| |
Treatment of Chronic Hepatitis C Virus
in African Americans With Genotypes 2 and 3
|
| |
| |
The American Journal of Gastroenterology
OnlineEarly Articles March 2007
Mitchell L. Shiffman et al, M.D.11Hepatology Section, Virginia Commonwealth University Medical Center, Richmond, Virginia and McGuire Veterans Administration Medical Center, Richmond, Virginia
ABSTRACT
Previous studies have documented that sustained virologic response (SVR) is significantly reduced in African Americans with chronic HCV genotype 1 following treatment with interferon and ribavirin when compared with Caucasians. The specific aim of the present retrospective study was to assess virologic response to interferon and ribavirin in African Americans with HCV genotypes 2 and 3. A review of our database identified 42 African Americans and 334 Caucasians with HCV genotypes 2 and 3. Patients coinfected with hepatitis B or human immunodeficiency virus, chronic renal failure, and recipients of an organ transplant were excluded. Thirty of the African Americans were treated with either standard interferon or peginterferon and ribavirin as initial treatment for chronic HCV. Ninety of the 334 Caucasians were matched to the African Americans with regards to genotype, cirrhosis, treatment regimen, sex, age, and body weight for comparison of virologic response. The proportion of patients with HCV genotype 2 was significantly greater (P < 0.001) in African Americans compared with Caucasians (81% vs 52%). End-of-treatment virologic response was observed in 94% of Caucasians compared with 80% in African Americans (P = 0.036). SVR was observed in 82% and 57% of Caucasians and African Americans, respectively (P = 0.012). Similar results were observed when patients who had been treated with only peginterferon and ribavirin were assessed. These results suggest that African Americans have a global defect in their ability to eradicate HCV infection following treatment with interferon and ribavirin which transcends across all genotypes.
The hepatitis C virus exists as a family of six genotypes, of which three predominate in North and South America, Western Europe, Japan, and Australia (1, 2). In the United States approximately 70% of persons have genotype 1; nearly equal amounts of the remaining patients have genotypes 2 and 3 (1-3). Genotypes 4, 5, and 6 tend to be found almost exclusively in those persons who immigrated to this country from regions of the world where these strains of HCV are commonly found. The exception to this is African Americans; 97% of whom are infected with HCV genotype 1 (4-6).
Patients with HCV genotype 1 are far less responsive to interferon therapy than are patients with genotypes 2 and 3. Large controlled clinical trials have demonstrated that approximately 40-50% of patients with genotype 1 achieve a sustained virologic response (SVR) following 48 wk of treatment with peginterferon and ribavirin (7-10). In contrast, SVR is observed in about 80% of patients with HCV genotypes 2 and 3 and this can be achieved with just 24 wk of therapy, a lower dose of ribavirin, and either standard or peginterferon.
As a group, African Americans respond far less well to interferon therapy than persons of other races. This observation was initially noted nearly a decade ago when standard interferon was administered three times weekly (t.i.w.) for treatment of chronic HCV (11). More recently three controlled clinical trials have clearly documented that the SVR rate in African Americans with HCV genotype 1 is approximately 40-50% lower than that observed in Caucasisans (12-14). However, since only 3% of African Americans have HCV genotypes 2 and 3, no data currently exist regarding the SVR rate of African Americans with these genotypes.
The present study is a retrospective analysis of African Americans with genotypes 2 and 3 who received treatment for HCV at our Center over the past decade. Their virologic response was compared to a group of Caucasian patients matched for genotype, the presence of cirrhosis, type of therapy, sex, and body weight, and who received treatment during this same time interval. Our results demonstrate that African Americans with HCV genotypes 2 and 3, like that previously reported for genotype 1, have a significantly lower SVR than observed in Caucasians.
RESULTS
Demographic and Clinical Features
Table 1 contains demographic features, baseline laboratory data, HCV genotype, and liver histology characteristics of the 49 African Americans and 334 Caucasians with HCV genotypes 2 and 3 identified in our databases. The overall age of the patients in this cohort was 45 yr, 75% were male, 15% had cirrhosis, and 79% had received treatment with either standard or peginterferon and ribavirin. No significant differences between these two groups existed with regards to the percentage of males, the two major risk factors associated with HCV acquisition, mean histology score, the percentage of patients with cirrhosis, and the percent who received HCV treatment. In contrast, African Americans were a few years older, weighed more, had a lower mean serum alanine aminotransferase (ALT), lower mean hemoglobin, and lower mean white cell count than Caucasians. A significant difference was also present between the two groups with respect to the distribution of HCV genotypes. Over 80% of African Americans were genotype 2 compared with 52% of Caucasians.
Table 2 compares demographic features, baseline laboratory data (just prior to starting treatment), HCV genotype, and liver histology characteristics of the 30 African Americans who received treatment for HCV to these of the 90 matched Caucasians. As per the matching criteria, the groups contained the same percentage of patients with genotype 2, cirrhosis, and were of similar age and body weight. The percentages of patients who received a ribavirin dose of 800 mg daily were also matched (80% recd 800mg) (30-35% recd IFN, 60-68% recd pegIFN). The number of patients treated with standard interferon with a plan for 24 wk of therapy was identical in both groups although not all of these patients could be matched for all other characteristics. The percentage of Caucasian females could not be matched 3:1 to the African Americans for genotype and cirrhosis but was very similar. The number of Caucasians who received peginterferon with a plan for 48 wk of therapy was also insufficient for matching to the two African Americans who received this treatment regimen. Caucasians who received peginterferon planned for 24 wk of therapy were therefore substituted. Those pretreatment factors, which were not purposely matched across the two groups, were similar to that observed for the two racial groups as a whole. This included a significantly lower mean hemoglobin and serum ALT in the African Americans. However, the percentage of patients with a normal ALT at the start of therapy was similar in the two groups.
Virologic Response
Virologic response rates for the African Americans and Caucasian cohorts are illustrated in Figure 1. Ninety-four percent of Caucasians were HCV RNA undetectable at the completion of treatment. SVR was observed in 82% of these patients. In contrast, 80% of African Americans achieved an end-of-treatment virologic response and only 57% achieved SVR. Both of these values were significantly lower than observed in Caucasians (P = 0.036 and 0.012, respectively). The rate of relapse in African Americans was more than twice that observed in Caucasians; 29% versus 13% (P = 0.53).
Because African Americans with genotype 2 and 3 may respond better to peginterferon and ribavirin than standard interferon and ribavirin both end-of-treatment and SVR were calculated for those genotype 2 patients who received peginterferon and ribavirin. Only two African American patients with genotype 3 received peginterferon and ribavirin. As a result, evaluation of the genotype 3 response rate in this racial group was not feasible. A total of 16 African Americans with genotype 2 were treated with peginterferon and ribavirin, all but one for 24 wk. Forty-eight matched Caucasians with genotype 2 also received peginterferon and ribavirin. In both racial groups the virologic responses observed in this subset of patients mirrored those observed for their entire populations, respectively (Fig. 2). In Caucasians, end-of-treatment virologic response and SVR were 94% and 83%, respectively. In African Americans the end-of-treatment virologic response was 75% and SVR 56%. Both end-of-treatment and SVR rates were lower than observed in Caucasians (P = 0.06 and 0.04, respectively).
METHODS
Identification of Patients
Two HCV databases, one maintained by the Hepatology Section and the other by the Molecular Diagnostics laboratory, were searched for persons whose race was self-reported to be either African American or Caucasian and who tested positive for HCV genotypes 2 or 3. Patients coinfected with either the hepatitis B or human immunodeficiency viruses, those with other coexistent liver disorders, chronic renal failure on dialysis, and patients who had previously received a liver transplant were excluded. This search yielded 42 African Americans and 334 Caucasians with HCV genotypes 2 or 3. The clinical records of these patients were reviewed in detail to assure that race and HCV genotype had been properly recorded and to identify those who had initiated treatment with either standard interferon alfa-2a or 2b (t.i.w.) and daily ribavirin or peginterferon alfa-2a or 2b once weekly and daily ribavirin. Thirty of the 42 African Americans received HCV treatment. Ninety of the 334 Caucasian patients were selected as a control group by matching to the African Americans 3:1 for the following features: genotype, presence or absence of cirrhosis, the type of interferon received (either standard interferon or peginterferon), the dose of ribavirin received, the planned duration of therapy, sex, and body weight within 5 kg. All patients included in this analysis (both the African Americans and Caucasians) who were HCV RNA undetectable 24 wk after treatment had been stopped also had to have been followed on a regular basis with the last HCV RNA measurement performed within the past 1 yr utilizing a sensitive HCV RNA assay. Approval to conduct this retrospective analysis was granted by the Committee on the Conduct of Human Investigation at the Virginia Commonwealth University Medical Center.
Demographic, biochemical, virologic, and histologic data at the time of the first office visit were recorded. In those patients who received HCV treatment these characteristics were also recorded at baseline, just prior to the initiation of therapy. The two major risk factors thought to contribute to HCV infection were also recorded. The liver biopsy slides of all patients in this study were reviewed by one of two liver pathologists at our institution. This also included all liver biopsy slides for those patients who had this procedure performed outside of our institution. Each biopsy was assigned a score for inflammation and fibrosis according to the Histologic Activity Index (HAI) of Knodell et al. (15).
Assessment of HCV RNA and Determination of Virologic Response
All quantitative and qualitative measurements of serum HCV RNA prior to the initiation of therapy, during therapy, and after therapy were recorded for each patient. However, since multiple different assays and units were utilized to measure and report serum HCV RNA at our Center during the past decade these data could not be tabulated and reported.
Virologic response was defined as being HCV RNA undetectable at the completion of treatment, whether this was for 24 or 48 wk in duration. The standard definition of SVR, undetectable HCV RNA in serum 24 wk following the termination of antiviral therapy, could not be routinely applied to all patients in this study because some of these patients had received their HCV treatment more than 5 yr ago, when less sensitive assays for HCV RNA in serum were utilized. However, all patients with "presumed" SVR at our Center are followed at 1-2 yr intervals with repeat HCV RNA determinations performed to ensure that SVR has been maintained. As a result, all patients with SVR in this study had at least three undetectable HCV RNA determinations performed since they discontinued HCV therapy (some patients had as many as six undetectable HCV RNA tests) with the most recent testing performed within the past 12 months utilizing Taqman PCR. The lower limit for detection of HCV RNA by this assay is 50 IU/mL.
Statistics
Values for various demographic factors, laboratory studies, and liver biopsy scores were reported as mean ± standard error. Differences between the means of specific variables were compared with a paired Student's t-test or Mann-Whitney rank sum test as appropriate. Virologic response rates were determined on the basis of the principle of intention-to-treat. Differences in the frequency of certain variables and virologic response rates between patient groups were compared with the χ2 analysis or Fisher's exact test as appropriate.
DISCUSSION
The present study is a retrospective evaluation of African Americans with HCV genotypes 2 and 3 who received treatment at our Center. Only 3% of African Americans with chronic HCV infection have genotypes 2 or 3 and it has therefore taken our Center nearly a decade to identify 42 such individuals. During this time virologic assays have become more sensitive and the treatment of HCV has evolved from standard interferon administered three times weekly to peginterferon administered once weekly. Although SVR rates in patients with genotype 1 infection have increased significantly with the evolution of improved treatment, this has not been the case for patients with genotypes 2 and 3 (7-9). Three large randomized controlled trials have demonstrated that SVR rates in patients with genotypes 2 and 3 were similar regardless of the treatment utilized (standard interferon vs peginterferon), the duration of this treatment (24 vs 48 wk), and the dose of ribavirin (800 mg vs 1,000-1,200 mg daily). We therefore felt it was reasonable to combine these treatment regimens when assessing virologic response in the current study. Our results demonstrated that African Americans with HCV genotypes 2 and 3 had an SVR rate of only 56% and this was significantly lower than the SVR rate observed in Caucasians. This lower rate of SVR was similar to that recently reported in a much smaller group of African Americans (N = 8) with genotypes 2 and 3 from a multicenter Veterans Administration Medical Center study (16). While combining treatment regimens was probably reasonable in Caucasians with HCV genotypes 2 and 3, it is certainly possible that African Americans could have had a better response when treated with peginterferon than standard interferon and ribavirin. For this reason we also analyzed virologic response in the 16 African Americans with genotype 2 treated with peginterferon and ribavirin. This was very similar to the SVR rate observed for the entire African-American cohort and significantly lower than the SVR rate observed in a matched group of genotype 2 Caucasians also treated with peginterferon and ribavirin.
Over the past decade, virologic assays for HCV RNA have become more sensitive. We therefore felt it was important to enure that the patients who had been treated years ago and were thought to have achieved an SVR with a less effective regimen have actually remained HCV RNA undetectable. It is for this reason that all patients treated for HCV at our Center are followed every 1-2 yr with repeat HCV RNA testing to ensure they remain HCV RNA undetectable. The same was true for those patients included in the present study. Thus, many of these patients have been repeatedly HCV RNA undetectable over several years since stopping HCV treatment. The most recent testing in all patients was performed by Taqman PCR with a lower limit for detection of 50 IU/mL. We are therefore very confident that the lower SVR rate reported for the African Americans in this study is as valid as possible given the limited sample size of this group.
Retrospective studies, particularly those that evaluate uncommon disorders in a limited patient population, are prone to bias, and this may lead to incorrect conclusions. This is clearly a possibility in the present study. To limit this as best as possible we matched each of the 30 African American who received HCV treatment to three Caucasians for as many of the factors thought to affect virologic response as was possible. These included genotype, the presence of cirrhosis, the type of interferon utilized, ribavirin dose, the planned duration of therapy, sex, age, and body weight. We were able to accomplish such extensive matching because the number of Caucasians with genotypes 2 and 3 in our database who received treatment for HCV exceeded the number of African Americans by nearly ninefold.
The reduced SVR rate to interferon therapy in African Americans was first observed when standard interferon monotherapy was utilized to treat chronic HCV. The SVR rate in African Americans with this treatment was only 2%, compared with 12% in Caucasians (11). The rate of SVR increased significantly when ribavirin was utilized along with standard interferon. However, this remained significantly lower in African Americans when compared with Caucasians; 27% versus 11%, respectively (4). More recently, three controlled trials have demonstrated that African Americans with genotype 1 also have a significantly lower SVR rate when treated with peginterferon and ribavirin compared with Caucasians (12-14). The decline in SVR rates observed in the African American population was secondary to both a decline in the rate of virologic response and to an increase in the relapse rate (12-14, 16, 17). A similar observation was made in the current study. This strongly supports the hypothesis that African Americans have a global defect in their ability to respond to interferon and eradicate HCV infection that transcends across all genotypes. The results of several studies support this hypothesis. The ability of interferon to inhibit viral replication and increase HCV clearance is significantly reduced in African Americans (18, 19). African Americans with chronic HCV have alterations in their cytokine profile, less efficient T-cell responses to HCV antigens, and clear acute HCV infection less frequently than do Caucasians (19-22). Various aberrations in the immune response and differences in HLA allelles are also present in African Americans compared with Caucasians (23). Alternatively, recent data from the VIRAHEP-C study have demonstrated that the reduced SVR rate in African Americans is not secondary to differences in body weight, economic, environmental, and/or virologic factors (14).
Previous studies have demonstrated that roughly equal percentages of HCV patients in the United States are infected with genotypes 2 and 3 (3, 6, 7). A similar observation was made in the current study; 52% of the Caucasian cohort were genotype 2 and 48% genotype 3. In contrast, 81% of African Americans were genotype 2 and only 19% had genotype 3. This interesting observation had not been previously recognized. This is not surprising given that only 3% of African Americans with HCV infection have genotypes 2 or 3 combined and that less than 15% of patients typically enrolled into large multicenter clinical trials were African Americans (3, 7-10). Thus, a large clinical trial of 1,000 patients would have included only 4-5 African Americans with genotypes 2 and 3 combined. Such studies would have been unable to determine if the distribution of genotypes 2 and 3 were indeed different in African American and Caucasian populations or evaluate the response of African Americans to antiviral therapy. The higher percentage of genotype 2 relative to genotype 3 in African Americans is equally as curious as the much higher percentage of genotype 1 in this population when compared with the genotype distribution observed in Caucasians. African Americans and Caucasians have similar risk factors for acquiring HCV and no previous study has ever suggested that genotype is related to the mode of infection (3, 5, 6, 10-12). We believe the most plausible hypothesis for this observation is that African Americans are capable of spontaneously resolving nearly all exposures to HCV genotype 3 and the vast majority of genotype 2 infections. Clinical trials in African Americans with acute HCV infection would be required to test this hypothesis.
In conclusion, the present study represents the largest cohort of African Americans with HCV genotypes 2 and 3 evaluated to date. Curiously, genotype 2 is found significantly more often than genotype 3 in this racial group, as opposed to Caucasians where genotypes 2 and 3 are found in roughly equal proportions. Furthermore, African Americans with HCV genotypes 2 and 3 appear to have a significantly lower SVR rate when treated with peginterferon and ribavirin than Caucasians. This is similar to what has been previously reported for African Americans with HCV genotype 1. These observations strongly suggest that African Americans have a global defect in interferon responsiveness which transcends across all genotypes and is likely to be genetically mediated. A multicenter prospective study will be required to confirm these observations and to test treatment strategies designed at overcoming this defect.
|
|
| |
| |
|
|
|
