 |
|
|
| |
RACIAL DIFFERENCES IN RISK FOR LIVER CANCER IN PATIENTS WITH HEPATITIS C CIRRHOSIS: A MULTICENTER CASE-CONTROL STUDY
Reported by Jules Levin
|
| |
| |
Digestive Disease Week 2003, May 17-23, Orlando. Abstract 121. Authors: Mindie H Nguyen, Ruel T Garcia, Jing Ning, Saraa A Tawfeek, Suzanna Lam, Ghasak Mahmood, Teresa L Wright, Emmet B Keeffe, Palo Alto, CA; San Francisco, CA.
M Nguyen, MD, gave an oral presentation of this study Sunday afternoon May 18 at DDW. Several observers in the audience criticized the study design and did not feel the conclusions were valid because of this. The study looks at patients with HCV cirrhosis. The criticism is that these patients were not followed over a period of years as individuals part of a study. This would allow them to be followed longitudinally and they all would then be more likely to receive similar access to care and treatment. But the study examined patients who already had HCV cirrhosis. African-Americans may be more likely to develop HCV cirrhosis because they may be less likely to receive the same care and treatment as Caucasians. And then they get referred to these sites for care and therefore there could be more HCV cirrhotics among African-Americans. This is the criticism offered by several observers in the audience, and here is the data presented by the study authors. In speaking with a co-author of this study they feel that Blacks probably are
more likely to progress to liver cancer after developing cirrhosis. They feel
that the criticism appears valid and contributes to the study finding, but
that another contributing factor in the study findings is due to Blacks
probably are more likely to go onto liver cancer after developing cirrhosis.
Hepatitis C is a major risk factor for liver cancer, accounting for up to 50% of all cases in the US. Her study asks: "among those with hepatitis C cirrhosis, are asian-Americans and African-Americans more vulnerable to liver cancer?"
This study is a retrospective case-control study from 1998-2002 and a criticism is that patients need to be followed prospectively to properly attempt to answer this question. Hepatitis B carriers and patients with chronic hepatitis B, HIV or other malignancies were not included in this study. Cases were confirmed by cytology, histology and/or by the presence of focal hypervascular hepatic mass (on biphasic CT, MRI and/or angiogram) and elevated AFP. HCC was ruled out in controls by negative AFP and imaging studies. Multivariate logistic regression was employed to examine associations between HCC and race. Adjustment was made for age, gender, severity of liver disease (MELD score, Child class), moderate-to-heavy alcohol use, and study centers. For all variables, values at diagnosis were obtained for HCC cases and at first negative AFP and x-ray for controls.
496 patients with HCV cirrhosis were included: 217 cases and 279 controls.
RESULTS
The cancer cases were expectedly older ( 59 vs 52 years old) (<0.001), more likely to be male (83% vs 71%, 0.001), more likely to have Child-more advanced liver disease-(0.001), and the median MELD score was 11 for the cancer patients vs 10 for the controls (0.05).
Adjustments were made for age, gender, Child class, MELD, and study center. The unadjusted odd ratio was 2.9 for African-Americans. The author said this was statistically significant. The adjusted odd ratio was 1.9 for the African-Americans vs the Caucasians; the author said this is a trend suggesting 4 times greater risk. The author said that this was not statistically significant and this could be due to the small number of patients in the African-American group: there were 105 Caucasians in the cancer group and 188 Caucasians in the controls group, but there were 11 African-Americans in the cancer group and 15 in the controls group. The adjusted odds ratio for Asian-Americans was 3.8 and the unadjusted odds ratio was 3.9.
The authors concluded that among patients with hepatitis C cirrhosis liver cancer risk is increased in African-Americans and and Asian-Americans.
The authors suggested limitations to the study. Potentially contributing but unmeasured factors: occult hepatitis B infection; duration of infection; alcohol and cigarette use was not able to be measured.
We need well designed longitudinal studies among diverse ethnic populations to address this question.
|
|
| |
|
|
|
|
|
