This article is more than 5 years old. Information may no longer be current.

Making the Case for Studying Factors Associated With Clearance and Relapse

One of the by-products of sustained viral response rates at or above 90% in most populations with hepatitis C is that the very nature of research into the disease is altered, perhaps irrevocably. For most of the history of HCV, while clinicians and researchers desperately searched for a cure — or at least improvement of side effects and end results with interferon-based therapy — concurrent research was conducted in other areas of HCV, including the factors associated with spontaneous clearance and those that cause recurrence or relapse. Today, with all eyes on DAA therapies, there seems to be less room for those peripheral analyses.

Many of the factors discovered in pre-DAA investigations still apply today, according to Sammy Saab, MD, MPH, professor of Medicine and Surgery, assistant professor of nursing and head of Outcomes Research in Hepatology at the David Geffen School of Medicine at UCLA. “We know that spontaneous clearance can happen in anywhere between 15% and 25% of individuals,” he said in an interview with HCV Next. “People who are HIV-negative, not drug or alcohol users and those who have the IL28B CC polymorphism are more likely to clear. Women are more likely to clear than men, but we still don’t know why.”

He added that, somewhat paradoxically, certain groups of injection drug users actually have higher rates of spontaneous clearance, but the reason for this is also unknown.

Predictors of failure to spontaneously clear include concomitant liver disease and HIV or hepatitis B coinfection, according to Saab. “If you have the IL28BTT allele, you have a significantly lower chance of clearance,” he said.

This is a fair body of evidence from which clinicians can draw. But Robert S. Brown Jr., MD, MPH, vice chair, Transitions of Care, and interim chief, division of gastroenterology and hepatology at Weill Cornell Medical College in New York, is not optimistic that it will grow. “People will be much less interested in factors that cause spontaneous clearance,” he said. “In the era before DAAs, we knew that early intervention with interferon could clear people with acute HCV. Now that you have therapies that are 98%, 99% effective, you can give patients time to clear on their own because there is no real advantage to treating someone at 1 month vs. 6 months unless we could show that a short course of therapy would clear patients.”

For Brown, this final point is of utmost importance. HCV therapies are expensive. If large numbers of patients can be treated and cleared in the acute phase with 6 or 4 or even 2 weeks of therapy, as opposed to a full 12-week course, the savings to the health care system will be considerable. “I would like to see a study showing which patients may be treated with substantially less therapy,” he said. “This would come with understanding who will spontaneously clear and who won’t.”

Regarding recurrence or relapse, the picture is even muddier, and the consequences of failure to elucidate it could escalate, according to Stacey A. Rizza, MD, associate professor of medicine in the Department of Infectious Diseases at the Mayo Clinic in Rochester, Minn. “We have terrific drugs, but we are still getting 5% who don’t respond, who either fail to respond at all or respond at first and relapse,” she said. “That 5% will add up over time, and at some point we will have a considerable number of patients to manage.”

Saab said, in general, there are some building blocks of knowledge about patients who fail or relapse. “The things that happen most in relapse are having resistance to therapy, unsuspected advanced liver disease, insufficient duration of therapy and medications that interfere with therapy or the duration of therapy,” he said.

But Rizza warned there are many unknowns. “We don’t know whether they are the same immunologic entity in terms of non-response or response and relapse,” she said. “How much has to do with the host vs. the virus? We don’t know. How important a factor is stage of liver disease? We don’t know that, either. We still have a lot to learn. We can’t say, here is one feature that causes people to respond or relapse, assuming the virus is sensitive to the drug and patients are compliant.”

Spontaneous Clearance

Saab said spontaneous clearance is a difficult area to study because most patients with acute HCV are asymptomatic. “You can never get a good sense of the true denominator,” he said.

Rizza suggested much of what is known about spontaneous clearance, with treatment completely removed from the picture, has been learned from chimpanzee models. “The research has taken a very immunological direction,” she said. “If a person has a strong innate immunological response, they are more likely to clear.”

Understanding the cellular immune response is critical to understanding spontaneous clearance, according to Rizza. “People who mount a stronger and broader T-cell response during the acute phase are more likely to clear than those who have a narrow, shallower immune response,” she said. “A broader immune system can target many targets on the virus, hitting as many epitopes as possible.”

The unknown, for Rizza, is what makes certain individuals have a stronger and broader immune response than others. “Obviously, in general terms, younger people are more likely to have a strong immune system, while pregnant women and people with HIV will have weaker immune systems,” she said. “But we still don’t know what it is that makes people have a particular immune response. There is pretty good understanding of the IL28B association, but genetic factors don’t tell you about immune response.”

There are also a lot of questions about the nature of the virus itself, and how that plays into spontaneous clearance. “We think that it has to do with a more diverse viral load or a less diverse viral load,” she said. “But overall, there are far more factors that we need to know.”

For Brown, there may be lessons to be learned from previous eras. “Why does prior interferon non-response predict non-response to direct acting antiviral?” he said. “If you take the immune system out of the treatment equation, why should it matter? It matters because the immune system probably is important.”

The IL28B Association

Huang and colleagues investigated HLA class I and class II genes, along with the IL28B SNP (rs8099917) genotype, in 231 blood donors in China who had cleared HCV spontaneously and 429 patients with chronic infection.

The analysis indicated that HLA-A*02:01 and DQB1*05:02 distributed differently between spontaneous clearers (OR = 1.839; Pc=0.024) and the patients who are chronically infected (OR = 0.547; Pc = 0.016) after Bonferroni correction. Multivariate analysis pointed to A*02:01 (OR = 1.798, P = .008) and DRB1*11:01 (OR = 1.921, P = .005) as factors associated with spontaneous clearance regardless of age, sex and IL28B status. “We concluded that in the Chinese population, HLA-A*02:01 and DRB1*11:01 might be associated with the host capacity to clear HCV independent of IL28B, which suggests that the innate and adaptive immune responses both play an important role in the control of HCV,” the researchers concluded.

“The challenge in interpreting data about this association is that they are mostly retrospective studies,” Saab said. “I am not aware of any prospective studies looking at this.”

The difficulty is simply that the early course of the infection is largely asymptomatic, so it is only when a patient is tested for something else and HCV antibodies are found that it is possible to determine that the patient was infected and cleared. But it also comes back to conducting research in a setting of limited resources, according to Saab. “With all the advances in HCV, a lot of the attention has shifted to treating people with disease,” he said. “It used to be that we would check the IL28B phenotype for the likelihood of spontaneous clearance, but it is no longer used commonly.”

Genetic studies are also being conducted. Waldron and colleagues investigated blood samples from 95 patients with chronic HCV infection and 62 patients who had spontaneously cleared. They conducted a genetic analysis of 12 pre-selected genes that had demonstrated previous association with host response to the virus. Results indicated that HLA-DQB1 (P = 1.76 X 10–5) and IL-6 (P = .0007) demonstrated significant associations with spontaneous clearance. Conversely, no association with spontaneous clearance was reported for IL28B (P = .17).

Findings like these simultaneously provide further information but also highlight so much of the uncertainty surrounding the genetics of spontaneous clearance.

Further Genetic Associations

Looking deeper, Singh and Dass investigated the KIR2DL3 and IFNL3 genes, along with the haplotype and copy number variation. The findings indicated that “mutations in H77Y of KIR2DL3 and R157Q, H156Y, S63L, R157W, F179V, H128R, T101M, R180C, and F176I of IFNL3 results in conservation, [root mean square deviation], total energy, stability, and secondary structures revealed a negative impact on the structural fitness,” Singh and Dass wrote. Other results indicated that alleles including rs270779, rs2296370, rs10423751, rs12982559, rs9797797, and rs35987710 of KIR2DL3 and rs12972991, rs12980275, rs4803217, rs8109886, and rs8099917 of IFNL3 may have a protective effect on clearance of the virus. “This roundup report specifies the effect of NK cell receptor, KIR2DL3 and IFNL3 variants that can have a better prospect in [genome-wide association studies] and immunogenetic studies leading to better understanding of HCV clearance and progression.”

In another study, Alves and colleagues aimed to determine how prevalent IFNL4 G > TT variants are and to assess whether they are associated with spontaneous clearance of HCV in a cohort of patients coinfected with HIV. The study included 138 individuals in Brazil. The spontaneous clearance rate in the cohort was 24.6%. The researchers reported significant differences in IFNL4 genotype distribution among patients who spontaneously cleared and those with chronic HCV disease (P = .002). Patients with the IFNL4 TT/TT genotype were 3.6 times more likely to spontaneously clear than those carrying the IFNL4 G allele (OR = 3.63; 95% CI, 1.51-8.89), according to the findings. “The IFNL4 G > TT polymorphism seems to be better than IFNL4 rs12979860 to predict spontaneous clearance of the HCV in Brazilian HIV-1 positive patients,” they concluded.

Saab acknowledged the association between IFNL4 and spontaneous clearance, but said that the details remain uncertain. “We don’t yet fully understand the association with the lambda interferon receptor, besides that it is near IL28B,” he said. “There is no causal element, only an association, so this does not play a big role in clinical practice.”

Duration of Therapy

For Brown, investigating issues surrounding spontaneous clearance and relapse or recurrence could pay dividends in terms of preserving resources.

“There is value in studying spontaneous clearance as part of a study of reinfection,” he said. “If we are going to treat high-risk individuals such as active injection drug users, which is something we never did in the interferon era, we can study risk factors for reinfection, strategies to decrease the risk of reinfection. It could be possible, then, to figure out who might clear spontaneously and who might be able to get a short course of treatment.”

A shorter course would translate to savings and have benefits far beyond the individual patient level. It is simple math. “If you could figure out a way to shorten the course from 12 weeks to 2 weeks, you could re-treat more people for the same price,” he said.

This approach could even appeal to payers, according to Brown. “We have a difficult time getting coverage for one course of therapy in IV drug users,” he said. “If they are not paying for 12 weeks of treatment, they may be more likely to cover.”

Genetic factors may return to the forefront of this discussion, as well. “This is the area where IL28B, which most people say is dead, might come back,” he said. “We might find that, say, patients with IL28B and genotype 2 or 3 disease can get away with a different or shorter course of therapy. This could be beneficial in terms of preserving resources. But these are difficult questions because knowing what to look at, you can only study so many regimens in high-risk groups.”

Brown is clear that failure to study these factors will have a financial impact. “There have been trials of 4 weeks of therapy that resulted in a 35% SVR rate,” he said. “Those studies were deemed failures. But for those 35%, they were a success. If you can figure out who those people were, you can save one-third of the cost of drugs. Not studying these factors is costing us money.”

Unfortunately, he said that until the pressure mounts, it is unlikely that such studies will be conducted. “Insurers might push us in the direction of using fewer drugs, but do we have enough information to do these studies and make these decisions?” he said. “We don’t understand enough about what is happening in the viral kinetics of patients who clear in a very short course of therapy, just like we don’t understand predictors of spontaneous clearance.”

Rizza summed it up: “Right now, there is no strong association to predict who will respond and who won’t. ... As a result, we are over-treating some patients and under-treating others.”

Small Numbers, Big Concerns

Perhaps the most unsettling thought about the small numbers of patients who failed to respond, responded and then failed, or relapsed, is that little is being done to ensure that options are available for salvage therapy.

“We have relapsers and non-responders,” Rizza said. “The world we are living in now is one where these patients are adding up.”

Saab expressed concern, as well. “I don’t have a good feeling about what to do with relapsers and non-responders,” he said. “There are going to be problems down the line with these failures. For people who fail current therapies, there is no consensus.”

For Brown, it is a matter, almost, of being victims of success. “When the SVR rate is 99%, you are never going to find predictive factors for clearance,” he said. “The same factors that predict clearance with therapy always predict spontaneous clearance. IL28B CC patients on the margin always have slightly better clearance. If you start looking at treatments that haven’t worked as well, you see women do a little better, non-diabetics do a little better. All the risk factors are still there, but we stopped paying attention because we have such high rates of cure. We have a good solution for virtually every patient now. Scientifically, it’s hard to study because there are so few patients. The tiny little differences that can fuel research aren’t there.”

Improved salvage therapies for treatment failures and relapsers may yield studies that will answer some of these questions, but the Catch-22 is that these studies are necessary to develop effective salvage therapies.

Unanswered Questions

“Women spontaneously clear a little better than men, but nobody knows why,” Rizza said. “In general, women tend to mount stronger immune responses. This has been questioned a lot with no conclusive answers. It may be worth noting that women tend to have higher rates of autoimmune disease, which could be something to look at.”

Brown added that women generally have higher cure rates with therapy. “With interferon, lower body weight was a factor in improved response,” he said. “They may also tend to get more mild disease or be more compliant with medications. But now that we are not using interferon, we have to assume that maybe there are other factors.”

He noted that in small trials involving traditionally difficult populations, women tended to perform better. “This could be related to their tendency toward more frequent spontaneous clearance,” he said. “If you are more likely to clear with therapy, it is possible that you are more likely to clear spontaneously. But we don’t know.”

In the absence of studies specifically looking at recurrence, it is necessary to glean information about it from other types of investigations, including those about resistant virus. Jimenez-Perez and colleagues noted that despite high SVR12 rates almost across the board, 10% to 15% of patients still experience therapeutic failure. While the researchers noted that pre-existing resistance-associated variants do not appear to predict response, those that persist after therapy may determine which second-line therapies may be useful. “As well as the presence of RAVs, virological failure to treatment with DAAs is generally associated with other factors related with a poor response, such as the degree of fibrosis, the response to previous therapy, the viral load or the viral genotype,” the researchers wrote. “Nonetheless, viral breakthrough and relapse can still occur in the absence of detectable RAVs and after the use of highly effective DAAs, so that the true clinical impact of the presence of RAVs in therapeutic failure remains to be determined.”

“Resistance is still a strong predictor of relapse,” Saab said.

Morishita and colleagues investigated the role of RAVs in the NS3 region in a cohort of 21 patients with HCV genotype 1b. Eligible participants were treated with simeprevir (Olysio, Janssen) after failure with telaprevir-based therapy. Results indicated a 50% overall SVR12 rate, after one patient discontinued treatment due to adverse events. Telaprevir-resistant variants were found in six patients, while no patients demonstrated variants at position 168, according to the findings. One patient demonstrated cross-resistance between telaprevir and simeprevir, but this patient reached SVR12. All patients who discontinued telaprevir-based therapy due to adverse events achieved SVR12, compared with just 29% among patients who did not discontinue telaprevir (P = .005). Similarly, patients who relapsed after peginterferon and ribavirin therapy reached a 100% SVR12 rate, while just 20% of those who did not respond to peginterferon and ribavirin reached SVR12 with simeprevir (P = .007). “Ultra-deep sequencing analysis revealed that [telaprevir] and/or [simeprevir]-resistant variants may have no influence on the effect of [simeprevir]-based therapy after failure of [telaprevir]-based therapy,” the researchers concluded. “Patients who discontinued treatment owing to adverse effects of [telaprevir]-based therapy and relapsers to previous [peginterferon and ribavirin] therapy would be good candidates for retreatment with [simeprevir]-based therapy.

“Resistant virus predicts relapse at a higher rate,” Brown said, but acknowledged that the mechanism of this is still poorly understood.

Saab agreed. “The genetic basis of recurrence is still up in the air, largely because it is such a rare event,” he said. “There are not a lot of data on this topic.”

Other Issues

“Patients who are jaundiced tend to clear at a higher rate,” Brown said. “This is presumably because they have more immune activity against the virus. This is why their liver gets more damaged.”

Further understanding this phenomenon may be beneficial, according to Brown. It may also be beneficial to understand whether the mode of transmission plays a role in either clearance or recurrence. “If we could look at patients who acquired the disease sexually versus those who acquired it through IV drug use, and maybe look at some genetic factors, we may see another basis to clearance,” he said.

Rizza offered a practical reason for continuing to explore these topics. “The reason we investigate spontaneous clearance is generally for vaccine development,” she said. “If we can figure out why some people’s immune systems mount a defense and clear, we try to replicate that in a vaccine.”

As for relapsers, she said, “We study them to figure out how to treat them.”

This, in fact, could be the overall goal for all of this research. This could be the goal for the studies that have been done and the studies that need to be done, according to Rizza. “Our job is to provide adequate therapies for patients who are predicted to be poor responders,” she said. “And to decrease treatment for those who are strong responders.” — by Rob Volansky

• References:
• Alves CF, et al. Genet Mol Biol. 2016;doi:10.1590/1678-4685-GMB-2015-0106.
• Garbuglia AR, et al. PLoS One. 2016;doi:10.1371/journal.pone.0158989.
• Huang J, et al. Sci Rep. 2016;doi:10.1038/srep31485.
• Jimenez-Perez M, et al. World J Gastroenterol. 2016;doi:10.3748/wjg.v22.i29.6573.
• Kwok RM, et al. Liver Transpl. 2016;doi:10.1002/lt.24614.
• Morishita N, et al. Hepatol Res. 2016;doi:10.1111/hepr.12817.
• Singh P and Dass JF. Mol Biol Rep. 2016;doi:10.1007/s11033-016-4044-5.
• Waldron PR, et al. J Immunol Res. 2016;doi:10.1155/2016/6530436.

• For more information:
• Robert S. Brown Jr., MD, MPH, can be reached at New York-Presbyterian/Weill Cornell Medical Center, 1305 York Ave., New York, NY 10021; email: rsb2005@med.cornell.edu.
• Stacey A. Rizza, MD, can be reached at 200 1st St SW, Rochester, MN 55905; email: Nellis.Robert@mayo.edu.
• Sammy Saab, MD, MPH, can be reached at Pfleger Liver Institute, 200 Medical Plaza, Suite 214, Los Angeles, CA 90095; email: ssaab@mednet.ucla.edu.

Disclosures: Brown reports a consulting role with Bristol-Myers Squibb and Gilead Sciences. Rizza reports no relevant financial disclosures. Saab reports being a speaker and a consultant for AbbVie, BMS, Gilead Sciences, Janssen and Merck.