IL28B gene variation predicts response to hepatitis C treatment: will it change treatment of co-infection?

Variations in the IL28B gene that have recently been linked to better treatment response among people with chronic hepatitis C virus (HCV) infection may also predict recovery in HIV/HCV-coinfected people, according to three studies presented at the 17^th Conference on Retroviruses and Opportunistic Infections (CROI) last week in San Francisco.

Andri Rauch began the session with an overview of the recent genetic discovery, which session moderator Ken Sherman said has "revolutionised the way we think about [hepatitis C] treatment and management issues in the past six months".

The IL28 gene encodes interleukin 28, also known as interferon lambda, a cytokine (chemical messenger) with antiviral activity. A related cytokine, interferon alfa, is current standard therapy for hepatitis C. Interferon lambda is being tested in phase 2b trials, and so far it appears to have similar efficacy but fewer side-effects.

Researchers found the gene by scanning the human genome to look for differences between people who responded well and those who did not respond to hepatitis C treatment.

At first they examined genes already known to play a role in immune control of HCV. When these were not confirmed, they broadened their search by doing genome-wide association studies, which look at full sets of genes from multiple individuals.

Dr Rauch and colleagues analysed participants in the Swiss Hepatitis C Cohort, whilst other teams looked at genotype 1 hepatitis C patients in the US (white and African American participants in the IDEAL trial), Australia and Japan.

Strikingly, Dr Rauch said, all the teams found an association between favourable treatment response and single nucleotide polymorphisms (SNPs) in the IL28 gene near the "B" subunit that encodes interferon lambda-3. This finding was "very consistent" among four studies published over the past six months, all looking at HIV-negative individuals.

Each research group identified several relevant SNPs, but one - known as rs8099917 or the "risk allele" - was common to all of them; another, rs12980275, was identified in three studies.

Because people inherit one copy of a gene from each parent, and may get either the high-risk or the protective variation, they can end up with one of three different genotypes: two copies of the high-risk variant, two copies of the protective variant or one of each. (These human genotypes are not to be confused with HCV genotypes, which also predict treatment response; referring to "favourable genotype" now has two distinct meanings.)

Across the studies, the high-risk genotype was approximately half as common amongst people who achieved sustained virological response to interferon-based therapy than amongst non-responders. It was even rarer amongst people who spontaneously cleared HCV without treatment.

Dr Rauch also noted that the high-risk variation occurred with higher frequency amongst people of African descent, intermediate frequency amongst Europeans and lower frequency amongst Asians, which he said "explains at least in part ethnic differences seen in HCV recovery rate both natural and with therapy".

Although its mechanism is not yet known, Dr Rauch speculated that the high-risk IL28 may have less effect on interferon-stimulated genes, which in turn leads to poorer immune response against the virus.

Researchers think, however, that the rs8099917 SNP is not the actual cause of poor response, but rather that it is linked to the causal gene(s) and therefore serves as a marker.

HIV/HCV co-infection

At the same session, three research teams reported results from studies looking at the IL28B gene in HIV/HCV-coinfected individuals.

Julia di Iulio and others from Rauch's team sought to determine the actual causal variant or genetic region tagged by the rs8099917 risk allele. They performed recombinant DNA mapping on three types of hepatitis C patients: those with two copies of the high-risk variant who developed chronic infection, those with two copies of the protective variant who cleared infection, those with two copies of the protective variant who nevertheless developed chronic infection, and those with two copies of the high-risk variant who nevertheless cleared infection.

The first three groups included 15 people each, but the last configuration was uncommon and the researcher could only find two such patients. All were HIV/HCV-coinfected save for six individuals with the protective genotype but chronic infection, who had HCV alone.

The researchers identified twelve haplotypes (sets of genes that occur together), clustering into two families. Most people who cleared HCV fell in the first family, whilst most with chronic infection were in the second family. Within the second family, they identified four SNPs that they proposed as possible candidates responsible for poor antiviral response.

In the second study, Jacob Nattermann and colleagues determined IL28B genotypes for 192 European HIV/HCV-coinfected individuals treated with pegylated interferon alfa, including 74 with acute HCV infection and 118 with chronic hepatitis C. In addition 156 people with HCV alone and 136 healthy individuals with neither virus served as control subjects.

Focusing on the SNP rs12979860, they found that IL28B genotype distribution did not differ significantly across the groups. Amongst the HIV/HCV-coinfected individuals, those with two copies of the protective gene variant were more likely to achieve sustained virological response than those with two copies of the high-risk variant or one of each type. However, this effect was not as strong as for HCV-monoinfected people and was significant only for chronic but not acute hepatitis C. They also found that HIV infection was associated with lower blood levels of interferon lambda.

In contrast, Norma Rallon and colleagues found a strong link between IL28B SNP rs12979860 variants and treatment response in HIV/HCV-co-infected individuals (90 sustained responders, 74 non-responders and 24 with spontaneous HCV clearance).

In this study, 75% of people with spontaneous clearance had the protective genotype, compared with 46% of those who developed chronic infection.

Moreover, amongst patients who achieved sustained response to treatment, 75% had the protective genotype whilst 38% did not. When they looked at HCV genotypes, however, the association was only significant for genotype 1, with a trend for genotype 4.

Clinical relevance

Dr Rauch and the other researchers discussed the clinical relevance for people with hepatitis C of what he called a "breakthrough in our understanding of genetic variations that influence outcomes."

In particular, some suggested that IL28 genotypic testing might help guide treatment decisions, since it could show who is most likely to respond and spare others futile therapy. An audience member expressed concern, however, that such a test might be used as a reason for doctors to deny treatment or for insurers in the US to refuse to pay for it.

"It's part of the puzzle," Dr Rauch said at an accompanying press conference. "It will certainly not be the only thing on which we base the decision to treat or not."

Further information

You can view these abstracts on the official conference website.

You can also view a webcast and slides of this session on the official conference website.