In patients with hepatitis C who develop hepatocellular carcinoma, the serum alpha-fetoprotein level around the time when the cancer is diagnosed is an independent predictor of mortality, Dr. Gia L. Tyson and her colleagues reported in the November issue of Clinical Gastroenterology and Hepatology.
Serum alpha-fetoprotein (AFP) level has been incorporated into at least three of the major staging and prognostic scoring systems for hepatocellular carcinoma, all of which were developed in patient populations outside the United States. But until now it was unknown whether AFP level would be predictive in U.S. patients in general, or in U.S. patients whose HCC is related to hepatitis C in particular.
This retrospective cohort study of 1,480 patients shows that these staging and prognostic systems are relevant in the United States, and that testing for serum AFP when HCC is diagnosed would be useful. Such testing is inexpensive, widely available, and easily interpretable, said Dr. Tyson of the Houston VA Health Services Research and Development Center of Excellence and Baylor College of Medicine, and her associates.
The investigators reviewed the records of adults enrolled in a national VA registry of hepatitis C patients who developed HCC between 1998 and 2007, and followed them through 2009 to determine whether serum AFP level correlated with mortality (Clin. Gastroenterol. Hepatol. 2011 November [doi: 10.1016/j.cgh.2011.07.026]).
As veterans, most of the patients (99%) were men and about half (56%) were white. The mean age was 58 years, and 40% of the subjects had cirrhosis. The interval between hepatitis C diagnosis and liver cancer diagnosis was a mean of 3.86 years.
Overall mortality was 87% during a mean follow-up of 1.8 years. After HCC diagnosis, median 1-year survival was 43%, median 3-year survival was 19%, and median 5-year survival was 12% in the study population as a whole.
Median survival was found to be significantly shorter in patients who had high AFP levels around the time of HCC diagnosis.
"This is the largest study in a United States HCV-infected cohort to report serum AFP levels are predictive of mortality after HCC diagnosis."
Specifically, the median survival was 709 days for patients with an AFP level less than 10 ng/mL, 422 days for those with an AFP level of 10-99 ng/mL, 208 days for those with an AFP level of 100-999 ng/mL, and 68 days for those with an AFP level of 1,000 ng/mL or more. Similarly, survival rates at 1 year, 3 years, and 5 years after HCC diagnosis were progressively lower as AFP levels increased.
The 5-year survival rate for HCC was low overall at only 12%, but it was extremely low, at 1%, in patients with serum AFP levels of 1,000 ng/mL or more. In comparison, those with AFP levels less than 10 ng/mL had a 24% survival rate at 5 years, Dr. Tyson and her colleagues said.
The risk of death increased with increasing AFP level independently of patient factors such as age, sex, and race/ethnicity; independently of clinical factors such as the presence of ascites, encephalopathy, and congestive heart failure; and independently of treatment received, including liver transplantation, resection, radiofrequency ablation, or transarterial chemoembolization.
The results of two sensitivity analyses confirmed that mortality increased significantly with increasing AFP level, with hazard ratios of 1.51 for an AFP level of 10-99 ng/mL, 2.29 for an AFP level of 100-999 ng/mL, and 4.22 for an AFP level of 1,000 ng/mL or more, compared with an AFP level less than 10 ng/mL.
"This is the largest study in a United States HCV-infected cohort to report serum AFP levels are predictive of mortality after HCC diagnosis," the investigators noted.
The dose-response relationship between AFP level and mortality risk further strengthens the role of AFP as a predictor of prognosis, they added. These findings are consistent with results from Italian, French, and Chinese patient populations.
"The main limitation" of this study was the lack of data concerning tumor size and staging. However, numerous studies elsewhere, including those in Italy, France, and China, have demonstrated that AFP as a prognostic factor is independent of tumor size and stage, Dr. Tyson and her associates noted.
This study also was limited in that nearly all of the patients were men and most were white, so the findings may not be generalizable to women and other racial/ethnic groups.
This study was supported in part by the National Institute of Diabetes and Digestive and Kidney Diseases, and the Houston Veterans Affairs Health Services Research and Development Center of Excellence. No financial conflicts of interest were reported.