but this approach has been only partially effective. Further, it depends on patients seeking health care which misses a large segment of those at risk. Hence, identification of silent carriers is the first impediment to eradication and new large-scale screening approaches are sorely needed. The second major impediment is access to treatment. DAA therapy, although cost-effective in the long run is very costly in the short term. This cost is beyond the means of most individuals and thus has to be borne by private or government insurance. Both of these insurers have been slow to take up this large financial burden, although the situation has improved recently as drug competition has emerged; as prices have declined (although still very high); as the long-term benefits of preventing cirrhosis, end-stage liver disease, and liver transplantation are being better recognized; and as the moral imperative to treat a potentially deadly, but highly curable disease has gained traction. The financial and logistical impediments to eradication in the US and Europe are magnified in the developing world where the health infrastructure is marginalized by poverty, where transmission routes are magnified by unsterile or ritualistic medical practices, and where the cost of treating millions, even at reduced drug prices, could consume or exceed national health budgets. Hence, whereas broad-scale screening and subsequent treatment might markedly reduce, if not eradicate, HCV infection in industrialized nations, the best global solution is to develop an HCV vaccine. Thus far, this has proven very difficult because of the quasispecies nature of the virus, the frequency of escape mutations, and the virus' ability to inhibit the HCV-specific immune response through a variety of mechanisms. Most attempts at developing an HCV vaccine have failed, but Houghton and colleagues have developed a vaccine based on HCV envelope proteins that in the chimpanzee appears to induce antibodies that, although they do not prevent infection (sterilizing immunity), lessen the propensity to develop chronic infection. It is unclear whether this vaccine will reach the marketplace or provide sufficient benefit to warrant universal implementation.

Despite many future challenges, steps toward the eradication of HCV infection have come a long way and advanced to a stage previously unimaginable. That HCV eradication is even in the discussion shows how far we have come in only 50 years since discovery of the Australia antigen, the beginning of modern-day hepatitis virus study, and only 20 years since the cloning of HCV. Hippocrates would be very pleased with our progress.