Clinical significance of elevated alpha‐fetoprotein in Alaskan Native patients with chronic hepatitis C

Abstract

The clinical significance of elevated serum alpha-fetoprotein (AFP) in patients with chronic hepatitis C virus (HCV) infection is not well defined. We analysed data from a population-based cohort of patients with HCV infection to assess the prevalence of elevated serum AFP, to determine its association with clinical and virologic parameters and with clinical outcomes. We defined a slightly elevated serum AFP level as 8 to or =15 microg/L. Among 541 HCV-RNA-positive persons, 61 (11%) had a slightly elevated or high AFP at the time of consent. AFP > or =8 microg/L was associated with the older age, aspartate aminotransferase/alanine aminotransferase ratio >1, and higher alkaline phosphatase levels, but not with heavy alcohol use, IV drug use, genotype, viral load or duration of HCV infection. Among 192 persons with an AFP at liver biopsy, 17% had an AFP > or =8 microg/L. The sensitivity/specificity of an AFP level > or =8 in detecting Ishak 3-6 fibrosis was 39%/95%. Among 372 persons with a minimum of four AFP measurements over 6 years, 5% had persistently elevated AFP >8 microg/L, 19% had both elevated and normal AFP measurements, and 76% had persistently normal AFP. Elevated AFP at consent was associated with hepatocellular carcinoma (HCC) and end-stage liver disease. Over 6 years of follow-up, persistently elevated AFP was associated with the development of HCC; no person with AFP persistently <8 microg/mL developed HCC. Serial AFP measurements appear to be useful in identifying persons with advanced fibrosis and help to determine who needs periodic screening with liver ultrasound to detect HCC.