Nonselective Beta-blockers Are Associated With a Lower Risk of Hepatocellular Carcinoma Among Cirrhotic Patients in the United States

Karn Wijarnpreecha; Fang Li; Yang Xiang; Xun Xu; Cong Zhu; Vahed Maroufy; Qing Wang; Wei Tao; Yifang Dang; Huy Anh Pham; Yujia Zhou; Jianfu Li; Xinyuan Zhang; Hua Xu; C. Burcin Taner; Liu Yang; Cui Tao


Aliment Pharmacol Ther. 2021;54(4):481-492. 

In This Article


This study demonstrated the significant association between NSBBs and the decreased risk of HCC among patients with cirrhosis compared with the no beta-blocker group using large-scale EHR data. Recent studies demonstrated that NSBBs can reduce the progression of liver cirrhosis and might partially reduce the development of HCC.[3,25,26] However, all studies did not find a statistically significant difference in HCC between NSBB and non-NSBB group.[3,25,26] It is difficult to estimate the direct carcinogenic protective effect of HCC from NSBBs in those studies. Thus, we performed subgroup analysis in cirrhosis with complications (including ascites, hepatic encephalopathy, esophageal varices, hepatorenal syndrome, portal hypertension, and spontaneous bacterial peritonitis) and found that the HCC protective effect of NSBBs still exists in decompensated cirrhosis. This suggests that the HCC protective effect of NSBBs is not only dependent on the stage of liver cirrhosis and likely operates via the direct effect of reducing carcinogenesis. Regarding the subgroup analysis among the etiology of liver cirrhosis (alcohol-related liver cirrhosis or non-alcoholic cirrhosis), or cirrhosis with complications status, NSBBs showed the protective effect of HCC when compared with no beta-blockers group. It should be noted that nadolol is the only NSBB that has HCC protective effect among patients with alcohol-related liver cirrhosis in subgroup analysis. Thus, the finding of this study supports the use of NSBBs, which has the potential additional benefit of HCC prevention compared with no beta-blockers.

NSBBs remained significantly protective effect of HCC among cirrhosis regardless of age, sex, ethnicity, or comorbidity status as shown in Table 3. Overall, effect modification shows that carvedilol, nadolol, and propranolol were associated with significantly decreased risk of HCC among patients with cirrhosis regardless of sex, age, comorbidities, or etiology of liver cirrhosis except for viral hepatitis B, which was likely due to limited sample size. Of note, the effect modification supports the use of carvedilol in Hispanics and African Americans since it significantly decreased HCC risk among these groups, which is not seen in the nadolol or propranolol group. A study reported that carvedilol has been increasingly used in clinical practice and has a greater portal pressure-reducing effect than propranolol and is safe to use in both compensated and decompensated cirrhosis.[27] Previous small studies reported improved survival in patients with liver cirrhosis who used carvedilol.[3,26] However, they did not find a statistically significant difference in HCC incidence between the carvedilol vs non-carvedilol group, likely due to the limited sample size.[3,26] The current study with a larger sample size addressed this limitation and demonstrated that patients with liver cirrhosis who take NSBBs had less incidence of HCC, which may benefit long-term survival. The current study also suggested that carvedilol may be the preferred NSBBs option for prevention of esophageal variceal bleeding or complications from cirrhosis given significantly additional benefit from decreased risk of HCC incidence compared with nadolol or propranolol.[28]

Although the exact mechanism that explains the reduced risk of HCC in NSBBs is not fully understood, there are plausible explanations for this association. A recent study demonstrated that propranolol was able to inhibit proliferation and promote apoptosis in liver cancer cells.[29] It has been suggested that the β2-adrenoreceptor is involved in the pathophysiology of liver cancer.[30–33] In addition, NSBBs have been shown to inhibit catecholamine-driven cancer cell migration, tumor angiogenesis, proliferation, and have cytotoxic properties for liver cancer cells.[29,34] It has been shown that carvedilol increased cytosolic free Ca2+ concentration in hepatoma cells by releasing Ca2+ from the endoplasmic reticulum and consequently causing Ca2+ influx via store-operated Ca2+ channels.[35] Ca2+ overload is a well-known cause of cell death.[36] The mechanism of carvedilol-induced cytotoxicity in hepatoma cells was shown in the previous study mainly mediated by Ca2+ influx and evoked apoptosis and cell death in a concentration-dependent manner.[35] Moreover, an in vitro study showed apoptosis of human liver carcinoma was promoted by gold nanoparticle combined with carvedilol via mitogen-activated protein kinase (MAPK)/AKT/mTOR pathway.[37] In addition to the proapoptotic and cytotoxic effect on hepatoma cells, carvedilol inhibits angiogenesis through the VEGF-Src-ERK signaling pathway.[38] Studies demonstrated that VEGF plays a vital role in angiogenesis and prognosis of HCC[39] and VEGF inhibitors suppress HCC angiogenesis and induced cell death via inactivation of VEGF/phosphoinositide 3-kinase (PI3K)/protein kinase B (AKT) pathway.[40] Interestingly, a recent in vitro study demonstrated carvedilol can also be a novel chemopreventive agent to prevent breast cancer by inhibiting the reactive oxygen species-mediated PI3K/AKT pathway.[14] In the rat model of CCL4 induced liver injury, carvedilol reduced both hepatic oxidative stress and serum/tissue IL-6 level,[41] which are tumorigenic and able to induce malignant transformation in the liver. In fact, it has been shown that HCC patients had significantly higher IL-6 than cirrhotic patients without HCC or patients with hepatitis.[42] IL-6 not only plays a role in the pathogenesis of HCC but also is a prognostic indicator of HCC.[43] Thus, we hypothesise that carvedilol may also protect against HCC via suppressing the oxidative stress and IL-6 pathway.

The current study has several major strengths. First, this study investigates the risk of HCC in patients with cirrhosis at different stages and the impact of carvedilol vs no beta-blockers, using a large-scale patient dataset. All information on diseases and drug exposures was recorded in the absence of a study hypothesis. Therefore, the current study is not susceptible to recall bias. Second, we did adjust the odds of HCC with multiple variables including host factors, comedications, and metabolic risk factors and comorbidities to eliminate confounding effects. In addition, this study used meticulous statistical procedures to minimise confounding factors by using time-dependent variables and the Cox regression analysis model. Third, the study minimised the potential selection bias by performing the PSM to match multiple factors. Fourth, we evaluated the effect modification of the protective effect of NSBBs and the risk of HCC using multiple factors to ensure that the protective effect of NSBBs persisted after subgroup analysis. The protective effect of NSBBs was not modified by multiple factors as a statistically significantly decreased risk of HCC remained among subgroup populations.

This study also has certain limitations that should be considered. First, the current study is dependent on ICD-9 and ICD-10. We could not control the possibility of coding variability, which may lead to missed or incorrect reporting of certain conditions. Second, we could not adjust or control unmeasured confounding or uncontrolled factors that could not be obtained from the Cerner database. The Cerner database does not have complete health behaviors or environmental exposure that may affect the risk of HCC, such as coffee consumption, alcohol consumption, or active smoking.[44,45] Lastly, the current study does not estimate the direct carcinogenic protective effect of HCC from NSBBs. The direct effect of reducing HCC development by NSBBs could be demonstrated by stratifying patients into different severity stages of cirrhosis (Child-Pugh classification) to see the protective effect of HCC regardless of the severity stages of cirrhosis. However, the Child-Pugh classification contains the degree of subjective assessment, including the severity of ascites and the grade of encephalopathy, which could not be obtained from the current Cerner database. Thus, a future study may focus on investigating the direct protective carcinogenic protective effect of HCC from NSBBs by stratifying patients into different Child-Pugh classifications.