Screening for hepatocellular carcinoma: From lack of evidence to common sense

Abstract

Davila et al. showed that 17% of the patients older than 65 years with cirrhosis underwent regular screening for hepatocellular carcinoma (HCC), in whom only 54% had only an ultrasound procedure. Gastroenterologists were more likely (4.5-fold) than primary care physicians to perform regular surveillance.1 Hepatologist associations recommended that “patients at high risk for developing hepatocellular carcinoma should be entered into surveillance programs (Level I)”.2 However, only inconclusive or negative observational studies are available. Trevisani et al. concluded that screening improved survival (5 months) despite raw data showing that screened patients died 18 months younger than nonscreened patients (length of time and lead time biases)!3, 4 Other examples are available: Kemp et al. reported a 26-month increase in survival in screened versus incidentally discovered HCC, but screened patients were 3 years younger.5, 6 Recently, I reviewed a large series with similar biases and the authors refused to resubmit a revised version against screening. The National Cancer Institute wisely stated (last revision on April 3, 2008) that “based on fair evidence, screening would not result in a decrease in mortality from HCC … based on fair evidence, screening would result in rare but serious side effects (Study Design: Randomized controlled trials and observational studies. Internal validity: Fair. Consistency: Multiple studies, large number of participants. External validity: Good/Fair.)”.7 In real life, screening is limited by patient compliance and operator reliability, and there is no consensus about the interval or the method. Screening is a complex issue which necessitates a national program to ensure a minimal participation of the population, quality controls, and evaluation of the results. The call, recall, and follow-up systems require major commitments, and in this case drop-outs are substantial. Finally, overdiagnosis, a well-known complication of screening, is an ignored critical issue. The U.S. Institute of Medicine recently issued a report8 that highlights the pitfalls of the federally sponsored cancer clinical trials system. However, it does not explain ineffective collaboration … recruiting is not an issue: HCC is the fifth most common cause of cancer. Screening advocates must understand that patients deserve evidence-based treatments and that poor evidence is a leading cause of poor compliance, a situation precluding efficiency for any screening policies. Errare humanum est, perseverare diabolicum (“to err is human, but to persist [in the mistake] is diabolical”). For the present time, clinicians must not forget that promoting smoking cessation, informing on limitation of alcohol intake, and vaccinating against hepatitis B virus are the three most cost-effective measures to prevent HCC. Cigarette smoking is an independent and a dose-related contributing factor for HCC worldwide, even in Asia.9 The mean relative risk is 1.5 but exposure is incredibly high. In France, tobacco, hepatitis, and alcohol are the three main risk factors for HCC, contributing 33%, 31%, and 26%, respectively, to HCC.10 How many gastroenterologists/hepatologists are promoting smoking cessation? Alain Braillon M.D.*, * Department of Public Health, University Hospital of Amiens, Amiens, France.