The purpose of this study was to evaluate the relationship between serum filamentous (F)-actin antibody titers and severity of hepatitis present in hepatitis C virus (HCV)-infected patients. Liver biopsy samples from 18 HCV monoinfected and 20 HCV-HIV coinfected patients were graded with respect to the degree of hepatitis activity and intensity of plasma cell infiltration using MUM-1 and CD138 immunostains. Of the 38 HCV-infected patients, 6 (16%) had F-actin antibody titers in excess of 30 enzyme-linked immunosorbent assay units. We found a positive trend between serum F-actin antibody levels and the mean number of plasma cells present in the portal tracts of patients with HCV infection (r = 0.31; P = .06) and a significant association between these factors in HCV-HIV coinfected patients (r = 0.64; P = .002). Our data suggest that elevated serum F-actin antibody titers are commonly encountered in HCV-infected patients and may reflect more active inflammation in liver biopsy samples, similar to autoimmune hepatitis.
Hepatitis C virus (HCV) is a positive-stranded RNA virus of the Flaviviridae family that causes chronic liver disease in most of the estimated 170 million people who are infected worldwide.1 Options for management of chronic HCV-related hepatitis are limited. Most protocols use some form of interferon-based therapy, which has direct antiviral and immunomodulatory actions.2 Many HCV-infected patients in the United States are coinfected with HIV, and in this setting, chronic HCV-related liver injury may be confounded by a number of agents, including concomitant drug toxicity due to antiretroviral therapy or other medications, and HIV-related parenchymal injury. Up to two thirds of patients with chronic HCV infection may also have elevated serum autoantibody titers, particularly anti–smooth muscle antibodies (SMAs) and antinuclear antibodies (ANAs), thereby simulating the clinical features of patients with immune-mediated liver injury.3–9 Smooth muscle autoantibodies are frequently detected among patients with autoimmune hepatitis, and elevation of the filamentous (F)-actin component of SMA is a specific (98%) and sensitive (100%) marker for this disease.3,10 However, the biologic importance of serum autoantibody titers in the setting of HCV infection has not been fully evaluated in a systematic manner.
We have noted that some patients with HCV monoinfection or HCV-HIV coinfection share overlapping clinical and pathologic features with patients with autoimmune hepatitis in that they have elevated serum autoantibody titers in conjunction with plasma cell–rich inflammation in liver biopsy samples. We hypothesized that elevated serum F-actin antibody titers may represent a marker of severe necroinflammatory activity in HCV-infected patients, thereby mimicking the clinicopathologic features of autoimmune hepatitis. Thus, we assessed the severity of chronic hepatitis necroinflammatory activity in a series of liver biopsy samples obtained from HCV monoinfected and HCV-HIV coinfected patients and evaluated the extent of portal plasmacytosis using MUM-1 and CD138 immunohistochemical stains to detect plasma cells. These results were correlated with serum F-actin antibody levels at the time of liver biopsy to evaluate the relationship, if any, between serum autoantibody titers and hepatitis activity in HCV monoinfected and HCV-HIV coinfected patients.
Materials and Methods
We retrospectively identified a series of 38 cases, including 18 patients with HCV monoinfection and 20 with HCV and HIV coinfection, who underwent liver biopsy performed at New York Presbyterian Hospital, New York, NY, between 2005 and 2007. All patients had detectable serum HCV RNA by qualitative assay (HCV TMA, Bayer Diagnostics, Berkeley, CA) and underwent serologic assessment for the presence of F-actin antibodies using the QUANTA Lite Actin IgG enzyme-linked immunosorbent assay (ELISA) (INOVA Diagnostics, San Diego, CA; performed at ARUP Laboratories, Salt Lake City, UT). For both groups, a serum F-actin antibody level in excess of 30 ELISA units was considered elevated, as recommended by the reference laboratory. This value was based on results obtained following assessment of 150 randomly selected normal serum samples evaluated by INOVA Diagnostics. In that study, the authors determined the mean value of their analyzed samples to be 7.3 ELISA units. They found the standard deviation and coefficient of variation for negative samples (mean, 14 ELISA units) to be 1.3 ELISA units and 9.7%, respectively, and 4.5 ELISA units and 4.1% for samples with F-actin antibody levels of more than 30 ELISA units (mean, 110 ELISA units).11
All patients were evaluated for the presence of HIV infection using an ELISA and confirmatory Western blot analysis. The data for patients who had undergone prior liver transplantation and patients receiving interferon-based therapy for HCV at the time of the liver biopsy were excluded from the study. Data regarding the presence or absence of other serum autoantibodies, patient age, and sex, were obtained from the patients’ medical records. Permission for the study was obtained from the institutional review board at Weill Cornell Medical College, New York, NY.
Pathologic Evaluation and Immunohistochemical Studies
H&E-stained slides were prepared from routinely processed liver biopsy specimens fixed in 10% buffered formalin. All of the cases were examined in a blinded manner and evaluated for portal and lobular disease activity using a 5-tiered scoring system Table 1.12 A Masson trichrome stain was performed for each case, and the extent of fibrosis was staged as follows: stage 0, no fibrosis; stage 1, fibrous expansion of portal tracts; stage 2, fibrous expansion of portal tracts with periportal (septal) fibrosis and limited or no bridging fibrosis; stage 3, well-developed bridging fibrosis with architectural distortion and early nodule formation; and stage 4, the presence of cirrhosis.12
The degree of plasmacytosis within portal tracts was evaluated by morphologic analysis and nuclear immunohistochemical staining for MUM-1 (MUM1p, dilution 1:100; DAKO, Carpinteria, CA) and cytoplasmic staining for CD138 (CD138 B-A38, dilution 1:25; Serotec, Raleigh, NC). Immunohistochemical stains were performed using standard techniques. The number of plasma cells present in each of the 3 most severely inflamed medium-sized portal tracts was recorded, and the mean was calculated for each case. Necroinflammatory disease activity within the lobules, grade of portal inflammation, degree of plasmacytosis, and stage of fibrosis were correlated with serum F-actin antibody titers, as described in the next section. Of 38 study patients, 12 had undergone prior liver biopsy. Histologic sections (H&E and Masson trichrome stains) of these cases were reviewed and compared with the current samples to assess for interval change.
The correlations between clinical features, serologic F-actin antibody levels, and pathologic findings present in liver biopsy samples were examined. The Pearson product moment method was used to evaluate the correlation between 2 continuous variables. The Spearman rank correlation method was used to examine the correlation between continuous and ordinal variables. Correlations associated with a P value of less than .05 were considered significantly different from 0. Correlation (r) with an absolute value below 0.3 was considered weak, between 0.3 and 0.5 was considered moderate, and greater than 0.5 was considered strong.
Clinical and Laboratory Features
The clinical and pathologic features of the study group are summarized in Table 2. The study group included 26 men and 12 women who ranged in age from 30 to 70 years (mean, 49 years; median, 50 years). The HCV monoinfected group included 15 men and 3 women (mean, 51 years; median, 48 years; range, 30–70 years), and the HCV-HIV coinfected group included 11 men and 9 women (mean, 47 years; median, 51 years; range, 32–59 years).
The presence of ANAs was evaluated in 16 of 18 HCV monoinfected and 17 of 20 HCV-HIV coinfected patients. In 3 (17%) of 18 monoinfected patients, ANA titers were elevated, including 1 with modest elevation (1:80) and 2 with substantial autoantibody titers (1:160 and 1:320). Notably, the patient with an ANA titer of 1:320 also had an F-actin antibody level of 58 ELISA units. Of 20 HCV-HIV coinfected patients, 2 (10%) had mildly elevated ANA titers (1:80 in both cases), including 1 with an F-actin antibody level of 36 ELISA units. Serum F-actin antibody titers ranged from 1 to 71 ELISA units among the 38 study patients but were higher in HCV monoinfected patients (mean, 23 ELISA units; median, 18 ELISA units; range, 6–71 ELISA units) compared with patients with HCV-HIV coinfection (mean, 16 ELISA units; median, 15 ELISA units; range, 1–36 ELISA units), although the difference was not significant. Serum F-actin titers in excess of 30 ELISA units were found in 6 (16%) of 38 patients, including 5 HCV monoinfected patients (28%) and 1 HCV-HIV coinfected patient (5%).
Mild elevations in aspartate and alanine aminotransferase levels were noted among monoinfected and coinfected patients. Patients with HCV monoinfection generally had lower hepatitis C viral loads (mean, 2.3 million IU/mL; median, 1.4 million IU/mL; range, 3,000-11.2 million IU/mL) compared with HCV-HIV coinfected people (mean, 5.4 million IU/mL; median, 4.2 million IU/mL; range, 52,000-24.3 million IU/mL), although the difference was not significant. The mean HIV copy number was 11,600 copies/mL (median, 158; range, 0–76,600) among coinfected patients. Three HCV-HIV coinfected patients had fewer than 200 CD4+ cells/μL (mean, 381/μL; median, 347/μL; range, 71–624/μL). We did not detect any relationship between HCV or HIV viral load and serum F-actin antibody levels in monoinfected or coinfected patients, nor did we identify a relationship between serum F-actin antibody titers and the absolute number of CD4+ cells/μL.
The number of evaluable portal tracts present within liver biopsy samples ranged from 3 to 19 (mean, 8 portal tracts per specimen). Most samples (∼80%) from both groups displayed minimal to mild inflammation in portal tracts and lobules Image 1A, although 16% (6/38) of cases showed moderate inflammation within the portal tracts in combination with fairly extensive interface hepatitis Image 1C. Plasma cells were detected in liver biopsy samples from all patients, and the mean number present in portal tracts ranged from 5 to 58 cells per portal tract Image 1B and Image 1D. There were no significant differences between HCV monoinfected and HCV-HIV coinfected patients with respect to the intensity of hepatic inflammation or the number of plasma cells present in liver biopsy samples. There was a moderate positive correlation between the mean number of plasma cells in the portal tracts and increasing grade of lobular inflammation among patients with HCV-HIV coinfection (r = 0.48; P = .03). There was no observed association between number of plasma cells present in portal tracts and fibrosis stage, sex, or age in either patient group.
A, A liver biopsy sample from a hepatitis C virus (HCV) monoinfected patient shows mild portal inflammation and scattered lobular inflammatory cells (H&E, ×200). B, A MUM-1 immunostain highlights plasma cells within the portal tract (×200). C, In contrast, a sample obtained from an HCV-HIV coinfected patient shows expansion of the portal area by a dense mononuclear cell–rich inflammatory infiltrate and extensive interface hepatitis (H&E, ×200). D, An immunostain directed against MUM-1 demonstrates numerous plasma cells within the inflammatory infiltrate (×200).
Relationship Between Serum F-Actin Levels and Severity of Hepatitis
We identified a positive correlation between increased serum F-actin antibody levels and the mean number of plasma cells present in hepatic portal tracts (r = 0.31; P = .06) among monoinfected and coinfected patients with chronic HCV infection, and the relationship was much stronger among HCV-HIV coinfected patients (r = 0.64; P =.002) Figure 1. Coinfected patients with elevated serum F-actin antibody levels (>30 ELISA units) also showed a trend toward increased necroinflammatory activity within hepatic lobules (r = 0.45; P = .05) compared with patients with normal or near normal serum F-actin antibody titers. We did not detect any relationship between F-actin antibody levels and fibrosis stage, sex, or patient age.
Assessment of Interval Progression of Disease
None of the patients in the study have undergone follow-up biopsy to assess for disease progression. However, liver biopsy samples had been obtained from 12 patients (32%) 3 to 6 years before the study, including 2 with HCV monoinfection and 10 with HCV-HIV coinfection. Histologic sections from these cases were obtained and compared with the current samples. All 12 cases showed mild inflammation in portal tracts and occasional necroinflammatory foci in lobules, similar to recent samples from each patient. In 6 patients, progression of fibrosis was identified: 5 patients had stage 1/4 fibrosis that progressed to stage 2/4, and in 1 patient with stage 2/4 fibrosis, bridging fibrosis (stage 3/4) developed 3 years after the initial biopsy sample. Six patients had stable liver disease with no increase in scarring. Plasma cells were more numerous in the portal tracts of liver biopsy samples from the 6 patients who had progressive fibrosis (mean, 32 plasma cells; range, 19–44 plasma cells) compared with patients who did not (mean, 24 plasma cells; range, 9–51 plasma cells). One patient with progressive fibrosis had a serum F-actin level of more than 30 ELISA units, but there were no significant differences with respect to F-actin titers between patients with progressive fibrosis (mean, 20 ELISA units) and patients with stable liver disease (mean, 20 ELISA units).
In this study, we evaluated the relationship between serum F-actin antibody titers and the degree of inflammatory activity and intensity of plasma cell infiltration in portal tracts of liver biopsy samples from patients with HCV monoinfection and HCV-HIV coinfection. We found that elevated serum F-actin antibody levels in excess of 30 ELISA units occurred with an overall frequency of 16% among patients with HCV infection and were more common among HCV monoinfected patients (28%). Although most patients had liver biopsy samples that showed minimal to mild inflammation and early-stage fibrosis, severity of portal plasmacytosis increased with progressive lobular inflammation among HCV-HIV coinfected patients (P = .03). We also identified a positive trend between serum F-actin antibody levels and the mean number of plasma cells present in portal tracts of liver biopsy samples from HCV-infected patients (P = .06), particularly among HCV-HIV coinfected patients (P = .002). Therefore, we conclude that elevated titers of serum F-actin antibodies are commonly encountered in HCV-infected patients and may reflect more active inflammation in liver biopsy samples from these patients, similar to patients with autoimmune hepatitis.
Relationship between serum F-actin antibody levels and the mean number of plasma cells in hepatic portal tracts of patients with hepatitis C virus (HCV) monoinfection and HCV-HIV coinfection. ELISA, enzyme-linked immunosorbent assay.
Autoimmune hepatitis is a progressive, chronic liver disease that is probably triggered by exposure to environmental factors, such as viruses or toxins, in genetically susceptible patients.13 The precise mechanisms of liver cell injury are unknown, but cell-mediated and antibody-dependent pathways are believed to be involved.14 The diagnosis is based on characteristic histologic features in liver biopsy samples that occur in combination with specific clinical and laboratory findings.
Liver biopsy samples from untreated patients often display plasma cell–rich mononuclear cell infiltrates centered on portal tracts and interface hepatitis with or without lobular injury.15 The patients often have elevated serum globulin and circulating autoantibody levels, particularly ANA and SMA, which typically lack disease specificity and tend to fluctuate but may be useful in the subclassification of disease.16 Of these, SMA is considered the most sensitive marker of autoimmune hepatitis, and antibodies that react with the F-actin component of SMA are typically found in patients with type 1 autoimmune hepatitis.10,17–19 In fact, F-actin autoantibody detection by ELISA is a highly sensitive (100%) and specific (98%) test for the presence of autoimmune hepatitis compared with standard SMA immunofluorescence testing, which shows a lower sensitivity (34%) but relatively high (98%) specificity.3,10
The presence of serum F-actin antibodies is also prognostically significant. Patients with elevated antibody titers may have a poorer response to corticosteroid therapy compared with patients who lack these antibodies.19
Emerging evidence suggests that some patients with chronic HCV infection have elevated autoantibody levels, including antibodies to ANA, SMA, and F-actin, presumably reflecting an element of molecular mimicry due to immunologic cross-reactivity between HCV antigens and human nuclear and smooth muscle antigens.3–9 Granito et al3 evaluated 51 patients with HCV infection and found that 49% had detectable antibodies to SMA. Of the HCV-infected patients, 30% had elevated F-actin antibody titers, which were in excess of 30 ELISA units in 10% of patients.3 In another study of the 34 patients with chronic viral hepatitis, Frenzel et al10 noted serum F-actin antibody titers greater than 30 ELISA units in 9% of patients, similar to the current study, in which we found 16% of patients with HCV infection to have elevated serum F-actin antibody titers. Most recently, Chrétien et al9 evaluated 186 treatment-naive patients with chronic HCV infection and found that 40% had detectable non–organ-specific autoantibodies, including SMA (15%), ANA (32%), antimitochondrial antibodies (0.5%), and anti-LKM1 antibodies (0.05%), although none of the patients with antibodies to SMA showed F-actin specificity by immunodot assay.9 It is important to note that several groups have noted a significant correlation between the presence of autoantibodies and both hepatic inflammation and fibrosis, increased serum globulin and IgG levels, and abnormal liver function tests, suggesting that elevated serum autoantibody titers may be a surrogate marker of hepatic injury in HCV-infected patients.6,9
Molecular mimicry may underlie the development of autoantibodies and autoimmune disease in patients with HIV infection. Massabki et al20 prospectively evaluated 100 HIV+ patients and 130 HIV– control subjects for the presence of serum autoantibodies and the development of autoimmune diseases. They found that 74% of HIV+ patients had detectable serum autoantibodies, the most frequent of which were anti–denatured DNA and anticardiolipin antibodies, compared with 31% of healthy control subjects. It is noteworthy that none of the patients in the study or the control group had elevated SMA titers.20 HIV-infected patients are at highest risk for the development of autoimmunity at 2 points during the evolution of HIV-related disease: acute HIV infection and following restoration of immune competence with antiretroviral therapy.21
It is important to note that patients with HIV infection may also be at increased risk for the development of immune-mediated liver injury. The combined results of 2 recent reports described 7 HIV-infected patients with autoimmune hepatitis, all of whom had liver biopsies that displayed a variable degree of interface hepatitis and lymphoplasmacytosis.22,23 Of the 7 patients, 4 had elevated ANA autoantibody titers, 5 had increased serum immunoglobulin levels, and 3 had elevated SMA antibody titers. Serum F-actin antibody levels were elevated in 3 patients. Antiretroviral therapy failed to ameliorate hepatitis in any of the patients, but corticosteroid therapy resulted in normalization of liver function test results in 5.22,23
Autoimmune hepatitis may also occasionally develop among HCV-HIV coinfected patients, in whom remission of autoimmune hepatitis may be attained by using immunosuppressive therapy, although management may be complicated by opportunistic infections and increasing HCV load.23,24 Overall, available data indicate that, although mildly elevated SMA antibody titers are common among patients with HIV infection, markedly elevated autoantibody titers are infrequently encountered. Indeed, we found that only 1 of 20 HCV-HIV coinfected patients had antibodies to the F-actin component of SMA in excess of 30 ELISA units.
The clinical importance of elevated serum F-actin antibody titers in patients with chronic hepatitis and, particularly, chronic HCV infection has been incompletely studied. In the previously cited report, Granito et al3 evaluated serum autoantibody levels in 78 untreated patients with type 1 autoimmune hepatitis and 160 control subjects, including 51 patients with HCV infection. They found that serum F-actin antibody levels exceeding 30 ELISA units were associated with higher γ-globulin and IgG levels in patients with autoimmune hepatitis but did not correlate with age, sex, inflammatory activity, fibrosis stage, or values obtained from liver function tests (aspartate aminotransferase, alanine aminotransferase, albumin, bilirubin, alkaline phosphatase, and γ-glutamyltranspeptidase). Unfortunately, the authors of the study did not report the relationship, if any, between F-actin antibody levels and any of the aforementioned parameters among HCV-infected patients.3 In our study, we detected a trend toward an association between increasing serum F-actin antibody levels and portal plasmacytosis among HCV-infected patients and a significant relationship between serum F-actin antibody levels and both portal plasmacytosis and lobular necroinflammatory activity among HCV-HIV coinfected patients. Thus, our data suggest that F-actin autoantibody titers may reflect disease activity in liver biopsy specimens from affected patients.
Our data indicate that a substantial number of patients with chronic HCV infection may have or develop elevated F-actin antibody titers, which are frequently associated with mildly increased liver function test results. Liver biopsy may reveal increased inflammatory activity in combination with numerous plasma cells within the portal tracts and parenchyma, thereby mimicking the clinical and histologic features of autoimmune hepatitis. In fact, the degree to which serum F-actin antibody levels are increased may reflect concomitant hepatic inflammation in some patients. Clinicians and pathologists should be aware of this potential pitfall that may be encountered during the evaluation of patients with chronic HCV infection. Further studies to determine the optimal management of HCV-infected patients with features of autoimmune hepatitis are warranted.
Upon completion of this activity you will be able to:
describe the setting in hepatitis C virus (HCV) when elevated F-actin levels may be present.
list the most common autoantibodies found in patients with HCV infection.
describe the clinical and pathologic features of autoimmune hepatitis.
recognize potential overlap between autoimmune hepatitis and chronic HCV infection with or without HIV coinfection in liver biopsies.
The ASCP is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The ASCP designates this educational activity for a maximum of 1 AMA PRA Category 1 Credit ™ per article. This activity qualifies as an American Board of Pathology Maintenance of Certification Part II Self-Assessment Module.
The authors of this article and the planning committee members and staff have no relevant financial relationships with commercial interests to disclose.