To the Editor
With interest we read the paper by Petta et al. evaluating IL28B in 160 patients with non-alcoholic fatty liver disease (NAFLD). The authors described association of IL28B with both inflammation and fibrosis, though the association with fibrosis was not significant after adjusting for inflammation [1]. Interestingly, steatosis the hallmark of fatty liver disease was not associated with IL28B in the study by Petta et al.
We had previously described an association of IL28B rs12979860 with serum lipids and steatosis [2,3], both of which was confirmed in subsequent studies [4–6]. While steatosis was concordantly found to be associated with the less favorable IL28B genotypes (CT and TT) in patients with hepatitis C, a relation to inflammation was not reported.
Though we believed the effect of IL28B to be relatively limited to HCV, in light of the findings of Petta et al., we evaluated the role of IL28B in our cohort of NAFLD patients. Patients were recruited at two different sites, the “Duke University Outpatient Fatty Liver Clinic” and “Durham Regional Hospital”, where patients are evaluated for bariatric surgery.
195 Caucasian patients with biopsy-proven NAFLD and DNA available for genetic testing were evaluated for IL28B rs12979860.
Histologic features as defined by the Nonalcoholic Steatohepatitis Clinical Research Network [7] include steatosis grade (0–1 vs. 2–3), fibrosis stage (0–2 vs. 3–4), lobular inflammation grade (0–1 vs. 2–3), hepatocyte ballooning (0–1 vs. 2) and NAFLD activity score (NAS, <5 vs. ≥5). Frequencies and means were generated using SAS version 9.2 using procedures PROC FREQ and PROC MEANS, respectively (SAS Systems, Cary, NC). Logistic regression (PROC LOGISTIC) was utilized to determine the relationship of IL28B and histological outcomes. Based on the Petta et al. approach, a dominant genetic model was employed such that the CC genotype was compared to CT/TT genotypes. Bivariate models were tested as well as models adjusting for age, gender, diagnosis of diabetes mellitus (DMdx), body mass index (BMI) and site of recruitment (Bariatric Surgery Program or Liver Clinic).
The distribution of rs12979860 genotypes in our data set was similar to the population described by Petta et al., and typical for a Caucasian population with 48.72% of subjects homozygous for the C allele, 41.54% heterozygous, and 9.74% homozygous for the T allele. However, in contrast to Petta et al., we did not find a significant role for IL28B in relation to histological characteristics of NAFLD (see Table 1A). Results did not change substantially when the analysis was adjusted for age, gender, DMdx, BMI and recruitment site. The adjusted model, however, confirmed the expected role for BMI and diabetes mellitus in regard to histological features of NAFLD (see Table 1B)
Table 1A.
IL28B rs12979860 genotype in relation to histological features of NAFLD.
| CC (%) | CT (%) | TT (%) | p value | ||
|---|---|---|---|---|---|
| Steatosis | 0–1 (n = 94) | 51 (54.26 ) | 34 (36.17 ) | 9 (9.57 ) | 0.1955 |
| 2–3 (n = 98) | 44 (44.9 ) | 45 (45.92 ) | 9 (9.18 ) | ||
| Inflammation | 0–1 (n = 148) | 75 (50.68 ) | 60 (40.54 ) | 13 (8.78 ) | 0.1933 |
| 2–3 (n = 34) | 13 (38.24 ) | 16 (47.06 ) | 5 (14.71 ) | ||
| Ballooning | 0–1 (n = 159) | 81 (50.94 ) | 63 (39.62 ) | 15 (9.43 ) | 0.3604 |
| 2 (n = 31) | 13 (41.94 ) | 15 (48.39 ) | 3 (9.68 ) | ||
| Fibrosis | 0–2 (n = 168) | 81 (48.21 ) | 70 (41.67 ) | 17 (10.12 ) | 0.7257 |
| 3–4 (n = 27) | 14 (51.85 ) | 11 (40.74 ) | 2 (7.41 ) | ||
| NAS 0–4 | (n = 143) | 76 (53.15 ) | 53 (37.06 ) | 14 (9.79 ) | 0.0845 |
| ≥5 (n = 49) | 19 (38.78 ) | 26 (53.06 ) | 4 (8.16 ) |
Table 1B.
Odds ratios (and corresponding p values in brackets) for rs12979860 (CC vs. CT/TT) and covariates: age, sex (male vs. female), DMdx (yes vs. no), BMI, and site of recruitment.
| rs12979860 | Site | Age | Sex | DMdx | BMI | |
|---|---|---|---|---|---|---|
| Steatosis grade (2–3 vs. 0–1) | 0.659 (0.192) | 0.251 (0.0029) | 0.992 (0.63) | 0.969 (0.92) | 1.233 (0.56) | 1.012 (0.63) |
| Lobular inflammation (2–3 vs. 0–1) | 0.522 (0.132) | 0.219 (0.0148) | 0.994 (0.79) | 0.634 (0.31) | 1.991 (0.13) | 0.979 (0.5) |
| Hepatocellular ballooning (2 vs. 0–1) | 0.795 (0.636) | 0.013 (<0.0001) | 1.033 (0.19) | 0.318 (0.024) | 1.883 (0.20) | 1.099 (0.019) |
| Fibrosis stage (3–4 vs. 0–2 as in Petta et al.) | 1.260 (0.622) | 0.073 (0.0005) | 1.038 (0.12) | 1.1577 (0.76) | 3.062 (0.022) | 1.095 (0.013) |
| NAS (5+ vs. 0–4) | 0.524 (0.094) | 0.102 (<0.0001) | 0.985 (0.41) | 0.399 (0.024) | 1.643 (0.24) | 1.009 (0.75) |
Importantly, the trend for inflammation was reversed. While 64% of patients with increased inflammation in the study by Petta et al. had the CC genotype, less than 40% of patients with severe inflammation in our study had the CC genotype. Furthermore, the NAFLD Activity Score, which showed borderline significance in the adjusted model (p = 0.0937), became less significant when an additive model (CC>CT>TT) was used (p = 0.2243). Similar to the trend observed with inflammation, the NAS tended to be lower in patients with rs12979860 CC genotype, and therefore in the opposite direction to the observation by Petta et al. Thus, in summary, our data do not show a significant association of IL28B and histological features in North American Caucasian patients with NAFLD.
However, almost all of our patients were obese, while only 40% of the patients in Petta et al.’s study were obese (BMI >30). It would therefore be interesting to know if the association reported by Petta et al. change when stratifying their cohort according to BMI group, <25, 25–30, and >30 BMI.
Acknowledgments
This research was supported by the NIH Recovery Act Limited Competition for NIH Grants: Research and Research Infrastructure Grand Opportunities (1 RC2 AA019399-01; PI: AM Diehl). Cynthia D Guy and Ayako Suzuki are additional members of the DUKE go-Grant NAFLD working group.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Petta S, Grimaudo S, Cammà C, Cabibi D, Di Marco V, Licata G, et al. IL28B and PNPLA3 polymorphisms affect histological liver damage in patients with non-alcoholic fatty liver disease. J Hepatol. 2012;56:1356–1362. doi: 10.1016/j.jhep.2012.01.007. [DOI] [PubMed] [Google Scholar]
- 2.Li JH, Lao XQ, Tillmann HL, Rowell J, Patel K, Thompson A, et al. Interferon-lambda genotype and low serum low-density lipoprotein cholesterol levels in patients with chronic hepatitis C infection. Hepatology. 2010;51:1904–1911. doi: 10.1002/hep.23592. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Tillmann HL, Patel K, Muir AJ, Guy CD, Li JH, Lao XQ, et al. Beneficial IL28B genotype associated with lower frequency of hepatic steatosis in patients with chronic hepatitis C. J Hepatol. 2011;55:1195–1200. doi: 10.1016/j.jhep.2011.03.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Clark PJ, Thompson AJ, Zhu Q, Vock DM, Zhu M, Patel K, et al. The association of genetic variants with hepatic steatosis in patients with genotype 1 chronic hepatitis C infection. Dig Dis Sci. 2012;57:2213–2221. doi: 10.1007/s10620-012-2171-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Clark PJ, Thompson AJ, Zhu M, Vock DM, Zhu Q, Ge D, et al. Interleukin 28B polymorphisms are the only common genetic variants associated with low-density lipoprotein cholesterol (LDL-C) in genotype-1 chronic hepatitis C and determine the association between LDL-C and treatment response. J Viral Hepat. 2012;19:332–340. doi: 10.1111/j.1365-2893.2011.01553.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Agúndez JA, García-Martin E, Maestro ML, Cuenca F, Martínez C, Ortega L, et al. Relation of IL28B Gene Polymorphism with Biochemical and Histological Features in Hepatitis C Virus-Induced Liver Disease. PLoS One. 2012;7:e37998. doi: 10.1371/journal.pone.0037998. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Brunt EM, Kleiner DE, Wilson LA, Belt P, Neuschwander-Tetri BA NASH Clinical Research Network (CRN) Nonalcoholic fatty liver disease (NAFLD) activity score and the histopathologic diagnosis in NAFLD: distinct clinicopathologic meanings. Hepatology. 2011;53:810–820. doi: 10.1002/hep.24127. [DOI] [PMC free article] [PubMed] [Google Scholar]