We share the response to the letter from Zhong and Yang by providing an empirical finding as follows: In a previous study, we demonstrated that the effect of diabetes on hepatocelluar carcinoma (HCC) is manifested in patients who are hepatitis C virus (HCV) negative but not in those who are HCV positive.1 Indeed, this finding strongly suggests that the pathway to HCC may not be merely through hepatitis virus infection but through metabolic pathways. In addition to the evidence mentioned above, it is very interesting to note that there is an inverse association between HCV infection and metabolic syndrome (MS) (age- and sex-adjusted odds ratio [aOR] = 0.91 [0.83–1.00]), found by reanalyzing the same dataset but using more subjects and a longer follow-up period as published in another study.2 By looking into individual components of MS, triglyceride (0.64 [0.58–0.70]) and blood pressure (0.91 [0.84–0.99]) were statistically significantly inversely associated with HCC. An inverse association between obesity and HCC was also noted, but was not statistically significant (0.98 [0.90–1.06]). It is also very intriguing to find the lower the high-density lipoprotein level, the higher the odds of having HCV infection (1.52 [1.39–1.68]). It is also very interesting to note that all individual components of MS were inversely related to hepatitis B virus infection.
Regarding how the effect of diabetes on risk of HCC was modified by the severity of hepatic damage among patients with chronic hepatitis C, by making use of the same data on alanine aminotransferase (ALT), aspartate aminotransferase (AST), and alpha-fetoprotein (AFP) excluding prevalent HCC cases as in the previous study,1 we found the effect of diabetes on the risk for HCC is remarkably seen in patients with lower AST (aOR = 2.41 [1.14–5.10]), ALT (aOR = 2.47 [1.12–5.45]), and AFP level (2.56 [1.15–5.71]) but not in patients who were HCV negative but had high levels of these three biological markers (Table 1). Among patients who were HCV positive, there was lack of any significant association between diabetes and the risk for HCC in either high or low level of the three biological markers. Although these findings were not consistent with the results reported by Veldt et al.,3 we think both of these findings are not contradictory to each other because our subjects with HCV infection were enrolled from community-based screening rather than from hospital-based patients and may not have had advanced fibrosis or cirrhosis as might the patients recruited in their study.
|Class 1||Class 2||RR of Diabetes/Non-Diabetes|
|HCV(+)||AST < 60 IU/L||2.02 (0.53–7.68)|
|AST ≥ 60 IU/L||0.18 (0.02–1.35)|
|ALT < 60 IU/L||1.59 (0.33–7.73)|
|ALT ≥ 60 IU/L||0.36 (0.09–1.53)|
|AFP < 20 ng/mL||0.60 (0.14–2.59)|
|AFP ≥ 20 ng/mL||1.13 (0.23–5.56)|
|HCV(−)||AST < 60 IU/L||2.41 (1.14–5.10)|
|AST ≥ 60 IU/L||0.54 (0.06–4.55)|
|ALT < 60 IU/L||2.47 (1.12–5.45)|
|ALT ≥ 60 IU/L||0.82 (0.17–3.87)|
|AFP < 20 ng/mL||2.56 (1.15–5.71)|
|AFP ≥ 20 ng/mL||0.93 (0.18–4.74)|
By combining all the findings, it may be possible that diabetes not only plays an independent role in the metabolic pathway leading to hepatocarcinogenesis in the absence of HCV infection, but also is a cofactor for subsequent progression to HCC among patients with chronic hepatitis C who have advanced fibrosis or cirrhosis.