In our series of 82 Asian Americans with HCC treated without transplantation at our tertiary cancer center between 1992 and 2005, we found that the rate of chronic HBV infection was high, at 61%. The rate of chronic HCV infection was significantly lower, at approximately 20%. The prevalence of other risk factors for HCC, such as cigarette use and alcohol abuse, was high, especially among patients with HCV. Survival outcomes were poor overall and in the HBV and HCV subgroups. The major predictors of worse survival were male sex, high alpha-fetoprotein levels, stage IV disease, and lack of treatment for HCC.
Nearly 94% of the patients in our study had previous HBV infection, as evidenced by a positive anti-HBc test result, and 61% of the patients had chronic HBV, as indicated by HBsAg seropositivity. The rate of chronic HBV in our study was lower than the 80% rate reported in a previous study by Hwang et al. of 79 Asian American HCC patients in California . The higher rate of chronic HBV in that study may reflect the overall higher prevalence of chronic HBV among the Asian American population in California, reported to be as high as 8.9% . Our cancer center is a tertiary referral center and sees cancer patients referred from various areas: about 58% are from Texas, 39% are from other states in the US, and 3% are international patients (unpublished institutional data). In another study of 220 Asian and Pacific Islander HCC patients in Hawaii, where the prevalence of chronic HBV may be less than 2% , Wong et al. reported that approximately 50% of the study patients had HBV , similar to the chronic HBV rate in our study (although the specific HBV test used in the Wong et al. study was not stated).
Our study also showed that the proportion of Asian American HCC patients with previous HBV did not decrease over time. This finding contrasts with the results of a study by Hassan et al., who studied 359 HCC patients seen at our institution during a similar period, 1993–1998, and found that the rate of previous HBV infection decreased over time . Hassan et al. reported HBV infection rates for the overall group and did not break down results by ethnicity . Given the paucity of population-based surveillance data describing HBV infection rates in the United States, our finding of no change in the rate of chronic HBV infection among Asian Americans with HCC has important implications in terms of interventions to reach this underserved group. All Asian Americans must be offered cost-effective HBV screening and vaccination . New and effective anti-HBV medications are available [22, 23], and patients found to have chronic HBV must have close surveillance and be considered for treatment. We believe that primary care physicians are the most logical solution to reduce the burden of HBV-associated HCC among Asian Americans because primary care physicians can institute systematic and cost-effective hepatitis screening and treatment for Asian Americans .
Although HCV is not the most prevalence risk factor for HCC, our study adds support to the body of evidence indicating that HCV is an important risk factor for HCC among Asian Americans, especially for individuals who do not have chronic HBV. Although our numbers are small, our study revealed the rate of HCV infection to be 20% among Asian American HCC patients, of whom 88% had a negative HBsAg test result. In the previously mentioned study by Hwang et al. in California, 20% of the Asian American HCC patients had a positive HCV result, and most of these patients (69%) had no evidence of HBV .
Since the observed prevalence of HBV among patients in our study was high, it is likely that Asian American patients with HCC are relatively less affected than other patients with HCC by other HCC risk factors, such as HCV infection, cigarette smoking, alcohol, and diabetes. Results from our ongoing case-control study where new HCC patients are prospectively enrolled [24, 25] indicate that non-Hispanic whites are more exposed to non-virus risk factors than Asian Americans. For example, the prevalence of cigarette smoking, alcohol drinking, and diabetes mellitus among non-Hispanic whites with HCC was 70.4%, 69.4%, and 31.3%, respectively. In our study of Asian American patients with HCC, we found the prevalence of these non-virus risk factors to be lower – 43.9%, 34.1%, and 11%, respectively. Our study findings were similar to those reported previously for Asian HCC patients in Hawaii (32% rate of alcohol use and 48% rate of cigarette use) . Diabetes mellitus, however, was more prevalent in the Hawaii study (29%) than in our study, and this is likely due to the significant prevalence of diabetes in Hawaii, reported to be approximately 6% . It has been shown that the Asian and Pacific Islander population in Hawaii is twice as likely to be diagnosed with diabetes as US Caucasians .
The median survival rate in our study of Asian American HCC patients, for whom transplantation was not available and most of whom received palliative therapy, was very poor: 9.2 months. This is lower than the median survival time of 1.56 years reported by Wong et al. in their study of Asians and Pacific Islanders in Hawaii . The difference may be attributed to the higher proportion of stage IV disease in our study (61%) than in Wong et al.'s study (32%). The stage of disease at presentation influences whether surgical options are available. Only 11% of the patients in our study had surgical resection; in contrast, in the study of Wong et al. , 54% of patients had stage I or II disease, and 24% of patients underwent liver resection and 7% liver transplantation. This supports the notion that patients with earlier-stage disease treated with surgical treatment options have more favorable survival outcomes [14, 15]. Similarly, in a study of 255 HCC patients by Barazani et al.  in which 18% of patients with HBV-associated HCC had liver transplantation, the 1-year survival rate was 85%, significantly higher than our 1-year survival rate of approximately 40%. Once HCC is advanced and transplantation is no longer an option, survival is poor.
We believe that our study is the only one to date to examine the impact of treatment, excluding liver transplantation, on the survival of Asian American HCC patients. We found that having received any type of palliative therapy or even liver resection was a predictor of longer survival. This again illustrates that it is critical to implement HBV screening, surveillance, and treatment programs for Asian Americans to ensure that if HCC develops, it will be detected early, when curative therapy options are still available.
Our study also found that elevated alpha-fetoprotein levels and stage IV disease were associated with shorter survival. These findings are similar to findings of Wong et al. . However, whereas we found that chronic HBV was not a significant predictor of survival, Wong et al. found that chronic HBV was associated with worse survival . It is unclear which HBV screening test(s) were used to define HBV disease in the Wong et al.  study, and differences in the testing methods may have contributed to the survival differences. Our study does support a previous finding by Hwang et al. , who described no significant survival differences between HCC patients with chronic HBV (positive HBsAg) and HCC patients with HCV (positive anti-HCV).
The proportion of patients in our study who had a history of alcohol use, 35%, is similar to the prevalence of 32% reported for Asian HCC patients by Wong et al. . However, in our study, alcohol use was not a significant predictor of survival, whereas in the Wong et al study, it was . The difference may be due to the definition of alcohol use. We defined alcohol use as at least 4 drinks monthly for 6 months during the patient's lifetime, whereas Wong et al. defined alcohol use as 2 drinks daily for 10 years.
We acknowledge that our study had limitations, including its retrospective nature and small sample size. Our study population represents various subcategories of Asian ethnic groups, and their prevalences may be unique because of their country of origin. In addition, the small number of Asian American HCC patients with HCV is problematic, and the related results discussed above should be interpreted with caution. Our institution is a referral center, and it may be difficult to retrospectively ascertain full details of previous treatments that referred patients received. Similarly, we may have underestimated the survival lengths because our analysis began with the treatments at M. D. Anderson; however, we think that this effect is likely minimal since our referral waiting periods are not lengthy. In addition, we think that our study may have underestimated the true effect of HBV, as we did not specifically examine patients with an isolated positive anti-HBc test result. Some such patients may have an undetectable level of HBsAg and may actually be chronically infected. However, we think that this would be a small number . We are aware that comparative studies of the clinical significance of hepatitis and predictors of survival in patients with HCC of different ethnicities would be helpful and could potentially lead to future advances in policy, practice, and research. We are currently conducting such comparative studies of patients treated at our institution, and a longer report is planned for the future. We are also aware that our institution's status as a tertiary referral center may have biased our study population as such centers may attract patients with later-stage disease and more complex cases.
Our findings in this study have important implications for clinical practice. We believe that primary care physicians are the most logical medical providers to break the burden of HBV-associated HCC among Asian Americans. We advocate the use of cost-effective strategies developed by Hutton et al., such as screening all Asian Americans for HBV, treating those who have evidence of chronic HBV, and subsequently screening and vaccinating family members of patients found to have chronic HBV . These strategies will systematically identify patients at risk for HCC and lead them to appropriate secondary cancer prevention. We also advocate close surveillance and treatment of HBV patients according to therapeutic endpoints . We anticipate that significant efforts will be necessary to implement these strategies in daily clinical patient care, and we call for collaboration among national, state, and local community organizations since up to 25% of Asian Americans in some parts of the country do not have a usual place for health care, most likely because of lack of insurance and/or other barriers .
In conclusion, we have shown that chronic HBV remains a significant problem among Asian American with HCC referred to our institution and that survival of patients with HCC treated without transplantation at our institution is poor. Early detection of HCC is vital for surgical options to be feasible, and systematic screening of Asian American individuals for HBV will be crucial in improving the outcomes of Asian American patients with HCC. Future cost analyses and quality-of-life studies will be necessary to further elucidate the burden of HCC among special populations and the impact of screening and early detection of this disease.