Introduction
Source: Richard B. Levine, Newscom.com
Waiting for Godot. Millions of people infected with hepatitis C await better therapies.
2007 was a good year for hepatitis C—for the virus, that is. Companies, including seven biotechs, abandoned or suspended at least eight antiviral drugs in clinical development over the course of the year. "There have been quite a few compounds that have bitten the dust lately," notes virologist Charles Rice of Rockefeller University in New York. The disappointments are nothing new: drugs specifically targeting the hepatitis C virus (HCV) have been in the works since the early 1990s, but no such compound has yet advanced to phase 3 trials.
But 2008 and beyond promise better. A compound from Cambridge, Massachusetts–based Vertex Pharmaceuticals should begin phase 3 trials by early in the year, and both biotech and pharma continue to pour resources into new therapies (Table 1). "People have been saying for the last 15 years that there's a new specific antiviral therapy that'll be approved in 5 years," says Rice. "That's turned out to be not very realistic. But I think we're finally in a place where that's likely to happen."
The two percent solution
New therapies are clearly needed. HCV has infected an estimated 170 million people worldwide, most of them chronically. New infection from blood transfusion has almost disappeared since 1990 thanks to screening, but the number of people facing death or serious liver disease from HCV is steadily rising because people often live decades with the virus before showing symptoms and because of new infections in illegal drug users.
Standard therapy for HCV, a 48-week course of interferon-
and ribavirin, can cure about half of patients, but it's a very hard treatment. Along with extreme fatigue and flu-like symptoms, patients often suffer persistent headaches, muscle aches and anemia. "Some of our patients who've gone through cancer chemotherapy say that cancer chemotherapy is easier than hepatitis C treatment," says Anna Lok, a hepatologist at the University of Michigan in Ann Arbor. Patients also can suffer devastating emotional side effects. Interferon treatment can lead to depression and, in rare instances, suicide. Many patients discontinue therapy, and others opt to simply wait out their infections rather than submit to a year of treatment. Because of a combination of nondiagnosis, treatment failure and treatment avoidance, and discontinuation, "only approximately 2% of the people with HCV in America are successfully treated with standard of care," noted Michel de Rosen, CEO of ViroPharma, in Exton, Pennsylvania, at a recent investor conference. "This means 98% of the people need something else, something better."
To kill a virus
Compared with HIV, for which targeted therapies have been on the market since the mid-90s, HCV has lagged far behind. Although HCV, first isolated by scientists at Chiron, in Emeryville, California, in 1989, immediately presented several tempting enzyme targets, progress on new therapies was long hampered by the inability to propagate the virus in cell culture. "You do an enzyme screen [for compounds], but then you've got to get into some cell-based assay, and we didn't have that for the first ten years," says Lawrence Blatt, chief science officer at InterMune in Brisbane, California. The only animal model for HCV, the chimpanzee, is too expensive for routine use.
The 1999 invention of a replicon system that enabled viral RNA replication in cell culture (but not new virus formation) led directly to the current wave of targeted therapies. "Now we're seeing an explosion of directed antivirals against HCV," says Blatt. In 2005, culture systems finally arrived that permit assembly and release of new virus. "Hopefully we are... past the growing pains of the field, which were agonizingly slow," says Rice.
Among the specific antivirals, inhibitors of the nonstructural protein 3 (NS3) protease are furthest along. As in HIV, the protease is an obvious target. HCV is a single-strand, positive-sense RNA virus whose tiny genome encodes a single polyprotein. This is cleaved by proteases, including NS3, to produce ten viral proteins. NS3 activity is absolutely necessary for viral replication. With its very shallow substrate-binding pocket, NS3 presented a much harder target than the HIV protease, but several companies have discovered potent and specific inhibitors that bind the enzyme's active site.
In 2003, an inhibitor from German firm Boehringer-Ingleheim produced a stunning 100–1,000-fold reduction in viral load in patients after just two days of treatment1. Although the drug was quickly withdrawn after cardiac toxicity appeared in animals, that trial led many companies to restart their stalled protease inhibitor programs. At least four protease inhibitors are now in clinical trials, with many others in preclinical development. It's almost certain that a protease inhibitor will be the first targeted antiviral on the market.
Vertex's NS3 inhibitor, telaprevir, is generating the most interest, not just because of its excellent potency (a median 3.4 to 4.8 log drop in viral RNA after a few days of treatment in phase 1 trials) but because the company is using the drug to cut current treatment time in half—a controversial approach. The drug, like all targeted therapies to date, is under development as combination therapy with interferon and ribavirin. Radically shortening therapy could be the key to eventual approval by the US Food and Drug Administration (Rockville, Maryland), as long as benefits from the triple combination outweigh toxicity. Phase 2 results, reported in November at the American Association for the Study of Liver Diseases meeting in Boston, showed that 12 weeks of triple therapy followed by 12 weeks of interferon and ribavirin alone produced a sustained virological response (SVR) rate of 61% and 65% in two trials, both for patients with genotype 1 HCV. Although control-arm results are not yet available, standard therapy typically leads to SVRs in 40–50% of genotype 1 patients.
SVR, defined as undetectable RNA six months after the end of therapy, is a very reliable indication of cure. That Vertex was able to achieve a higher SVR rate than interferon and ribavirin alone in half the time is good news for the compound and has broader scientific implications.
Those implications involve how the virus is destroyed. A seemingly bedrock truth about HCV is that it's not enough to stop RNA replication (the goal of targeted therapies like telaprevir); one must also eradicate the virus from the body by harnessing the immune system to clear infected cells. (Unlike with HIV, merely suppressing HCV viral load has not been shown to improve clinical outcomes.) No one knows why roughly half of patients don't respond to interferon and ribavirin, but it's assumed they have intrinsic immune system defects that prevent them from clearing infected cells. Such defects, by definition, can't be overcome using targeted antivirals like telaprevir because all they do is stop RNA replication, and thus the virus remains, ready to rebound.
The 24-week telaprevir treatment is controversial—and potentially paradigm shifting—because it challenges conventional wisdom, which holds that the adaptive immune system needs the full year to clear infected hepatocytes. Telaprevir's superior SVR rates in phase 2 compared with interferon and ribavirin alone suggests that the immune system plays a much smaller role than previously thought and that just stopping RNA replication may be enough to cure patients.
Targeted therapies like telaprevir "are all about driving down viral replication and targeting the virus, but not touching the immune system," says Vertex executive VP and chief medical officer John Alam. Viral RNA, its replication halted, may simply degrade within liver cells, Alam speculates. If the patient's adaptive immune system is not dictating their ultimate response, "maybe, eventually, there is no limit to who we can actually cure" with targeted therapy, he says.
Looming obstacles
Theoretical implications aside, Vertex must still deliver results in phase 3 trials. Toxicity is one concern. Treatment discontinuation rates were significantly higher for the telaprevir arms compared to control arms in phase 2. If side effects push up discontinuation rates in phase 3, that could affect Vertex's ability to prove clear superiority to current treatment.
There's an even bigger uncertainty. Vertex's two completed phase 2 trials were for treatment-naive patients only. It's an open question whether triple therapy with telaprevir or anything else can help the large body of patients who have already failed standard therapy—the very people who need a new therapy most. (Vertex is now completing such a phase 2 trial.) Offering interferon and ribavirin again probably does not help these patients, says Lok: "There's no point in giving someone a treatment if they've already had it."
Proving her wrong may depend on a second function of the NS3 protease, discovered in 2003: it blocks the double-stranded RNA-sensing pathway of interferon regulatory factor-3 (IRF-3), at least in vitro2. By giving telaprevir, speculates Alam, "you're de-repressing those pathways" and enabling interferon to work again against the virus. "This may work in partial responders to interferon and ribavirin",11 agrees Rice, "but I'm not sure it makes sense for the null responders."
Interferon or not, some kind of combination therapy will be necessary. Telaprevir and other specific antivirals won't work as monotherapy because of viral resistance. HCV is very prolific, producing about a trillion particles per day in a chronically infected individual. And, like all RNA viruses, it has an error-prone replication strategy that produces a swarm of quasi-species variants. As a result, drug-resistant variants arise very early in therapy, within seven to ten days. "Viral replication and the viral load is so much higher than in HIV, [so] resistance comes up much faster," says Norbert Bischofberger, chief scientific officer at Gilead Sciences, in Foster City, California.
Thus the need for combination therapy. Results so far for telaprevir/interferon/ribavirin are encouraging, with only a 5% rate of viral breakthrough in phase 2. Short-term studies at Vertex show that variants that are highly resistant to telaprevir tend to be less fit for replication and hopefully can be wiped out by interferon and ribavirin. The phase 3 telaprevir trial should give definitive answers.
Although specific amino acid substitutions conferring resistance to both protease inhibitors and polymerase inhibitors have now been mapped, "whether or not [resistance] will emerge as an important determinant of response, the way it has in HIV, we need to wait for data to find out," Intermune's Blatt says. Drug resistance did not cause the recent wave of clinical failures.
Deconstructing failure
Four of the failed drugs were inhibitors of NS5B, the RNA polymerase. In July, Idenix Pharmaceuticals, based in Cambridge, Massachusetts, placed its phase 2 polymerase inhibitor, valopicitabine, on clinical hold "based on the overall risk/benefit profile," according to a press release, and the compound has since been dropped. Both poor potency and severe gastrointestinal side effects (nausea and vomiting) reportedly sank the drug. In August, ViroPharma and its development partner, Wyeth, announced discontinuation of phase 2 dosing of their polymerase inhibitor, HCV-796, because of elevated liver enzymes in some patients. HCV-796 had shown excellent potency and efficacy, making its problems particularly painful. GlaxoSmithKline, in London, and XTL Biopharmaceuticals, of Valley Cottage, New York, discontinued their polymerase inhibitors in phase 1 earlier this year.
These failures are a serious setback for the field, but they don't indicate a class effect, say experts. "Each drug is its own beast and has its own characteristics, and unfortunately we've had some bad luck," says John McHutchison, a liver specialist at Duke University Medical Center, in Durham, North Carolina. "I'm hoping with the next wave we won't have those... problems."
At least three polymerase inhibitors remain in clinical development (Table 1) with several others in the pipeline. Early results for the newer polymerase inhibitors look promising. In September, Pharmasset, in Princeton, New Jersey, reported that its nucleoside polymerase inhibitor, R7128, produced a mean 2.7 log decrease in viral RNA at the highest phase 1 dose without any serious adverse effects. According to Pharmasset's VP for pharmaceutical research, Michael Otto, R7128 doesn't have significant activity against other flaviviruses, unlike previous nucleoside inhibitors, suggesting that it's very specific for HCV polymerase and thus runs less theoretical risk of side effects.
The company has now launched a 28-day combination trial with interferon and ribavirin. Nucleosides, which prevent HCV replication by latching onto—and preventing further growth of—the replicating viral RNA chain, often lead to bone marrow or mitochondrial toxicity, but Pharmasset has seen no evidence of that to date.
Nucleoside inhibitors are less prone to drug resistance than non-nucleoside inhibitors because they bind to the enzyme's active site, which is more conserved and less genetically variable than the allosteric (noncatalytic) sites bound by non-nucleosides. But mutation may not be the only resistance mechanism. Researchers led by Matthias Gotte at McGill University in Montreal just published evidence for a new mechanism of viral resistance to nucleoside inhibitors: pyrophosphorolytic excision3. The RNA polymerase, in the presence of cellular pyrophosphate, can literally cut the nucleoside analog drug off the chain, allowing the chain to resume growth. How much this happens in HCV is not yet known.
Combination treatment, as in HIV, is an obvious way to get around resistance. But "the challenge in HCV is that your combination today is with interferon and ribavirin, which are not particularly potent agents," says Alam, at least early in treatment, when resistance emerges.
Getting rid of interferon
Once we have them, will combining specific, potent antivirals—without interferon and ribavirin—work in HCV? No one knows; no such clinical trial has yet taken place. But, for the HCV field, the prospect of a combination of, say, a protease inhibitor and polymerase inhibitor is beginning to loom large. Clinicians are desperate for something new to give patients.
But a protease inhibitor and polymerase inhibitor combination alone won't defeat viral resistance, believes McHutchison. "The virus is churning and making mutations—every mutation in the genome—every day," he says. "The chances of being able to obliterate every virus before there's an opportunity for resistance to one or another or both is very unlikely. I think so, anyway. That's why it's very important that we have an immunomodulator... such as interferon."
InterMune's Lawrence Blatt agrees. "I don't think we'll ever find a day when we're not using interferon in hepatitis C, one way or another," he says. "Eradication requires an activation of the immune system and requires the immune system to actually clear out hepatocytes harboring HCV. And I don't think that's going to be achieved without a biologic response modifier like interferon."
But this view is hardly universal. Rice believes that specific antivirals in combination should be able to cure HCV, although it may take three drugs instead of two. "I would be surprised if this wasn't possible," he writes in an e-mail. Gilead's Bischofberger feels that specific antivirals should be enough. "Put enough agents in, [and] at one point you will be able to reduce viral replication to zero," he says. "And at that instance you cannot develop resistance."
Rice and Bischofberger do not think immune system clearance of infected hepatocytes is central to a cure. They point out that most studies do not show adaptive immune system activation during interferon treatment. (Interferon may work instead through direct antiviral effects.) In cell culture, Rice points out, one can eliminate HCV RNA without getting rid of the cells.
The debate is more than academic. Combining targeted agents without interferon is risky because it involves a much longer road to drug approval and is very hard to arrange because few, if any, companies have both protease inhibitors and polymerase inhibitors at the same stage of development. So theoretical arguments could easily derail already problematic clinical trials.
The success of telaprevir after a short treatment course, indicating that adaptive immunity plays a lesser role, is good reason to try specific agents in combination, Alam believes. "The more we show that it's about the virus and not about the immune system, the more compelling it becomes to try those combinations," he says. Can the immune system, and interferon, be removed from the equation? The only answer is empirical, says Alam, who adds that Vertex hopes to eventually test telaprevir together with a polymerase inhibitor in patients, against a combination that includes interferon.
Other HCV protein targets are now emerging. "All of the [viral] proteins are potentially interesting targets," says Rice. Gilead and Achillion, based in New Haven, Connecticut, are collaborating on a preclinical program testing inhibitors of the NS4A protein, which is required for proper assembly of the HCV replication complex. (A previous version was tried briefly in the clinic.) Inhibitors of the NS3 HCV helicase, which unwinds the HCV RNA molecule, and the p7 protein, which creates a putative ion channel and is essential for infectious virus production, are in preclinical development. Rice's laboratory solved the crystal structure of the NS2-3 protease catalytic domain in 2006, and Blatt expects companies to now target the enzyme. The NS5A protein, which appears important for both viral replication and new virus production, is another attractive target, although no functional biochemical assay yet exists. A microRNA, miR-122, appears important for HCV replication and is likely to be targeted. Despite the recent drug trial failures, industry activity is still on the rise.
Now that cell culture systems are available, researchers predict a rush of new papers—and ultimately new drugs. "I think there'll be an exponential amount of information about the viral life cycle, about the interactions of the virus with the host immune system, because those tools are available now," says McHutchison. "And then subsequently new targets, new compounds." Some of these will involve viral entry and viral budding, which can only now be studied in culture. (Viral entry inhibitors against HIV, in contrast, have been in clinical trials for years.) "In three to five years, we'll start seeing the fruits of that effort" in HCV, says Blatt.
Given the scale of the effort and the sheer number of drug targets, it seems inevitable that interferon will eventually give way to some less brutal form of treatment. That day can't come soon enough for patients. But, in the wake of this year's disappointments, it seems further away than ever.