Transcript HCV genotype 1 - Canadian Co

Favourable IL28B Genotypes are Associated with Spontaneous Clearance but
not Aboriginal Ancestry in HIV/Hepatitis C (HCV) Co-infected Individuals from
the Canadian Co-infection Cohort
Nasheed Moqueet
Dept of Epidemiology, Biostatistics
and Occ Health
1020 Pine Avenue West
Montreal, Quebec H3A 1A2
[email protected]
Phone: 514-934-1934 #32227
Nasheed Moqueet*1,2, James Young2, Connie Lisle3, Ron Carter3 and Marina B. Klein1,2
1McGill
University, Montreal, QC Canada; 2 Montreal Chest Inst., McGill Univ. Health Ctr, Montreal, QC Canada; 3McMaster University, Hamilton, ON Canada
Results
Abstract
Background Homozygous recessive SNPs (rs12979860 CC, rs8099917 TT) in the human Interleukin 28B (IL28B) gene
have been associated with spontaneous HCV clearance and better treatment response in various populations worldwide,
possibly via functional variants at rs8103142. These relationships and genotypic distributions have not been studied in
Canadian HIV/HCV co-infected populations where aboriginal people are overrepresented. Previous studies have also
suggested that aboriginals (First Nations, Metis, Inuit) may have greater rates of spontaneous clearance.
Methods A case-control study nested in a prospective multicentre Canadian Co-infection Cohort (n=1020). Only those
who have never been treated for HCV with at least 2 consecutive RNA tests (n=436) were included. We matched on selfreported ethnicity and used incidence density sampling to select for each HCV negative case (n=62), 2 HCV+ controls
within 2 years of when a case occurred. Conditional logistic regression was used with multiple imputation to account for
missing data and adjusted for sex and HCV genotype (1 or 4 vs. 2 or 3). In addition, all aboriginal participants were
genotyped (n=135).
Results Overall, 73% were male with median HCV duration of 18 years; 15% reported aboriginal ancestry. 62
participants cleared spontaneously in 967.5 person-years of risk (6.4/100 person-years, 95% CI 5-8.2/100 p-y). The
adjusted ORs (95% CI) for spontaneous clearance for each SNP analysed were---rs12979860: 3.44 (1.70, 6.95); female,
1.73 (0.82, 3.65); and HCV genotype 1 or 4, 0.64 (0.24, 1.69); at rs8099917: 6.61(2.20, 19.89); female, 1.89 (0.90, 3.98);
and HCV genotype, 0.68 (0.24, 1.93), and at rs8103142: 4.19 (1.93, 9.08); female, 1.80 (0.82, 3.92), HCV genotype, 0.52
(0.20, 1.38). Ethnicity matching precluded analysis of the association of aboriginal ethnicity with spontaneous clearance.
However, among aboriginals, at rs12879860, 74.5% had the C allele, at rs8099917, 81.4% had the T allele, and at
rs8103142, 75% had the T allele. The corresponding values among Canadian whites are 72.8%, 81.9%, and 75%,
respectively. These allele frequencies are very similar to those previously reported among European populations, all
lower than in East Asians.
Conclusions Genotypes rs12979860 CC, rs8099917 TT, and rs8103142TT in the IL28B gene are strongly linked to
spontaneous clearance in HIV/HCV co-infected Canadians. These genotypes do not appear to be more common among
Canadian aboriginals. Resolvers were also more likely to be female and infected with HCV genotypes 2 or 3, though
these associations did not reach statistical significance.
Background
. Table 1: Baseline Characteristics of CCC, the source population, n=1,020
Variable
Summary
Median follow-up time, years (IQR)
Mean age at baseline, yrs (SD)
Male, n(%)
Ethnicity, n (%)
White
Black
Aboriginal
Other
2 (0.7-3)
45 (8)
744 (73%)
Median HCV duration, yrs (IQR)
Injection drug user, n (%)
HCV genotype 1, n (%)
Ever treated for HCV, n (%)
18 (11-25)
822 (81%)
556 (72%)
175 (17%)
782 (78%)
38 (4%)
155 (15%)
29 (3%)
Table1
• Most people have been HCV infected for a long time.
Figure 1:
• Majority of the participants are male, older (mean age=45 yrs) and have a
history of drug use.
o Prevalence of all beneficial alleles is fairly common in both populations
o IL28B allele frequencies are almost identical among Canadian whites and aboriginals
• Very few had ever been treated for HCV at first visit.
o Higher clearance rates among aboriginals are not due to differential frequency of IL28B alleles
Table 2: Characteristics of matched case control sets
Ethnicity-matched
Controls (n=97)
Female
31 (27%)
Aboriginal
20 (17%)
HIV and HCV co-infection affects over 10 million individuals worldwide. Adverse outcomes such as HCV treatment failure and
viral persistence are higher in co-infected individuals compared to HCV monoinfected patients. Thus, there is a critical need to not
only understand the pathogenesis of HCV in co-infection, but also identify factors that would aid in clinical decision-making.
HCV genotype, n (%)
1
2
3
 Some factors independently associated with higher rates of spontaneous HCVclearance include female sex, aboriginal ethnicity
and infection with HCV genotypes 2 and 3.
Event rate
(95% CI)
 Certain homozygous recessive SNPs (rs12979860 CC, rs8099917 TT) in Interleukin 28B (IL28B) are also associated with
spontaneous HCV clearance and/or better response to treatment.
rs12979860
**CC, n (%)
CT, n (%)
TT, n (%)
 These SNPs are tightly linked with the SNP at rs8103142, which leads to a nonsynonymous amino acid substitution (K70R) in
the IL28B protein.
 These relationships have not been characterized in Canadian HIV/HCV co-infected populations, where aboriginal populations
are over represented.
rs8099917
**TT, n (%)
GT, n (%)
GG, n (%)
Objectives:
To evaluate the association of IL-28B Single Nucleotide Polymorphisms (rs12979860, rs8099917, rs8103142) with spontaneous
HCV clearance in individuals enrolled in the Canadian Co-infection Cohort Study (CCC)
rs8103142
**TT, n (%)
CT, n (%)
CC, n (%)
To measure and compare the distribution of IL28B alleles among Canadian whites and aboriginals enrolled in the CCC
Methods
78 (79 %)
7 (7%)
14 (14%)
Cases
(n=62)
23 (37%)
12 (19%)
9 (53%)
1 (6%)
7 (41%)
6.4 per 100 person-years
(4.99-8.2)
IL28B genotypes
49 (45%)
50 (45%)
11 (10%)
43 (71%)
10 (16%)
2 (3%)
71 (63%)
32 (28%)
10 (9%)
51 (46%)
51(46%)
8 (7%)
49 (80%)
10 (16%)
2 (3%)
Study population: Only individuals who have never been treated for HCV (n=845) with at least 2 HCV RNA tests available were
included. Visits after HCV treatment initiation were censored.
• Cases less likely to be infected with HCV genotype 1 than the others.
16 due to missing ethnicity
390 for < 2 HCV RNA measures
The final study population was 436 individuals contributing 967.5 person-years of risk.
Case definition: A case was HCV-RNA negative on 2 consecutive PCR tests, while a control was HCV-RNA positive on 2
successive tests within two years of when cases occur.
Control selection and time axis: We matched on self-reported ethnicity and used incidence density sampling to select 2 controls
for each case.
Table 3: All SNPs are in Hardy-Weinberg Equilibrium in
OR (95% CI)
rs12979860 CC
3.4 (1.73, 6.69)
rs8099917 TT
6.18 (2.13, 17.96)
rs8103142 TT
3.92(1.90, 8.11)
female
1.54 (0.79, 3.02)
HCV genotype 1
0.61 (0.24, 1.58)
white and aboriginal controls (p>0.05).
o Favourable genotypes at all SNPs are linked with
higher odds of clearance
o These effects are statistically significant
o Females have 54% higher odds of clearing than males
and transgenders
o This effect was not statistically significant.
o Infection with HCV genotype 1 or 4 is linked with a
39% lower odds of clearance than genotypes 2 or 3
Table 4: Multivariate analysis of factors associated with spontaneous HCV clearance
rs12979860 CC
56 (92%)
5 (8%)
0 (0)
Table 2:
3 due to missing HCV duration
Table 3: Univariate analysis of factors associated with spontaneous HCV clearance
Adjusted OR (95% CI) 3.44 (1.70, 6.95) **
Study design: Nested case control
Setting/Source population: The Canadian Co-infection Cohort Study (CCC, n=1,020) is a prospective cohort of HIV/HCV coinfected individuals recruited from 16 centres across Canada, representing 20% of the co-infected population under care. Data and
samples are collected at visits every 6 months. Median follow-up is 2 years with an IQR of 0.7-3 years.
Of the 845 individuals, we excluded
Figure 1. IL28B Favourable Allele Frequency in Whites and Aboriginals from the CCC
• Clearance is rare, at a rate of 6.4/100 person-years (4.99-8.2/100 p-y).
• Cases more likely to be female compared to controls (37% vs. 27%).
rs8099917 TT
rs8103142 TT
6.61(2.20,19.89) **
4.19 (1.93, 9.08) **
Female
HCV genotype
1.73 (0.82, 3.65)
1.89 (0.90, 3.98)
1.80 (0.82, 3.92)
(1/4 vs. 2/3)
0.64 (0.24, 1.69)
0.68 (0.24, 1.93)
0.52 (0.20, 1.38)
Table 4:
o Because the SNPs are tightly linked, they were analysed separately.
o All IL28B genotypes ** are linked with spontaneous clearance, with effects of similar
magnitude.
o Effects of all variables are very similar as in univariate analyses
o Indicating that effect of the SNP is independent of gender and HCV genotype.
o Not enough power to detect any gender or HCV genotype specific differences in the effect of
IL28B.
• Cases are more likely to have the favourable genotype ** at all SNPs than
controls: CC at rs12979860, TT at rs8099917, and TT at rs8103142.
Limitations
 Because the study participants have been infected with HCV for a long time, impossible to say if clearance is first or repeat occurrence.
 We did not have sufficient power to detect interactions between the SNPs and gender or SNPs and HCV genotype.
 Functional assays and other studies are needed to verify the effect of the rs8103142 variant on the IL28B protein.
Conclusions
 Results indicate that, consistent with trends in other populations, the CC genotype at rs12979860 and TT genotype at rs8099917 are associated with over a two-fold increase in
spontaneous HCV clearance.
HCV duration, which is the time axis, was based on date of HCV seroconversion, if known, or on the year of first injection drug
use or blood product exposure.
 The effect is also similar with TT genotype at the functional variant rs8103142, which leads to an arginine-lysine substitution at position 70 in the IL28B protein.
Exposure: IL28B genotypes at SNPs rs12979860, rs8099917, and rs8103142 from stored plasma, analyzed as dichotomous (having
the favourable genotype vs. not)
 Being female is also associated with a higher rate of spontaneous clearance, though this effect lacked statistical significance, likely due to small sample size.
Analysis: Data were analyzed using conditional logistic regression
Software:All analyses were conducted using Stata v11.2 (College Station, TX: StataCorp LP. 2009).
 Distribution of IL28B alleles is very similar in Canadian whites and aboriginals.
 HCV genotype 1 or 4 is linked with lower clearance, though the effect did not reach statistical significance.
We acknowledge: Rhyan Pineda, Laurence Brunet, Kathleen Rollet, Claire Infante-Rivard, and the participants, co-investigators & coordinators of the HIV-HCV Canadian Cohort (CTN 222).
Funded by: