Genome-wide association study of primary sclerosing cholangitis identifies new risk loci and quantifies the genetic relationship with inflammatory bowel disease.

Primary sclerosing cholangitis (PSC) is a rare progressive disorder leading to bile duct destruction; ∼75% of patients have comorbid inflammatory bowel disease (IBD). We undertook the largest genome-wide association study of PSC (4,796 cases and 19,955 population controls) and identified four new genome-wide significant loci. The most associated SNP at one locus affects splicing and expression of UBASH3A, with the protective allele (C) predicted to cause nonstop-mediated mRNA decay and lower expression of UBASH3A. Further analyses based on common variants suggested that the genome-wide genetic correlation (rG) between PSC and ulcerative colitis (UC) (rG = 0.29) was significantly greater than that between PSC and Crohn's disease (CD) (rG = 0.04) (P = 2.55 × 10-15). UC and CD were genetically more similar to each other (rG = 0.56) than either was to PSC (P < 1.0 × 10-15). Our study represents a substantial advance in understanding of the genetics of PSC.

Anti-kelch-like 12 and anti-hexokinase 1: novel autoantibodies in primary biliary cirrhosis.

BACKGROUND & AIMS: Using high-density human recombinant protein microarrays, we identified two potential biomarkers, kelch-like 12 (KLHL12) and hexokinase-1 (HK1), in primary biliary cirrhosis (PBC). The objective of this study was to determine the diagnostic value of anti-KLHL12/HK1 autoantibodies in PBC. Initial discovery used sera from 22 patients with PBC and 62 non-PBC controls. KLHL12 and HK1 proteins were then analysed for immunoglobulin reactivity by immunoblot and enzyme-linked immunosorbent assay (ELISA) in two independent cohorts of PBC and disease/healthy control patients. METHODS: Serum samples from 100 patients with PBC and 165 non-PBC disease controls were analysed by immunoblot and samples from 366 patients with PBC, 174 disease controls, and 80 healthy donors were tested by ELISA. RESULTS: Anti-KLHL12 and anti-HK1 antibodies were each detected more frequently in PBC compared with non-PBC disease controls (P < 0.001). Not only are both markers highly specific for PBC (≥95%) but they also yielded higher sensitivity than anti-gp210 and anti-sp100 antibodies. Combining anti-HK1 and anti-KLHL12 with available markers (MIT3, gp210 and sp100), increased the diagnostic sensitivity for PBC. Most importantly, anti-KLHL12 and anti-HK1 antibodies were present in 10-35% of anti-mitochondrial antibody (AMA)-negative PBC patients and adding these two biomarkers to conventional PBC assays dramatically improved the serological sensitivity in AMA-negative PBC from 55% to 75% in immunoblot and 48.3% to 68.5% in ELISA. CONCLUSIONS: The addition of tests for highly specific anti-KLHL12 and anti-HK1 antibodies to AMA and ANA serological assays significantly improves efficacy in the clinical detection and diagnosis of PBC, especially for AMA-negative subjects.

The IRF5-TNPO3 association with systemic lupus erythematosus has two components that other autoimmune disorders variably share.

Exploiting genotyping, DNA sequencing, imputation and trans-ancestral mapping, we used Bayesian and frequentist approaches to model the IRF5-TNPO3 locus association, now implicated in two immunotherapies and seven autoimmune diseases. Specifically, in systemic lupus erythematosus (SLE), we resolved separate associations in the IRF5 promoter (all ancestries) and with an extended European haplotype. We captured 3230 IRF5-TNPO3 high-quality, common variants across 5 ethnicities in 8395 SLE cases and 7367 controls. The genetic effect from the IRF5 promoter can be explained by any one of four variants in 5.7 kb (P-valuemeta = 6 × 10(-49); OR = 1.38-1.97). The second genetic effect spanned an 85.5-kb, 24-variant haplotype that included the genes IRF5 and TNPO3 (P-valuesEU = 10(-27)-10(-32), OR = 1.7-1.81). Many variants at the IRF5 locus with previously assigned biological function are not members of either final credible set of potential causal variants identified herein. In addition to the known biologically functional variants, we demonstrated that the risk allele of rs4728142, a variant in the promoter among the lowest frequentist probability and highest Bayesian posterior probability, was correlated with IRF5 expression and differentially binds the transcription factor ZBTB3. Our analytical strategy provides a novel framework for future studies aimed at dissecting etiological genetic effects. Finally, both SLE elements of the statistical model appear to operate in Sjögren's syndrome and systemic sclerosis whereas only the IRF5-TNPO3 gene-spanning haplotype is associated with primary biliary cirrhosis, demonstrating the nuance of similarity and difference in autoimmune disease risk mechanisms at IRF5-TNPO3.

Optimising risk stratification in primary biliary cirrhosis: AST/platelet ratio index predicts outcome independent of ursodeoxycholic acid response.

BACKGROUND & AIMS: Outcomes in primary biliary cirrhosis (PBC) can be predicted by biochemical response to ursodeoxycholic acid (UDCA). Such stratification inadequately captures cirrhosis/portal hypertension, recognised factors associated with adverse events. METHODS: We evaluated a cohort of PBC patients (n=386) attending the Liver Unit in Birmingham (derivation cohort), seeking to identify risk-variables associated with transplant-free survival independent of UDCA-response. A validation cohort was provided through well-characterised patients attending the Toronto Center for Liver Diseases (n=479) and Jena University Hospital (n=150). RESULTS: On multivariate analysis, factors at diagnosis associated with liver transplant (LT)/death were patient age (HR:1.06; p<0.001), elevated bilirubin (HR:1.27; p<0.001), early-onset cirrhosis (HR:2.40; p<0.001) and baseline AST/platelet ratio index (APRI) (HR:1.95; p<0.001). At 1-year, UDCA biochemical non-response predicted poorer transplant-free survival, and additional factors (multivariate) associated with adverse outcome were age (HR:1.02; p<0.05) and 1-year APRI (HR:1.15; p<0.001). Obtaining a cut-point from our derivation cohort, APRI >0.54 at baseline was predictive of LT/death (adjusted HR: 2.40; p<0.001), and retained statistical significance when applied at 1-year (APRI-r1, adjusted HR:2.75; p<0.001) despite controlling for UDCA-response. Across both cohorts, transplant-free survival was poorer for biochemical-responders with an APRI-r1 >0.54 vs. biochemical-responders with a lower APRI-r1 (p<0.01 and p<0.001, respectively); non-responders with high APRI-r1 had the poorest outcomes (p<0.001 and p<0.001). CONCLUSION: In PBC, elevated APRI is associated with future risk of adverse events, independently and additively of UDCA-response. This cross-centre, robustly validated observation will contribute to ongoing efforts to refine existing risk-stratification tools, as well as direct focus for new therapies in patients with PBC.

Good maternal and fetal outcomes for pregnant women with primary biliary cirrhosis.

BACKGROUND & AIMS: Up to 25% of patients diagnosed with primary biliary cirrhosis (PBC) are of childbearing age. However, little is known about disease course during pregnancy. METHODS: We performed a retrospective analysis of women with PBC during pregnancy using a representative large cohort of patients attending the Liver Center at Toronto Western hospital from January 1979 through June 2009 (n = 306). Statistical analysis was performed by using R statistical software. RESULTS: We identified 32 women (50 pregnancies) who either became pregnant after a diagnosis of PBC or in whom pregnancy led to diagnosis. Liver biochemistry remained stable in most patients (70%) throughout pregnancy. However, 23 of 32 patients (72%) had a flare in biochemical disease activity post partum, which was unrelated to biochemical disease activity before conception (P = .53), or during the gestational period (P = .14). No adverse maternal events were observed during pregnancy or post partum, and only 2 of 32 of women (6%) developed progressive disease after delivery. De novo pruritus developed during pregnancy in 17 of 32 women (53%), whereas itch that existed before conception worsened for 4 patients. Fifteen of 21 women (71%) with pregnancy-related pruritus required symptom-specific therapy. Twenty-nine of 32 women (91%) had at least 1 successful live birth; adverse fetal outcome was not influenced by biochemical disease activity before conception (P = .24) or during pregnancy (P = 1.00). CONCLUSION: Pregnancy in women with PBC is frequently symptomatic but mostly uneventful. The majority of women maintain stable liver biochemistry during pregnancy, although postpartum biochemical exacerbations are common.

IgG4 related disease - a retrospective descriptive study highlighting Canadian experiences in diagnosis and management.

BACKGROUND: Appreciating the utility of published diagnostic criteria for autoimmune pancreatitis, when compared to the characteristics of patients clinically managed as having disease, informs and refines ongoing clinical practice. METHODS: Comparative retrospective descriptive evaluation of patients with autoimmune pancreatitis including dedicated radiology review. RESULTS: 66 subjects with radiographic OR clinical features of autoimmune pancreatitis were initially identifiable (Male: n = 50), with 55 confirmed for evaluation. The most common presentation included pain (67%), weight loss (65%), and jaundice (62%). Diffuse enlargement of the pancreas was evident in 38%, whilst multifocal, focal, or atrophic changes were seen in 7%, 33% and 9% respectively. 13% had no pancreatic parenchymal involvement. Peripheral rim enhancement was seen in 23 patients (42%). Where discernible, disease was a) Sclerosing pancreatitis and cholangitis, n = 21; b) Sclerosing cholangitis, n = 9; c) Sclerosing pancreatitis, n = 4; d) Sclerosing pancreatitis and cholangitis with pancreatic pseudotumour, n = 7; e) Sclerosing cholangitis with hepatic pseudotumour, n = 3; f) Sclerosing pancreatitis with pancreatic pseudotumour, n = 1. 56% of the patients had systemic manifestations and the median serum IgG4 at diagnosis was 5.12 g/L. The Korean criteria identified most patients (82%) compared to HISORt (55%) or the Japan Pancreas Society (56%). The majority (HISORt 60%; Japan Pancreas Society 55%; Korean 58%) met diagnostic criterion by radiological findings and elevated serum IgG4. Treatment and response did not differ when stratified by diagnostic criteria. CONCLUSION: Our descriptive and retrospective dataset confirms that in non-expert practice settings, autoimmune pancreatitis scoring systems with a focus on radiology and serology capture most patients who are clinically felt to have disease.

A validated clinical tool for the prediction of varices in PBC: the Newcastle Varices in PBC Score.

BACKGROUND & AIMS: Gastro-oesophageal varices (GOV) can occur in early stage primary biliary cirrhosis (PBC), making it difficult to identify the appropriate time to begin screening with oesophageo-gastro-duodenoscopy (OGD). Our aim was to develop and validate a clinical tool to predict the probability of finding GOV in PBC patients. METHODS: A cross-sectional retrospective study analysing clinical data of 330 PBC patients who underwent an OGD at the Freeman Hospital, Newcastle was used to create a predictive tool, the Newcastle Varices in PBC (NVP) Score, that was externally validated in PBC patients from Cambridge (UK) and Toronto (Canada). RESULTS: 48% of the Newcastle, 31% of the Cambridge, and 22% of the Toronto cohorts of PBC patients had GOV. Twenty-five percent (95% CI 18-32%) of the Newcastle cohort had GOV diagnosed at an index variceal bleed. Of the others, 37% (95% CI 28-46%) bled after a median of 1.5 years (IQR 3.75). Transplant-free survival was significantly better in those without GOV than in those with GOV (p<0.001), but similar in patients with GOV that bled and those that did not (p=0.1). The NVP score (%Probability)=1/[1+exp^-(9.186+0.001*alkaline phosphatase in IU-0.178*albumin in g/L-0.015*platelet × 10(9)) was validated in 2 external cohorts and was highly discriminant (AUROC 0.86). Cost consequences analyses revealed the NVP score to be as accurate as, but more economical than using either OGD directly or other risk scores for screening PBC patients. CONCLUSIONS: The NVP score is an inexpensive, non-invasive, externally validated tool that accurately predicts GOV in PBC.

Immunochip analyses identify a novel risk locus for primary biliary cirrhosis at 13q14, multiple independent associations at four established risk loci and epistasis between 1p31 and 7q32 risk variants.

To further characterize the genetic basis of primary biliary cirrhosis (PBC), we genotyped 2426 PBC patients and 5731 unaffected controls from three independent cohorts using a single nucleotide polymorphism (SNP) array (Immunochip) enriched for autoimmune disease risk loci. Meta-analysis of the genotype data sets identified a novel disease-associated locus near the TNFSF11 gene at 13q14, provided evidence for association at six additional immune-related loci not previously implicated in PBC and confirmed associations at 19 of 22 established risk loci. Results of conditional analyses also provided evidence for multiple independent association signals at four risk loci, with haplotype analyses suggesting independent SNP effects at the 2q32 and 16p13 loci, but complex haplotype driven effects at the 3q25 and 6p21 loci. By imputing classical HLA alleles from this data set, four class II alleles independently contributing to the association signal from this region were identified. Imputation of genotypes at the non-HLA loci also provided additional associations, but none with stronger effects than the genotyped variants. An epistatic interaction between the IL12RB2 risk locus at 1p31and the IRF5 risk locus at 7q32 was also identified and suggests a complementary effect of these loci in predisposing to disease. These data expand the repertoire of genes with potential roles in PBC pathogenesis that need to be explored by follow-up biological studies.

The specificity of fatigue in primary biliary cirrhosis: evaluation of a large clinic practice.

UNLABELLED: Quality of life is an important concern for patients with chronic liver disease. We sought to describe the frequency, severity, and associations of fatigue, in patients with primary biliary cirrhosis (PBC). We performed association testing between PBC-40 multidomain disease-specific quality of life responses and clinical findings. Three hundred twenty-seven patients from a single clinic with PBC (94% female, 92% AMA-positive) were evaluated. The average age was 57 years and average disease duration 7.2 years. Verbally reported fatigue was noted in 48% but present in the overwhelming majority on PBC-40 completion, with 44% having moderate or severe symptoms. Of those not complaining of fatigue clinically, 25% documented moderate or severe fatigue by questionnaire. Age had an inverse relationship with fatigue (P < 0.01), whereas body mass index (BMI) was positively associated (P < 0.01), as was the presence of pruritus (P < 0.001), sicca symptoms (P < 0.001), depression (P < 0.001), fibromyalgia (P < 0.004), and scleroderma (P < 0.05). For those with varices (P < 0.05) or cirrhosis clinically (P < 0.05), higher fatigue scores were noted, although those who initially presented with noncirrhotic disease had higher scores at the time of testing (P < 0.005). Fatigue was associated with greater use of prescription medication (P < 0.01), in particular for antipruritics (cholestyramine: P < 0.001; rifampin: P < 0.001), proton pump inhibitors (P < 0.002), beta-blockers (P < 0.02), and antidepressants (P < 0.001), whereas those taking calcium and vitamin D appeared less fatigued (P < 0.05). In a multivariate model, calcium and vitamin D use, BMI, stage of disease at diagnosis, as well as symptomatic fatigue or pruritus, were significant. Biochemical response to UDCA was not associated with lower fatigue scores. CONCLUSION: Attempts at defining the biological basis of fatigue in patients with PBC, and improving its treatment, must account for its multifactoral causes.

Variants at IRF5-TNPO3, 17q12-21 and MMEL1 are associated with primary biliary cirrhosis.

We genotyped individuals with primary biliary cirrhosis and unaffected controls for suggestive risk loci (genome-wide association P < 1 x 10(-4)) identified in a previous genome-wide association study. Combined analysis of the genome-wide association and replication datasets identified IRF5-TNPO3 (combined P = 8.66 x 10(-13)), 17q12-21 (combined P = 3.50 x 10(-13)) and MMEL1 (combined P = 3.15 x 10(-8)) as new primary biliary cirrhosis susceptibility loci. Fine-mapping studies showed that a single variant accounts for the IRF5-TNPO3 association. As these loci are implicated in other autoimmune conditions, these findings confirm genetic overlap among such diseases.

Baseline ductopenia and treatment response predict long-term histological progression in primary biliary cirrhosis.

OBJECTIVES: Laboratory and pathological predictors of future histological progression in primary biliary cirrhosis (PBC) are needed for routine practice and clinical trials. We sought to develop clinically meaningful markers for those with predominantly early disease at risk of progressive liver damage. METHODS: Patients with PBC (n=69) with a follow-up liver biopsy performed approximately 10 years after initial histological diagnosis were identified and reviewed. RESULTS: Histological progression in the stage of fibrosis observed in paired liver biopsies from the same patient was associated with the absence of biochemical response to ursodeoxycholic acid (UDCA) at 2 years: alkaline phosphatase (ALP) >1.67 × ULN (upper limit of normal) (P=0.001, odds ratio (OR) 12.14, 95% confidence interval (CI) 2.69-54.74) when defined as an increase in one stage and ALP > 1.76 × ULN (P=0.03, OR 5.07, 95% CI 1.17-21.95) when defined as an increase in two stages. Ductopenia (>50% loss), as formally evaluated through blinded biopsy review of liver tissue obtained at initial diagnosis in a subset of 34 patients, predicted histological progression (P=0.012), along with biochemical response to UDCA (P=0.002). The presence of interface hepatitis in the same biopsies did not. CONCLUSIONS: Patients with PBC who fail to show a biochemical response to UDCA or who have ductopenia on baseline biopsy progress histologically during extended follow-up. Such patients may benefit from novel treatments, with our exploratory data providing a means of identifying these individuals early in their disease.

Value of autoantibody analysis in the differential diagnosis of chronic cholestatic liver disease.

BACKGROUND & AIMS: It is challenging to diagnose patients with chronic cholestatic liver diseases when all the classic criteria are not fulfilled. We evaluated the performance of the recently developed MIT3-based enzyme-linked immunosorbent assay (ELISA), which detects antimitochondrial autoantibodies (AMAs), together with ELISAs for other autoimmune liver disease-related antibodies in patients with chronic cholestatic liver disease. METHODS: Sera from 281 patients with chronic cholestatic conditions, including primary biliary cirrhosis (PBC), primary sclerosing cholangitis, AMA-positive autoimmune hepatitis, and "undetermined cholangiopathy" were tested for the following PBC-associated autoantibodies: anti-gp210, anti-sp100, conventional anti-M2, anti-M2 (MIT3) IgG, anti-M2 (MIT3) IgA, as well as anti-centromere, anti-soluble liver antigen, and anti-chromatin. RESULTS: Of 57 patients with PBC who were AMA-negative by conventional M2 ELISA, 14 were found to be AMA-positive by the MIT3-IgG assay. Furthermore, on the basis of the data from 3 tests (MIT3-IgG, gp210, and sp100), PBC was confirmed in 20 of 57 (35%) patients diagnosed with AMA-negative PBC. We confirmed that sp100 and gp210 antibodies were detected only in patients with PBC and AMA-positive autoimmune hepatitis, whereas gp210 was detected more frequently in patients known to have had a poor outcome. Of the 11 patients with an undetermined cholangiopathy, 3 (27%) tested positive for PBC with the MIT3-IgG assay. In contrast to previous findings, anti-centromere antibodies were not found to be associated with poor outcome in PBC. CONCLUSIONS: ELISAs for AMAs and antinuclear antibodies are useful in diagnosis and prognosis of patients with features of PBC who lack conventional AMA and in patients with a cholangiopathy of undetermined etiology.

Primary biliary cirrhosis associated with HLA, IL12A, and IL12RB2 variants.

BACKGROUND: Primary biliary cirrhosis is a chronic granulomatous cholangitis, characteristically associated with antimitochondrial antibodies. Twin and family aggregation data suggest that there is a significant genetic predisposition to primary biliary cirrhosis, but the susceptibility loci are unknown. METHODS: To identify genetic loci conferring a risk for primary biliary cirrhosis, we carried out a genomewide association analysis in which DNA samples from 2072 Canadian and U.S. subjects (536 patients with primary biliary cirrhosis and 1536 controls) were genotyped for more than 300,000 single-nucleotide polymorphisms (SNPs). Sixteen of the SNPs most strongly associated with primary biliary cirrhosis were genotyped in two independent replication sets. We carried out fine-mapping studies across three loci associated with primary biliary cirrhosis. RESULTS: We found significant associations between primary biliary cirrhosis and 13 loci across the HLA class II region; the HLA-DQB1 locus (encoding the major histocompatibility complex class II, DQ beta chain 1) had the strongest association (P=1.78x10(-19); odds ratio for patients vs. controls, 1.75). Primary biliary cirrhosis was also significantly and reproducibly associated with two SNPs at the IL12A locus (encoding interleukin-12alpha), rs6441286 (P=2.42x10(-14); odds ratio, 1.54) and rs574808 (P=1.88x10(-13); odds ratio, 1.54), and one SNP at the IL12RB2 locus (encoding interleukin-12 receptor beta2), rs3790567 (P=2.76x10(-11); odds ratio, 1.51). Fine-mapping analysis showed that a five-allele haplotype in the 3' flank of IL12A was significantly associated with primary biliary cirrhosis (P=1.15x10(-34)). We found a modest genomewide association (P<5.0x10(-5)) with the risk of disease for SNPs at the STAT4 locus (encoding signal transducer and activator of transcription 4) and the CTLA4 locus (encoding cytotoxic T-lymphocyte-associated protein 4) and 10 other loci. CONCLUSIONS: Our data show significant associations between primary biliary cirrhosis and common genetic variants at the HLA class II, IL12A, and IL12RB2 loci and suggest that the interleukin-12 immunoregulatory signaling axis is relevant to the pathophysiology of primary biliary cirrhosis. (ClinicalTrials.gov number, NCT00242125.)

CTLA4/ICOS gene variants and haplotypes are associated with rheumatoid arthritis and primary biliary cirrhosis in the Canadian population.

OBJECTIVE: The co-occurrence of different autoimmune diseases in patients and their families suggests the presence of shared genetic risk factors. Two compelling candidate autoimmune disease susceptibility genes are those that encode CTLA4 and inducible costimulator (ICOS), immunoregulatory proteins. Associations of CTLA4 polymorphisms with various autoimmune diseases have been reported, but for rheumatoid arthritis (RA) and primary biliary cirrhosis (PBC), the association data are inconsistent and have largely excluded analysis of polymorphisms in the ICOS gene adjacent to CTLA4. We undertook this study to examine whether CTLA4 and ICOS influence RA and PBC susceptibility by testing CTLA4/ICOS polymorphisms for association with these diseases in Canadian subjects. METHODS: Caucasian RA patients (n = 1,140), PBC patients (n = 481), and controls (n = 1,248) were typed for 21 biallelic polymorphisms across the CTLA4/ ICOS genes using a multiplex genotyping array, and the results were analyzed using a false discovery rate method to correct for multiple testing. RESULTS: Significant associations of multiple CTLA4 and ICOS gene polymorphisms with RA and PBC were observed, with the strongest association signals for both diseases coming from a CTLA4/ICOS intergenic single-nucleotide polymorphism, rs17268364 (corrected P [P(corr)] = 6.0 x 10(-4) and P(corr) < 1.0 x 10(-4), respectively). Significant associations, which were common to both diseases, were also observed with other alleles and haplotypes across 3 linkage disequilibrium blocks within the CTLA4 gene, the intergenic region, and the ICOS gene. CONCLUSION: Our results provide evidence for RA and PBC association with the CTLA4/ICOS locus and suggest that the risk allele(s) within this region may be common to both diseases.

Gene expression profiling of early primary biliary cirrhosis: possible insights into the mechanism of action of ursodeoxycholic acid.

OBJECTIVES: Primary biliary cirrhosis (PBC) is a poorly understood disease, both in terms of its pathogenesis and the mechanism of action of its most common treatment, ursodeoxycholic acid (UDCA). We used gene expression profiling to compare liver tissue from treatment-naïve and UDCA-treated patients in order to outline some of the molecular changes associated with PBC and its treatment. PATIENTS AND EXPERIMENTAL DESIGN: Liver biopsy specimens from non-cirrhotic, treatment-naïve (n=11) patients were compared with biopsies from UDCA-treated patients (n=20) and with 10 normal, healthy female controls. Gene expression was determined using a 19K cDNA microarray. In order to determine whether the observed changes in gene expression levels were specific to PBC or generic to liver damage overall, PBC samples were also compared with chronically diseased [48 hepatitis C virus (HCV), 18 hepatitis B virus (HBV)] and acutely stressed liver tissue (25 liver biopsies taken after reperfusion of liver transplant grafts). RESULTS: We found a gene signature specific to PBC (P

Hepatic gene expression discriminates responders and nonresponders in treatment of chronic hepatitis C viral infection.

BACKGROUND & AIMS: Pegylated interferon (IFN)-alpha plus ribavirin is the most effective treatment of chronic hepatitis C but has unpleasant side effects and high costs. A large proportion of patients do not respond to therapy for reasons that are unclear. We used gene expression profiling to investigate the molecular basis for treatment failure. METHODS: Expression profiling was performed on percutaneous needle liver biopsy specimens taken before therapy. Gene expression levels were compared among 15 nonresponder, 16 responder, and 20 normal liver biopsy specimens. Differential gene expression was confirmed using real-time polymerase chain reaction. RESULTS: We identified 18 genes whose expression differed significantly between all responders and all nonresponders (P < .005). Many of these 18 genes are IFN sensitive and 2 (ISG15/USP18) are linked in a novel IFN-regulatory pathway, suggesting a possible rationale for treatment resistance. Using a number of independent classifier analyses, an 8-gene subset accurately predicted treatment response for 30 of 31 patients. The classifier analyses were applicable to patients with genotype 1 infection and were not correlated with viral load, disease activity, or fibrosis. CONCLUSIONS: Hepatic gene expression profiling identified consistent differences in patients who subsequently fail treatment with pegylated IFN-alpha plus ribavirin: up-regulation of a specific set of IFN-responsive genes predicts nonresponse to exogenous therapy. These data may be of use in predicting clinical responses to treatment.