Should Surveillance for Liver Cancer be Modified in Hepatitis C Patients After Treatment-related Cirrhosis Regression?

Roberta D'Ambrosio; Massimo Colombo


Liver International. 2016;36(6):783-790. 

In This Article

Can HCC be Accurately Predicted in Non-viraemic Patients?

One major issue in SVR patients with advanced fibrosis is represented by the residual risk of developing HCC. In retrospective studies, the annual rate of HCC in SVR patients appears to be significantly attenuated compared to viraemic patients (0.4–2.5%),[8,12–14,16–18,27,30,64] providing the rationale for the recommendation of maintaining lifelong surveillance for HCC in this patient population.[31,33] To optimize surveillance, many efforts have been made to identify predictors of HCC in SVR patients. Among cured patients, age,[8,19,20,22–25,28,65] pretreatment F3/F4 fibrosis[20,23–25,65,66] and low platelet count[22,24,30] have been recognized as risk factors for the development of HCC (Table 2), whereas in other studies, alcohol abuse, diabetes and steatosis have emerged as predictors of HCC in SVR patients[20,23,28,66] (Table 2). In a large study in Taiwan, the strongest predictor of HCC development in SVR patients was the presence of cirrhosis [HR 4.98 (95% CI: 2.32–10.71)], age [HR 1.06 (95% CI: 1.02–1.11)] and baseline γGT values [HR 1.01 (95% CI: 1.00–1.01)], whereas in non cirrhotics, high baseline γGT values [HR 6.44 (95% CI: 2.20–18.89)] and age >60 years [HR 3.68 (95% CI: 1.33–10.17)] emerged as independent risk factors.[25] Among the latter patients, HCC developed more frequently in those cumulating more than one risk factor [HR 9.06 (95% CI: 2.01–40.92) for 1 risk factor; HR 20.62 (3.77–112.77) for 2 risk factors], whereas HCC risk was similar in patients with and without cirrhosis who carried two independent risk predictors, like age and elevated γGT. Results from that study were consistent with a previous study in Taiwan, showing increased risk of HCC among SVR patients with advanced fibrosis (F3–F4), older age (≥60 years), high AFP levels (≥20 ng/mL) and low platelet count (<150 × 109/L). Interestingly enough, diabetes mellitus, which was diagnosed in 21% of the patients, emerged as a risk factor for HCC at univariate but not at multivariate analysis.[24] This allowed to develop a predictive model with an acceptable predictive value for HCC (ROC 84.4%) [(ScoreHCC) 5 (if age ≥60 years) + 4 (if PLT <150 × 109/L) + 4 (if AFP ≥20 ng/mL) + 6 (if fibrosis F3–F4)], whereby patients could be stratified into three risk groups: low risk (ScoreHCC ≤10), intermediate risk (ScoreHCC 11–15) and high risk (ScoreHCC ≥16) of HCC. The corresponding 5-year risk for HCC was 1.4% vs. 9.1% vs. 30.8% (P < 0.0001). Age is a relevant predictor of HCC risk in Caucasian patients with an SVR, too, where the 5-year incidence of HCC peaked 12.2% in those >60 years (5.4–19.1%).[26] However, there are also studies suggesting that HCC risk is not only a function of age [HR 2.60 (1.48–4.58); P = 0.001] and sex [HR 3.42 (1.01–11.63); P = 0.049] but also of alcohol abuse [HR 2.68 (1.14–6.34); P = 0.024] and presence of type 2 diabetes [HR 4.76 (1.60–14.10); P = 0.005].[28] Findings, therefore, suggest that such a score made to predict HCC in SVR patients should be adopted according to geographical regions to take into account the potential impact of both environmental and ethnic-related risk factors.