Table 7.

Comparison among four two-stage community-based HCC screenings

References*Model 1 (9)*Model 2 (10, 2, 15)*Model 3 (19)*Model 4
First authorWu JCChen CJ, 1995Lu SNCurrent study
Year published1988Chen TH, 2002 Yang HI, 20022006
Sample size of screening1,894Male/female: 12,026/1,800(Low) 1,694, (High) 4,61656,702
Screening makers in first stageAFPHBsAg, anti-HCV, AFP, aspartate aminotransferase, alanine aminotransferase, family historyPlatelet(A), (P), (B)
% High-risk subjects identified from population1%Male/female: 30.9%/34.6%(Low) 6.1%, (High)17.9%(A) 0.64%, (P) 5.33%, (B) 5.70%
Estimated sensitivity for HCC detectionNot available>90%48%(A) 50.5%, (P) 54.5%, (B) 71.3%
Second-stage Screening toolsUS in hospitalUS in communityUS, AFP in communityUS in community
PPV of first-stage marker(s)21%Not available(Low) 0%, (High) 4.27%17.8%, 1.89%, 2.41%
Surveillance program for high-risk groupNilCirrhosis every 3 mo, others every 6 moNilLimited to HBsAg(+), anti-HCV(+), or cirrhosis
  • * (9), Wu JC, Liver 1988; (10), Chen CJ, J Formosan Med Assoc 1995; (2), Chen THH, Int J Cancer 2002; (15), Yang HI, NEJM 2002; (18), Lu SN, Cancer 2006.

  • (Low) and (High), low and high prevalence areas of hepatitis C and hepatocellular carcinoma.

  • (A), AFP alone; (P), platelet count alone; (Both), combination of both AFP and platelet count.