| |
Influence of ethnicity in the outcome of hepatitis C virus infection and cellular immune response
|
| |
| |
Indeed, 35% of (AA) African American patients responded to at least one HCV antigen compared with only 14% of CA patients. HCV-specific CD4 T-cell response in AA patients also was more multispecific and more frequently directed against the NS HCV antigens..... T-cell response to core in the recovered AA subjects was particularly poor compared with their response to the NS antigens...., among all AA subjects, HCV clearance was associated with a significant increase in CD4 T-cell response to the NS antigens but not core Such differential responsiveness was not apparent among the recovered CA subjects who showed augmented responsiveness to all HCV antigens including coreÉ.. suggesting an unexpected ethnic difference in the immunologic requirement for HCV clearance... despite an apparently greater HCV-specific CD4 T-cell response in AA patients, our results suggest that most of these T cells are functionally impaired.
Hepatitis C virus (HCV) infection results in chronicity in the majority of patients and may progress to cirrhosis and hepatocellular carcinoma over many years of infection. Both host and viral factors contribute to the clinical outcome of HCV infection. Among these, the vigor, breadth, and endurance of HCV-specific T-cell responses have been identified as key determinants in viral clearance and disease resolution during acute hepatitis C. In most patients, however, HCV persists in association with a weak or narrowly focused virus-specific T-cell response that cannot clear the virus, but may nonetheless contribute to clinical outcome and treatment responsiveness.
The race or ethnic origin of the patient also may contribute to the clinical outcome. Indeed, a number of recent studies have reported differences in natural history and treatment responsiveness between HCV-infected African Americans (AA) and Caucasian Americans (CA). In particular, AA ethnicity has been associated with a higher propensity for chronicity, liver cancer, and antiviral treatment resistance, although paradoxically, with milder biochemical and histologic liver disease relative to CAs. Thus, host ethnicity is becoming an important consideration in prognostication and management of HCV-infected patients. The underlying mechanisms for these clinical observations are not yet known. One hypothesis is that they are based on the cellular immune response, an important determinant in the outcome of viral infections.
To test the hypothesis that host ethnicity contributes to clinical outcome and antiviral T-cell response, we compared the CD4 T-cell response with multiple HCV antigens in AA and CA subjects with either persistent HCV infection or spontaneous HCV clearance. Our results showed a previously unidentified differential immune responsiveness to HCV based on ethnicity, and suggest that greater HCV persistence and interferon (IFN) resistance reported among AA patients is not caused by absence of HCV-specific T-cell response.
This study was performed to examine the immunologic basis for the apparent ethnic difference in clinical outcome of hepatitis C virus (HCV) infection
between African Americans (AA) and Caucasian Americans (CA). To this end, we recruited 99 chronically HCV-infected and 31 spontaneously
HCV-cleared subjects for clinical, virologic, and immunologic analysis. In particular, CD4-proliferative T-cell response to genotype 1-derived HCV
antigens (core, NS3-NS5) was examined in 82 patients chronically infected with genotype 1 (54 AA, 28 CA) and in all HCV-cleared subjects (14 AA, 17
CA). HCV-specific Th1 response also was examined in 52 chronic and 13 recovered subjects. Our results showed that HCV clearance was associated
with a vigorous HCV-specific Th1 response irrespective of ethnic origin. Although the HCV-specific CD4 T-cell response clearly was weaker during
chronic infection, AA ethnicity in this setting was associated with a significantly greater CD4-proliferative T-cell response to HCV, particularly to the nonstructural antigens (22% AA vs. 0% CA, P = .007) as well as better clinical parameters of liver disease. Interestingly, most HCV-specific CD4 T-cell proliferative responses in AA patients were unaccompanied by concurrent interferon (IFN-) production, suggesting a dysregulated virus-specific, CD4 T-cell effector function during chronic HCV infection. In conclusion, our results suggest that host ethnicity does influence the clinical outcome and antiviral T-cell response during HCV infection. AA ethnicity is associated with a more robust antiviral CD4 T-cell response than CA ethnicity, although these T cells are limited in direct virus or disease control due to their dysfunctional nature. (HEPATOLOGY 2003;37:590-599.)
Subjects
Subjects were recruited through the outpatient clinics at the Philadelphia Veteran's Affairs Medical Center (PVAMC), University of Pennsylvania, Temple
University, and the Department of Transfusion Medicine at the National Institutes of Health according to protocols approved by the institutional review committees. Ninety-nine HCV antibody and RNA-positive chronic patients (98% men) were recruited between 1999 and 2001 (mostly from PVAMC) based on clinical diagnosis of chronic HCV infection.
Thirty-one healthy, recovered, HCV-seropositive but HCV-RNA negative persons (87% men) with presumed spontaneous HCV clearance (without IFN therapy) and normal liver function test results also were enrolled after at least 2 negative reverse-transcription polymerase chain reactions for HCV viremia. Because of their rarity, recovered subjects were recruited from both PVAMC (n = 19) and other sites (n = 12). Twenty-three HCV-seronegative and RNA-negative normal controls with normal liver function and no HCV exposure also were recruited. Exclusion criteria included human immunodeficiency virus or hepatitis B virus co-infection, history of antiviral or immunosuppressive therapy, autoimmune hepatitis, primary biliary cirrhosis, and conditions precluding research blood donation (e.g., anemia, severe medical illness).
All subjects were assessed for clinical parameters of liver disease including alanine aminotransferase (ALT) activity, total bilirubin, albumin, prothrombin time (international normalized ratio [INR]), and platelet count. ALT values also were expressed as an ALT ratio (ALT divided by upper limit of normal),
compensating for variations in normal range. History of excessive alcohol use was documented through careful chart review including progress notes
documenting drinking as a medical concern, positive CAGE questionnaires, patients' own admission of their alcoholism, or recovery from heavy drinking or history of treatment for alcoholism. Ethnic background was determined through history and physical examination. The chronic patients were assessed for HCV viremia by Roche COBAS qualitative or quantitative reverse-transcription polymerase chain reaction (Roche Diagnostics, Branchburg, NJ); HCV genotype was assessed by INNOLIPA (Innogenetics, Gent, Belgium). In recovered subjects, lack of HCV viremia was confirmed by Roche COBAS qualitative polymerase chain reaction and HCV serotype by the third-generation HCV serotyping assay (kindly performed by Drs. David Parker and Lara Sanders, Abbott Laboratory, London, United Kingdom).
Clinical liver function parameters are better preserved among AA than CA patients during chronic HCV infection
We began by examining the clinical and virologic parameters relative to ethnicity in patients with chronic HCV infection. The chronic group consisted of 60 AA and 39 CA, predominantly male (98%), subjects between the ages of 40 to 60 years, reflecting the regional patient demographics. Both ethnic groups displayed relatively compensated liver function, although they differed in a number of areas. For example, the AA patients displayed a
significantly lower mean ALT activity (61 vs. 105 IU/L, P = .017) and ALT ratio (AA 1.1 vs. CA 1.8, P = .033) compared with the CA patients. Also, the AA
patients had significantly lower total bilirubin and INR but higher platelet count compared with the CA patients. Although the mean values for these
parameters were not particularly abnormal for either group, their differences were highly significant statistically and suggested a worse clinical outcome for the CA patients in our cohort. Significant thrombocytopenia (platelet count <130) was more frequent in CA than in AA patients (21% vs. 3%, P = .006) as was clinically evident cirrhosis (e.g., albumin <3.5, total bilirubin > 2.0, INR >1.3, or platelet count <130,000). Accordingly, histologic cirrhosis was slightly more frequent in CA (38%, 3 of 8) than in AA (14%, 1 of 7) patients, although the number of biopsies was too low for statistical comparison. Notably, clinical cirrhosis was more frequent among patients with abnormal ALT activity than those with normal ALT activity (22% vs. 4%,
P = .010). This trend was apparent in both ethnic groups and statistically significant for the AA subjects (14% vs. 0%, P = .027).
As for other host or viral factors, the 2 ethnic groups differed in their HCV genotype distribution: 90% of AA had genotype 1 vs 72% for CA (p=0.019).
HCV-specific CD4 T-cell response is significantly greater and more broad among AA than CA patients with chronic HCV infection
We examined the influence of ethnicity on the HCV-specific CD4 T-cell response by using 5 recombinant HCV proteins spanning almost two thirds of the HCV polyprotein and known to be immunogenic in HCV-infected patients. Because these antigens were genotype 1 derived, the immunologic comparison was limited to the 82 patients with genotype 1 infection. We also studied 31 recovered subjects (positive controls for HCV clearance) and 23 normal HCV-seronegative individuals (negative controls). HCV-specific CD4 T-cell response was significantly greater (particularly for the nonstructural or NS antigens) among the recovered subjects compared with the chronic patients and normal controls.
Indeed, 35% of AA patients responded to at least one HCV antigen compared with only 14% of CA patients (P = .046). HCV-specific CD4 T-cell response in AA patients also was more multispecific (17% AA vs. 0% CA, P = .022) and more frequently directed against the NS HCV antigens (22% AA vs. 0% CA, P = .007). However, the ethnic groups did not differ significantly in their response to control tetanus toxoid, PHA, and Candida.
HCV clearance in AA subjects is associated with a vigorous and multispecific CD4 T-cell response to the HCV NS but not core antigen
We then asked if ethnicity also influences the antiviral T-cell response during successful HCV clearance. HCV-specific T-cell response was significantly greater and broader in the recovered subjects compared with chronically infected patients, irrespective of ethnicity, sex, or recruitment site (e.g., VA or non-VA). Notably, the recovered and chronic subjects did not differ significantly in their response to control antigens including tetanus toxoid. The response to NS antigens tended to be greater in magnitude than the response to HCV core in both ethnic groups. However, T-cell response to core in the recovered AA subjects was particularly poor compared with their response to the NS antigens (21% vs. 86%, P = .0018). Furthermore, among all AA subjects, HCV clearance was associated with a significant increase in CD4 T-cell response to the NS antigens (86% recovered vs. 22% chronic, P < .0001) but not core (21% recovered vs. 26% chronic, P = NS). Such differential responsiveness was not apparent among the recovered CA subjects who showed augmented responsiveness to all HCV antigens including core. Thus, HCV clearance in AA subjects was associated with CD4 T-cell response to HCV NS but not core antigen, suggesting an unexpected ethnic difference in the immunologic requirement for HCV clearance.
HCV-specific type 1 T-cell response is markedly weaker during chronic HCV infection compared with recovery, irrespective of ethnic origin
HCV-specific CD4 Th1 IFN-y response to 3 HCV antigens (core, NS3-4, and NS5) was quantified ex vivo in 19 recovered, 52 chronic, and 13 normal control
subjects using IFN-y elispot assay. The combined Th1 response to these antigens in the recovered subjects was approximately 500 per million circulating PBMCs for both ethnic groups, equivalent to 0.05% of circulating lymphocytes.
Similar to the pattern in CD4 T-cell proliferation, recovered AA subjects showed a vigorous Th1 response to the NS antigens (>200/million) but not HCV core (12/million). The combined HCV-specific Th1 response in chronic patients was lower by almost 10-fold compared with recovered subjects and no different from the normal controls. Chronic patients responded poorly to the NS antigens in both ethnic groups (<10/million PBMC), although the AA patients showed slightly greater Th1 response to HCV core (54 AA vs. 26 CA per million PBMCs, P = .04).
CD4 T-cell response during chronic HCV infection is dysfunctional in coordinated HCV-specific proliferation and IFN- production
The weak Th1 response to the NS antigens in the chronic AA patients was surprising given their CD4 T-cell proliferative responses to the same antigens. In fact, for the NS3-4 antigen (the most immunogenic NS antigen), none of the 5 positive proliferative responses in the chronic AA patients was accompanied by concurrent Th1 response, whereas 9 of 14 proliferative responses among the recovered AA patients showed concurrent Th1 response (0% vs. 64%, P = 0.03). Furthermore, when the proliferative and Th1 responses to all 3 HCV antigens were assessed together, we found that only 25% of the positive proliferative responses in chronic AA patients were associated with an efficient IFN- response, compared with 64% in the recovered subjects (P = .017).
Similar comparison was not possible for the CA patients who showed minimal proliferative response. However, the disparity in HCV-specific T-cell proliferation and Th1 response persisted when both ethnicities were combined. Conversely, an efficient HCV-specific Th1 response was not associated often with a concurrent proliferative response among the chronic patients compared with recovered subjects (data not shown), consistent with a dyscoordinated HCV-specific T-cell proliferation and IFN- production during chronic HCV infection. Thus, despite an apparently greater HCV-specific CD4 T-cell response in AA patients, our results suggest that most of these T cells are functionally impaired.
Discussion
Clinical outcome of a viral infection is determined by the interplay between host, viral, and environmental factors. One important component of host response is the cellular immune response, including CD4 and CD8 T cells. Although CD8 T cells are the primary effector cells directly eliminating the virus-infected cells, CD4 T cells play a central regulatory role, providing signals needed to prime and maintain the antiviral CD8 T cells as well as B-cell antibody production. The significance of antiviral CD4 T-cell response in HCV infection is apparent in the sustained vigorous and broad CD4 T-cell response in patients who clear HCV spontaneously and the weak or focused response in patients who cannot. During chronic HCV infection, the apparently
ineffective T cells may nonetheless contribute to the clinical outcome, similar to human immunodeficiency or murine lymphocytic choriomeningitis virus infection. For example, global CD4 T-cell defects in human immunodeficiency virus-infected individuals or immunosuppression in liver transplant recipients is associated with accelerated HCV-associated liver disease progression and mortality as well as high HCV titer.29-33 Furthermore, HCV-specific CD4 T-cell response has been associated with better clinical outcome of liver disease and IFN treatment response.
Based on recently reported ethnic differences in clinical outcome of HCV infection, we examined the influence of ethnicity on HCV-specific CD4 T-cell
response and clinical outcome in both chronically HCV-infected and spontaneously recovered patients. Among our cohort with chronic HCV infection, AA ethnicity was associated with a more benign clinical outcome than CA ethnicity, consistent with the study by Wiley et al.,14 who reported milder histologic and biochemical liver disease in AA than non-AA patients with chronic hepatitis C. The differences in laboratory parameters were specific to chronic HCV infection because they were absent in recovered and normal subjects. They were not likely to be caused by alcohol because both groups showed equally high prevalence of heavy drinking, although a more detailed quantitation of alcohol intake is needed to specifically address this question. Furthermore, they were not caused by differences in gender distribution between the ethnic groups because both groups were equally male predominant. Finally, these differences were not caused by genotype variations because the trends persisted in genotype 1-infected patients.
Relevant to our clinical findings, the ethnic groups differed in their CD4 T-cell responsiveness to HCV. Although a recent study suggested an ethnic difference in nonspecific T-cell responsiveness to PHA, ours directly examined the HCV-specific T-cell response in patients with different outcomes of HCV infection. We found that the AA patients responded more readily to HCV than the CA patients among those with chronic HCV infection. However, the ethnic groups responded similarly to PHA, tetanus, and Candida, suggesting that their difference in HCV-specific response is not from a global discrepancy in immune function. Although both groups responded best to HCV core, they differed sharply in their response to the NS antigens (0% CA vs. 22% AA, P = .007). Based on the clinical and immunologic findings, it is tempting to associate the 2 findings and speculate that the improved clinical outcome in AA patients is caused by their greater immune responsiveness. Furthermore, these results suggest that the poor response to IFN-based antiviral therapy in AA patients is not caused by a lack of HCV-specific T-cell response.
However, it is likely that additional factors contribute to liver disease progression and virus control during chronic HCV infection. Notably, there was no direct linear relationship between the CD4 T-cell response and individual clinical parameters or viral titer. This may be explained partially by the inability of HCV-specific CD4 T cells to proliferate and produce IFN- in a coordinated manner. This lack of IFN- was not caused by a type 2 response in our patients (data not shown), consistent with published reports of predominantly type 1 T-cell response in HCV infection. Another possibility is that CD8 T cells are involved more directly by directly killing virus-infected cells and producing antiviral cytokines, as shown in lymphocytic choriomeningitis virus, Epstein-Barr virus, and HBV infection. Based on the contribution of CD4 T cells in the overall immune response and in viral pathogenesis, the dysfunctional amplification and IFN- production of HCV-specific CD4 T cells may be responsible indirectly for an ineffective HCV-specific CD8 T-cell response. Finally, the peripheral CD4 T-cell responses may be only a partial reflection of the antiviral T-cell response to HCV because we could not examine the intrahepatic compartment in our patients.
HCV-specific CD4 T-cell response also differed between AA and CA subjects who cleared HCV. Although both ethnic groups responded more vigorously to
HCV NS than core antigen, the recovered AA subjects responded less well to HCV core (21% AA vs. 59% CA subjects, P = .0356). In fact, the recovered and
chronic AA patients were similar in their response to HCV core (21% vs. 26%, P = NS), contrasting with the statistically significant difference between the
recovered and chronic CA subjects (59% vs. 14%, P < .001). Thus, HCV clearance was associated with a significant CD4 T-cell response to HCV core in CA but not AA subjects, suggesting a differential requirement for HCV clearance. There are several potential explanations for this finding. For example, recovered AA subjects may have been exposed to viral isolates that differ from genotype 1. However, this is unlikely because HCV core is highly conserved across the known genotypes and because most recovered subjects displayed serotype response to genotype 1 (data not shown). Moreover, the same AA subjects responded vigorously to the less-conserved NS antigens. HCV core-specific T cells may have been selectively deleted in vivo, given the association of HCV core with tumor necrosis factor receptor and apoptosis. Another interesting consideration is the recently described alternate HCV core antigen with an entirely different amino acid sequence produced by frame-shift mutations. Finally, HCV core may modulate the antiviral T-cell response through the complement pathway. It will be interesting to examine whether these pathways are more active among AA than CA patients.
Taken together, our results show a novel ethnic difference in CD4 T-cell responsiveness to HCV. Despite speculations about the decreased spontaneous HCV clearance among AA patients, those that do clear HCV showed a robust and broad type 1 CD4 T-cell response to the NS antigens similar to the CA subjects who cleared HCV. By contrast, the lack of concurrent HCV core-specific CD4 T-cell response among the AA subjects suggests that the immunologic requirements for HCV clearance may vary between the 2 ethnic groups. During established chronic HCV infection, AA patients displayed a significantly better clinical parameter of liver disease as well as a greater (albeit mostly dysfunctional) HCV-specific CD4 T-cell response than CA patients. Thus, our study provides the first immunologic analysis of a large number of HCV-recovered and chronically infected patients, and sets the stage for further studies on host-virus interaction, including during antiviral therapy and both CD4 and CD8 T-cell response.
|
|
| |
| |
|
|
|
