From Medscape Education Infectious Diseases

Understanding Resistance in Chronic HCV Infection: Lessons Learned From HIV and HBV

CME Released: 11/30/2011

CME-Transcript

Jeffrey Burack, MD, MPP, BPhil: Hello. I'm Dr. Jeffrey Burack, Associate Clinical Professor of Medicine at the University of California, San Francisco and Berkeley, and Medical Director of the East Bay AIDS Center in Oakland, California. Welcome to this CME-certified activity titled Understanding Resistance in Hepatitis C: Lessons Learned from HIV and Hepatitis B.

Joining me today are Dr. Steven Flamm, Professor of Medicine at the Northwestern Feinberg School of Medicine and Medical Director of Liver Transplantation at Northwestern Memorial Hospital in Chicago, and Dr. Nancy Reau, Associate Professor of Medicine at the University of Chicago Pritzker School of Medicine.

****CME begins four minutes into the audio.
Sit back and listen to the CME, follow along by viewing the slides provided below.
Click below to begin

Steve, you have a nice analogy for understanding how viruses become resistant to antiviral medications. Would you share it with us?

Steven L. Flamm, MD: Consider that certain enzymes within the virus are critical to its replication and if those enzymes can be inhibited the replication rate can be reduced and, therefore, the viral level can be lowered.

To envision the interaction between the enzyme and the medication, think of the enzyme and its active site as a 3-dimensional structure with a specific conformation -- like a lock, with the medication designed to fit into this lock like a key. If everything works correctly, the lock and key fit together perfectly and the key (the medication) inhibits the ability of the lock (the enzyme) to function. As a result, the virus is unable to replicate well and this causes the viral load to drop.

Dr. Burack: What happens to the "lock and key" in a medication-resistant situation?

Dr. Flamm: Medication resistance occurs when there is a mutation in the enzyme, typically at the active site. The mutation creates a slight conformational change in the enzyme (lock). The medication (key) stays the same and no longer fits into the lock. Consequently, the enzyme (and the virus itself) is able to escape inhibition by the medication and begin functioning again, resulting in increased replication and viral load.

Dr. Burack: In the past 20 years, we have learned a lot about medication resistance from our experience with HIV.

HIV is a rapidly replicating virus whose replication enzymes are highly error prone. While many of the mutations caused by these errors are detrimental to the virus, some may confer an advantage by helping the virus escape from either immunologic or medication control. This is particularly likely to happen when the medication concentration is not high enough to suppress viral replication but sufficient to exert selective pressure on future generations of virions. Thus, we have learned that we need to use multiple medications that target sufficiently different points in the HIV life cycle, or at least at points where cross-resistance does not occur, as well as the importance of optimizing treatment to maintain medication levels, and especially the importance of medication adherence.

Nancy, how is the hepatitis C virus (HCV) similar to HIV? How do the viruses differ?

Nancy Reau, MD: It is very important to not only apply the lessons learned from another virus but also to appreciate the uniqueness of each and every virus. For example, HIV and hepatitis B virus (HBV) are very similar to HCV in that all of these viruses replicate very rapidly and are highly error prone, leading to a potential error during each round of replication cycle.

Some of these mistakes are beneficial, allowing the virus to adapt or change in order to escape detection by the immune system or avoid medication pressure, while other mistakes create virions that are unfit and do not survive. However, as long as the virus continues to replicate so rapidly, mistakes will continue to be made, increasing the likelihood that some of the virus will survive, which is the case for HIV, HBV, and HCV.

The unique aspect of HCV is that, in contrast to HIV and HBV, it doesn't have a mechanism for saving or memorizing these mutations. For example, HIV integrates itself into the host DNA, while HBV is able to sequester itself in the host nucleus in the form of covalently closed circular DNA (cccDNA); using these mechanisms, the virus genetic information is carried forward every time the host cells divide. By contrast, HCV replication takes place in the cytoplasm and the virus never enters the nucleus. Because of that, HCV lacks a mechanism to preserve over time any mistakes that are made, including those that are potentially advantageous.

Dr. Burack: What you just described is what we in the field of HIV term "archived resistance"; that is, once resistance to a medication has developed the patient is presumed to be permanently resistant to that medication. Is that the same with HCV?

Dr. Reau: It is difficult to say with certainty because the only medications available to treat HCV infection until recently -- pegylated interferon and ribavirin -- do not directly target the virus and so have not been associated with emergent medication resistance.

Therefore, HCV resistance is not something we used to worry much about. With that caveat in mind, the hypothesis at the moment is that there is no mechanism by which HCV can archive itself. However, I think it will take time for us to truly understand whether HCV is able to select for mutations and whether these mutations are longer lived than we currently believe them to be.

Dr. Burack: Where are we today in HCV therapy? You mentioned pegylated interferon and ribavirin, but the therapeutic landscape has changed greatly in the past couple of years. How has it evolved over time?

Dr. Reau: Unlike HIV, where a treatment regimen can be designed and individualized to fit the specific patient and the specific virus, the approach in HCV has been to add on to existing therapies.

For example, we have modified the original interferon therapy over time and then added ribavirin to that interferon backbone. Now we have a third class of agents that we can use, but so far we have not been able to get rid of the interferon-ribavirin backbone without losing treatment response.

Although the mechanism by which interferon exerts its therapeutic effect in HCV infection is still unclear, overall it appears to be via immune modulation of endogenously occurring cytokines. Similarly, the mechanisms of action for ribavirin are not fully understood. We know that it has mutagenic potential and we also know that it boosts the host immune response. Which of these actions is most important for HCV therapy remains to be uncovered, but, in the meantime, we cannot yet eliminate ribavirin from our backbone of therapy because of its therapeutic benefit. Importantly, the mechanisms of action of interferon and ribavirin do not induce resistance. Thus, while some patients may be more or less sensitive to the interferon/ribavirin therapy, using these medications does not alter the host in a way that makes this therapy less effective if used again in the future.

Dr. Burack: What about the new direct-acting antivirals?

Dr. Reau: That's why we're having this discussion.The direct-acting antivirals are a very new development in the field of HCV therapy. As Steve said earlier, these medications are aimed at a very specific target within the virus. If we go back to the analogy of the lock and a key, the success of these medications is predicated on a perfect fit between the lock (viral enzyme) and the key (the medication), which may be negated by mistake-induced minute conformational changes (ie, resistance), making these new agents less effective.

Dr. Flamm: And to stress what Nancy said earlier, the reason we haven't discussed resistance in HCV therapy to date is that the medications we have been using for the last 15 or so years do not act directly on the virus, and that is why resistance does not occur with their use. By contrast, there are many medications in the pipeline that we are going to talk about later, as well as the 2 that were just approved, that are targeted directly against one of the viral enzymes. And it's also important to know that the virus has many enzymes that help it replicate. The 2 recently approved antivirals act against just one of those enzymes, a protease.

Dr. Burack: Which protease is that?

Dr. Flamm: The NS3/4A protease; mutations in that enzyme can confer resistance against these new medications. Of course, once the next generation of direct-acting antivirals comes along, resistance to those agents could also develop. The concept of medication resistance is something we are going to have to deal with going forward.

Dr. Burack: Now that we have discussed how some of these mutations arise, let's talk about how resistance can be evaluated.

In HIV-infected patients, there are 2 different ways to look for resistance. One approach is to use genotypic testing to look for mutations that are known to associate with medication resistance.

The other approach is to look more directly at enzyme function with phenotypic testing. Are there similar methods available to test for HCV medication resistance?

Dr. Reau: Yes. In research studies, models have employed both genotypic and phenotypic resistance testing, and we are now able to order genotypic resistance testing for HCV in the clinic. I should note that this assay is only able to detect the presence of a mutation if the prevalence of the mutation within the virus population is at least 20%, meaning that at least 20% of the virus in that patient must harbor the mutation. That's a limitation of this type of genotypic resistance testing because we know that every possible genetic error should be present within the viral pool and if we look hard enough we should be able to find it. The test is available if physicians wish to order it.

Dr. Burack: In research settings, where deep sequencing or clonal sequencing can be done, would you expect to find additional mutations that are not detected by population sequencing?

Dr. Flamm: You certainly could. It's important to understand that every single virion is not examined during resistance testing via this assay, as Nancy explained.

On the other hand, clonal sequencing is able to detect mutations present at a much lower frequency, approximately 5% of a population, and ultra-deep sequencing can detect mutations with an even lower prevalence.

It is important to understand what the capabilities and clinical implications of this test are and how to interpret the results and compare them with results published in the medical literature.

Dr. Burack: Let's discuss some of the clinical implications of resistance in HCV therapy, namely baseline medication resistance. In managing HIV, we have learned that we have to test for the presence of resistance-associated mutations (vs resistance-associated variants for HCV) at baseline, either before therapy is initiated or, ideally, even earlier when the patient is first diagnosed with HIV infection.

The reason is that some patients may have what's called primary or transmitted medication resistance. In other words, they have been infected with an HIV virus that already has some medication resistance.

Nancy, what do we know about transmitted resistance in HCV and how big of an issue is it for patients initiating treatment?

Dr. Reau: There is what we know today and there is what we might learn in the future. Today, no matter how potent the available direct-acting antivirals are, they are still added on to our current therapy -- pegylated interferon and ribavirin -- which, as Steve pointed out earlier, are not associated with medication resistance. So, the presence of a resistance-associated variant at baseline does not mean that this variant will not be sensitive to interferon or ribavirin. Rather, it is the presence of a resistance-associated variant combined with baseline insensitivity to interferon and ribavirin that predicts failure of therapy with a resistant variant. In the phase 3 boceprevir trials, 66 patients with baseline resistant variants were identified. The presence of these variants would potentially impact the treatment course, but they found that it was only the combination of the pre-existing major variant combined with a baseline insensitivity to interferon and ribavirin that could predict whether patient a was going to fail therapy with a resistant variant.

Dr. Burack: Were these patients who responded poorly during the 4-week lead-in phase?

Dr. Reau: That's right. We also need to keep in mind that this study looked at a population of more than 1000 boceprevir-treated patients, and this combination of interferon/ribavirin insensitivity and baseline resistance mutations was present in only 66 patients (or about 7% of the total patient population). That means that you would have to test a very large number of patients to identify the few for whom baseline resistance would actually be an issue. Moreover, even if you obtained the baseline resistance profile, you would still be missing important clinical information, such as interferon sensitivity, which you need to predict treatment response more accurately.

Dr. Burack: It sounds like there is still a lot we don't know about the clinical utility of baseline resistance testing. Steve, what do we know about the emergence of medication resistance during HCV therapy?

Dr. Flamm: Emergence of resistance-associated variants during therapy is closely associated with treatment failure, whereas some patients who have resistance-conferring mutations at baseline still respond to therapy. The fact that those mutations are acquired while on therapy is a very bad sign associated with treatment failure. It behooves us to try to limit the development of resistance. In addition to dealing with failure of a patient's current therapeutic regimen, we also need to consider the long-term implications of emergent medication resistance. As Nancy mentioned, the fact that archiving does not appear to occur in HCV might imply that the long-term implications of medication resistance in HCV may not be profound. However, we cannot be sure of that yet because these agents are still very new and more data are needed to draw more definite conclusions.

Dr. Reau: We also know that a mutation can affect the fitness of the virus variant. So, although apparently the virus population eventually returns to close to its baseline composition after treatment is stopped, if you allow enough compensatory mutations to occur you could actually select a variant that is very fit and may not go away. It's not due to archiving; it's simply a reflection of the fitness so that you want to identify the person who is going to fail therapy, not necessarily just through resistance testing. If you don't stop the medication and you continue to offer that pressure, compensatory mutations will occur and you are going to change that virus even more.

Dr. Burack: We hear a lot about futility rules in HCV treatment or stopping therapy promptly when a patient remains viremic. Is that the reason or are there other reasons that we need to pay attention to that?

Dr. Reau: If you treat a patient with interferon and ribavirin, even without adding a direct-acting antiviral agent, the ability to stop an expensive medication that has side effects at a time when it is inevitably not going to work benefits the patient, decreases cost, and benefits the office staff who are trying to maintain that patient. The earliest you can determine that therapy is ineffective, the better.

Dr. Flamm: We have had futility rules for at least the past 10 years. The reasons to stop the medicine are that the medicines cost a lot and they have a lot of side effects. If they are not working, you should not expose the patient to them. In the past, the futility rules have not always been followed. It is even more important now because when you have treatment failure, the vast majority if not all patients develop resistance. The longer you expose a patient to a medication, such as a direct-acting antiviral agent, that isn't working (because there is already resistance), the more you allow for the development of additional mutations, which make the resistant enzyme even more difficult to block. If that enzyme persists and leads to a fit virus, or if it persists over the long term, it could have implications for future medical therapy because of what we call "cross resistance," or resistance created earlier to a medication that appears later. So it is imperative that when patients meet the criteria for the futility rules they are adhered to by the healthcare team and the medications are stopped immediately because of the potential for the development of resistance.

Dr. Burack: If patients stop therapy under those circumstances do those mutations go away?

Dr. Flamm: That's a great question. We don't have a lot of long-term information on this, but patients in the HCV protease inhibitor trials who failed therapy and have resistance-associated variants have been followed over the long term. The final data have not been presented, but from the preliminary reports it appears that the vast majority of patients lose the resistance-associated variants in less than 2 years and the wild type returns. Again, there are limitations to resistance testing. We can only detect resistance-associated variants if they occur at a certain prevalence in the patient. The preliminary information tells us that these resistance-associated variants seem to disappear over the relative short-term.

Dr. Burack: One group of investigators used clonal sequencing to examine this. What did they find?

Dr. Flamm: They found the same thing when they looked at some of the patients they had been following with population sequencing (for which you need larger amounts of resistant virus), but when they looked at smaller amounts they found that the resistance-associated variants seem to disappear and the wild-type variants overtook the resistance-associated variants. The resistance-associated variants were less fit and reverted back to the baseline viral population.

Dr. Burack: Given that there is no archiving with HCV as there is with HIV or HBV, might this suggest that it is at least theoretically possible that a patient who developed medication resistance could be treated again in the future with the same medication in a different regimen?

Dr. Reau: I don't think you would want to treat the patient with the exact same regimen because if they failed that regimen it's very likely that it would be unsuccessful in the future. But as we know from HIV and HBV, if you add a complementary medication that is active against the mutation that is likely to be selected by reintroducing the HCV protease inhibitor that the patient previously failed, you could possibly control and eliminate that.

Conversely, it is very difficult to adhere to a futility rule, but the price of not adhering in the past was cost and side effects. We now know that there is an additional cost, and we also know, most importantly, that the futility rules truly are futility rules; just because you see a nice decline in the virus, if it hasn't declined robustly this is usually a sign that resistance has already developed. As hard as it is to convince your patient, who has had a monumental decrease but not quite enough of a decrease, to stop these medicines that they feel are successful, it has to be done.

Dr. Burack: If patients stop therapy under those circumstances do those mutations go away?

Dr. Flamm: That's a great question. We don't have a lot of long-term information on this, but patients in the HCV protease inhibitor trials who failed therapy and have resistance-associated variants have been followed over the long term. The final data have not been presented, but from the preliminary reports it appears that the vast majority of patients lose the resistance-associated variants in less than 2 years and the wild type returns. Again, there are limitations to resistance testing. We can only detect resistance-associated variants if they occur at a certain prevalence in the patient. The preliminary information tells us that these resistance-associated variants seem to disappear over the relative short-term.

Dr. Burack: One group of investigators used clonal sequencing to examine this. What did they find?

Dr. Flamm: They found the same thing when they looked at some of the patients they had been following with population sequencing (for which you need larger amounts of resistant virus), but when they looked at smaller amounts they found that the resistance-associated variants seem to disappear and the wild-type variants overtook the resistance-associated variants. The resistance-associated variants were less fit and reverted back to the baseline viral population.

Dr. Burack: Given that there is no archiving with HCV as there is with HIV or HBV, might this suggest that it is at least theoretically possible that a patient who developed medication resistance could be treated again in the future with the same medication in a different regimen?

Dr. Reau: I don't think you would want to treat the patient with the exact same regimen because if they failed that regimen it's very likely that it would be unsuccessful in the future. But as we know from HIV and HBV, if you add a complementary medication that is active against the mutation that is likely to be selected by reintroducing the HCV protease inhibitor that the patient previously failed, you could possibly control and eliminate that.

Conversely, it is very difficult to adhere to a futility rule, but the price of not adhering in the past was cost and side effects. We now know that there is an additional cost, and we also know, most importantly, that the futility rules truly are futility rules; just because you see a nice decline in the virus, if it hasn't declined robustly this is usually a sign that resistance has already developed. As hard as it is to convince your patient, who has had a monumental decrease but not quite enough of a decrease, to stop these medicines that they feel are successful, it has to be done.

Dr. Burack: So those patients are not only destined to have virologic failure, they will develop even more medication resistance as you continue to treat them. Is there utility in HCV resistance testing at the time of treatment failure when the patient is being taken off the medications?

Dr. Reau: There are times when you might want to know this. If we had other agents that you might re-expose the patient to, then you would need to know what that patient does and does not have sensitivity to. Telaprevir and boceprevir are complementary with enough cross resistance that you can't use one when the other fails. If a patient failed treatment and you tested them for resistance and you didn't see it, you would assume that it was because the test wasn't accurate enough, so I'm not sure that information obtained at the discontinuation of therapy would be clinically useful today.

Dr. Flamm: There are questions about resistance testing. First, should it be checked at baseline? The very limited data we have suggest that even if a patient has baseline resistance, they may still respond to therapy, so you would not withhold therapy from that patient. There may be times when baseline resistance testing is useful, but at this time we really don't know what to do with the information. If you follow patients out for a year or 2, the majority appear to lose the resistance-associated variants. It's difficult at this time to ascribe a clinical utility for resistance testing. It may be true in the near future that when a patient fails therapy it is important to know what they developed resistance to, because even if they lose the resistance-associated variants over time, it may have an impact if they later choose another therapeutic regimen, but right now we don't know.

Dr. Burack: Another interesting concept taken from HIV therapy is the barrier to resistance of different medications. Some medications require more than 1 amino acid change, for example, in order to be highly resistant to a given medication. Some mutations might come with a significant fitness cost. There is also the idea that some resistance-associated mutations require more than 1 nucleotide change. Steve, could you explain? I understand that this is part of the reason that we see variation in susceptibility of different HCV genotypes to some of these medications.

Dr. Flamm: The protease inhibitors have raised an issue that we didn't worry about before, the subtype of the HCV genotype. I should mention that these medications are only approved for genotype 1; they are not approved for genotypes 2, 3, or 4. Our test results have always told us whether the person was HCV genotype 1a or genotype 1b, and I would be asked whether it mattered. My answer was always "no," but now it may be "yes" because genotype 1a and 1b have different thresholds for developing resistance.

We need to remember how resistance develops. The changes in the lock that we mentioned earlier are mutations of nucleotides, amino acid substitutions, and for genotype 1a only 2 steps are required (a relatively small number of steps to develop a mutation that sticks or confers resistance). Four steps are required for genotype 1b; it's a much more complicated process to develop an effective mutant resistance-associated variant to confer resistance. Genotype 1a is very different from genotype 1b, at least in terms of the development of resistance-associated variants with the protease inhibitors.

Dr. Burack: Let's talk about what is in the pipeline. What do you foresee as the future of HCV therapy?

Dr. Flamm: One of the long-term strategies to prevent the development of resistance is to add more medications that have complementary modes of action, similar to how we treat HIV when single medications have not been effective because of the development of resistance.

Where multiple complementary medications work well for HIV, I think the same will be true in HCV; there are many medications in the pipeline in other classes: enzyme inhibitors, polymerase inhibitors, NS5A inhibitors, and even cyclophilin inhibitors against the host target. My suspicion is that there will be combinations of these medications that will improve efficacy by diminishing the development of these resistant mutations

What about pegylated interferon and ribavirin? It appears that ribavirin is still necessary. The studies done to date show that when you eliminate ribavirin from the mix, there are more variants and the sustained virologic response is lower, so it appears that ribavirin will be necessary in the future. Interferon, probably not. A study presented at the 46th Annual Meeting of the European Association for the Study of the Liver showed that 4 out of 11 (36%) genotype 1 null responders (people who had previously been on pegylated interferon and ribavirin and had no response) given only 2 oral medications without ribavirin had a sustained virologic response rate after only 6 months. There was a presentation at the 62nd Annual Meeting of the American Association for the Study of Liver Diseases this year on data from 10 Japanese patients with genotype 1, subtype 1b, who were treated with these same 2 oral medications without ribavirin; all 10 had a sustained virologic response. It seems that if we get the right combination of complementary medications we will eliminate or limit the development of resistance and may not need interferon to obtain a sustained virologic response. That, of course, is the Holy Grail in the treatment of chronic HCV infection.

Dr. Reau: This is also likely to come with a lot more thought in the beginning of therapy. We only have a couple of choices and we have to use a combination of what exists. We are not only differentiating between genotypes 2 and 3 compared with genotype1, but we are talking about subtypes. This might be a place for baseline resistance testing to identify whether a patient is a good fit for an oral regimen or not, so we might be coming a lot closer to the HIV world.

Dr. Burack: Let's give some capsule advice. How do we prevent the emergence of resistance?

Dr. Flamm: Liver doctors need to incorporate some of the concepts that the HIV doctors use. It is very important when you use a direct-acting antiviral agent that it be taken exactly as prescribed; in this case, this means that you must give it with pegylated interferon and ribavirin. The medications need to be taken 3 times daily with food, as prescribed. If a lot of doses are skipped, the virus will have a chance to increase replication because the lock and the key are not together and resistance can occur.

If you have to discontinue either of those medications for any reason, the protease inhibitor also needs to be discontinued or resistance will develop. The medications cannot be stopped for a week or 2 and restarted, and the dose cannot be reduced. This is another foreign concept for doctors because when a medication side effect occurs our knee-jerk reaction is to lower the dose. In this case, we cannot. We need to try to manage the side effects and if we can't, then we should stop the medication completely. Lowering the dose is a very bad idea.

The way to optimize therapy for HCV -- and this is more important than ever -- is to counsel the patient properly before you start therapy and during therapy.

If patients are educated and clinicians adhere to the principles that we discussed, you will be able to optimize the therapeutic regimen for patients with chronic HCV infection.

Dr. Burack: The great promise for these patients, of course, is that although this might require intensive work on their part and on yours, it is time-limited treatment that may cure chronic HCV infection.

Dr. Reau: It is a finite therapy with a much better sustained virologic response rate, which impacts cancer development and transplant choices.

Dr Flamm: This discussion of resistance is not to imply that we should be scared to treat patients with chronic HCV infection. The bottom line is that approximately 65%-80% of patients who are treated and adhere to therapy will clear HCV, so this is not to dissuade physicians to treat chronic HCV infection. It is to make them aware of how to best care for their patients.


Dr. Burack: I want to thank my colleagues and thank you for participating in this activity.

To proceed to the online CME test, click on the Earn CME Credit link on this page.