Focus.

摘要

AFP was first described by Abelev et al. in the early 1960s [1,2]. Over the next 30 years, and before the introduction of modern imaging techniques, AFP measurement was frequently used for screening of patients at risk, confirmation of diagnosis of hepato-cellular carcinoma (HCC), and even surveillance. However, in the past decade, it became progressively clear that serum AFP measurements do not correlate reliably with tumor size, stage, and prognosis in HCC [3]. At a cutoff of 20 ug/ml, AFP has a 60% sensitivity and 91% specificity as a diagnostic tool for HCC with positive and negative predictive values of 25% and 98%, respectively (at an HCC prevalence of 5%) [4]. These data lead to a 2001 Editorial published in this Journal by M. Sherman entitled: "Alpha-feto protein: an obituary", who suggested "bidding a fond adieu to AFP as a test for diagnosis and surveillance of HCC" [5]. Consequently, the recent AASLD guidelines for the management of HCC adopted this policy verbatim and excluded the recommendation for testing and monitoring of AFP levels for diagnosis and surveillance of HCC in patients at risk [6]. Despite these reservations, many clinicians still check AFP in their routine clinical practice, knowing that markedly elevated AFP levels are a poor prognostic sign in HCC.