Objectives: Hepatocellular carcinoma is the most frequent primary malignant tumor of the liver and the third most common cause of all cancer-related mortalities. There is a need to develop new strategies to prevent hepatocellular carcinoma, as the incidence of this cancer continues to increase despite all advancements. In this study, our aim was to determine the effects of propranolol treatment on the incidence of hepatocellular carcinoma in cirrhotic patients waiting for liver transplant.
Materials and Methods: We retrospectively reviewed the data of patients waiting for liver transplant with cirrhosis due to various causes registered at the Hepatocellular Carcinoma Surveillance Program between June 2011 and December 2017 in our center. These data were compared between patients using propranolol and those not using propranolol.
Results: Of the 231 patients, 135 (58.4%) were male and 96 (41.6%) were female. The mean age was 58.1 ± 14 years. We noted that 153 of total patients (66.2%) were using propranolol. Three patients (2%) were using 20 mg propranolol, 125 (81.7%) were using 40 mg propranolol, 10 (6.5%) were using 60 mg propranolol, and 15 (9.8%) were using 80 mg propranolol. Of total patients, 36 (15.6%) developed hepatocellular carcinoma, including in 12 patients (7.8%) using propranolol and 24 patients (30.8%) who did not use this agent (P < .001). Thus, the hepatocellular carcinoma frequency was 5.22 times lower in patients receiving propranolol than in those not receiving propranolol.
Conclusions: Although causes of cirrhosis and initial stages were similar in both groups using and not using propranolol, incidence of hepatocellular carcinoma was significantly lower in the propranolol group than in the group without propranolol. This result showed that propranolol treatment has a protective effect for hepatocellular carcinoma in patients waiting for liver transplant with cirrhosis.
Key words : Beta blocker, Esophageal varices, Liver cancer
Currently, the definitive treatment for liver cirrhosis is liver transplant. The most common indications for liver transplant include decompensated cirrhosis and hepatocellular carcinoma (HCC) due to cirrhosis.1,2 Unfortunately, the inadequacy of the number of deceased donors in Turkey has resulted in prolonged periods on liver transplant wait lists for patients with cirrhosis. This condition has increased the risk of developing HCC with other complications of cirrhosis.
Hepatocellular carcinoma is the most frequent primary malignant tumor with an epithelial origin in the liver. Globally, it is the fifth most common cancer in men, the seventh most common cancer in woman, and the third most common cause of all cancer-related deaths. Hepatocellular carcinoma is a complex disease associated with many risk factors and causes. Hepatitis B and C viruses (HBV and HCV), chronic alcohol consumption, and nonalcoholic steatohepatitis (NASH) are the most common risk factors for HCC development.3 However, regardless of the cause, cirrhosis is the greatest risk factor for HCC development. The incidence of HCC has increased over the past 2 decades, and it is predicted that, if new preventive strategies cannot be developed, it will continue to increase, despite recent significant advances made to prevent HCC.4
Although the mechanisms of action are not precisely known, both experimental and human studies have shown that beta-adrenergic activity contributes to the formation and development of tumors.5-10 The relationship between nonselective beta blockers (NSBBs) use and cancer has been extensively studied in melanoma, breast, prostate, ovarian, and colon cancers. Although NSBB use has been shown to reduce recurrence, metastasis, and mortality in many studies, conflicting results have been reported in some population-based cohort studies.11-15 The anti-angiogenic effects of NSBBs in hemangiomas are also well known.16 The number of studies evaluating the effect of NSBBs on HCC development in patients with cirrhosis is limited; therefore, further research investigations are needed.17-20 The aim of this study was to determine the effects of propranolol treatment on the incidence of HCC in patients with cirrhosis waiting for liver transplant.
Materials and Methods
We retrospectively reviewed the data of cirrhotic patients with various causes registered in our HCC Surveillance Program between June 2011 and December 2017 in our center. Patients under the age of 18 years, patients with deficient data, patients with impaired follow-up according to the HCC Surveillance Program, patients with HCC or suspected HCC, patients with any malignancy other than HCC, and patients using beta blockers other than propranolol were excluded from the study. The demographic information; clinical characteristics; hematologic, biochemical, and coagulation parameters; cirrhotic duration; Child-Pugh Turcotte (CPT) and the Model for End-Stage Liver Disease-sodium (MELD-Na) scores; imaging results; histopathologic findings; and deaths and causes of deaths were recorded for patients who met the required criteria. All data were collected until the end of the study (December 2017), exitus, or liver transplant. Patients were divided into 2 groups: propranolol users and nonusers. Data were compared between groups.
Endoscopic examinations were periodically performed on all patients after diagnosis of cirrhosis to determine the presence and stage of esophageal varices. According to the protocol applied at our center, if there were no contraindications, propranolol was initiated as a primary prophylaxis in patients with moderate to large esophageal varices. In addition, propranolol therapy was initiated either alone or in combination with endoscopic treatment as secondary prophylaxis in patients with at least 1 previous esophageal variceal bleeding. Propranolol doses were determined according to patients’ heart beats per minute, blood pressure values, and tolerances. For every patient that used propranolol, the dosage and usage period were recorded. The number of patients with HCC development and the rate of HCC development were compared between the 2 groups using and not using propranolol.
In accordance with our HCC Surveillance Program, patients with cirrhosis underwent alpha-fetoprotein (AFP) tests and transabdominal ultrasonography scans at 3- to 6-month intervals. Computed tomography and/or magnetic resonance imaging were performed in patients who were suspected of having HCC on the basis of ultrasonography and/or elevated AFP levels. Biopsies were performed on the lesion(s) in the liver that did not fulfill the radiologic criteria of HCC. The number of foci, lesion location, size of lesion, and time from the diagnosis of cirrhosis to the time of HCC development were recorded in each patient who developed HCC.
Statistical analyses were performed with SPSS software (SPSS: An IBM Company, version 20.0, IBM Corporation, Armonk, NY, USA). Chi-square tests, Fisher exact tests, t tests, and Mann-Whitney U tests were used for univariate analyses. For multivariate analysis, predictors of treatment outcomes were examined using logistic regression analysis of factors determined from previous analyses. Hosmer-Lemeshow test was used for model adaptation. Outcomes in which type 1 error levels were below 5% were considered statistically significant.
Between June 2011 and December 2017, there were 231 eligible study patients with cirrhosis due to various causes who were on our liver transplant wait list and followed in the HCC Surveillance Program. Cirrhosis causes are shown in Table 1. The most common causes of cirrhosis were HBV in 58 patients (25.1%), cryptogenic disease in 57 patients (24.7%), and NASH in 32 patients (13.9%). Table 2 summarizes overall patient demographic and clinic features; hematologic, biochemical, and coagulative parameters; duration of cirrhosis; and CPT and MELD-Na scores. Of total patients, 135 (58.4%) were male and 96 (41.6%) were female. The mean age was 58.1 ± 14 years, and the mean duration of cirrhosis was 7.7 ± 4 years. The mean CPT score was 7.1 ± 1.4, and the mean MELD-Na score was 14.3 ± 4.6. Smoking and alcohol consumption were documented in 50 (21%) and 27 patients (11.7%), respectively. There were 72 patients (31%) with diabetes mellitus.
Of the 231 total patients, 153 (66.2%) were using propranolol and 78 (33.8%) were not using propranolol. Table 2 shows the demographic and clinical characteristics; hematologic, biochemical, and coagulation parameters; cirrhotic durations; and CPT and MELD-Na scores of patients receiving and not receiving propranolol therapy. We observed no statistically significant differences between the 2 groups in terms of these features, except for mean value of platelets. The mean platelet count was 86.68 ± 46.72 in the propranolol group and 113.05 ± 57.08 in the group that did not use it (P < .001). As shown in Table 1, causes of cirrhosis were proportionally similar in patients who used or did not use propranolol. Mean duration of propranolol usage was 6.75 ± 3.47 years. Of patients who used propranolol, 3 (2%) used 20 mg propranolol, 125 (81.7%) used 40 mg propranolol, 10 (6.5%) used 60 mg propranolol, and 15 (9.8%) used 80 mg propranolol.
Table 3 shows the number of patients and rate of HCC development, the average time to development of HCC from cirrhosis, the average size of lesion or lesions, regardless of the total number of foci in each patient, the average number of foci, and HCC mortality rates over our 6.5-year period follow-up. Of the 231 total patients, 36 (15.6%) developed HCC. In patients who developed HCC, the mean time to HCC development was 6.5 ± 3.4 years. The mean number of HCC foci in these patients was 1.97 ± 1.79; 22 patients (61.1%) had single focus, 9 patients had ≤ 3 foci, 2 patients (5.6%) had ≤ 5 foci, and 3 patients (8.3%) had > 5 foci. Regardless of the number of HCC foci, the average size of the lesion or lesions was 5.17 ± 4.43 cm. Treatment of the 36 patients who developed HCC was as follows: 13 (36.1%) received transarterial chemoembolization (TACE), 8 (22.2%) received local ablation, 3 (8.3%) received TACE and local ablation combination, 10 (27.8%) received liver transplant, and 2 (5.6%) had surgical resection.
As shown in Table 3, HCC developed in 12/153 patients (7.8%) who used propranolol and in 24/78 (30.8%) who did not used propranolol (P < .001). This showed that HCC frequency was 5.22 times lower in patients who used propranolol than in those who did not use propranolol. Regardless of the number of foci, the mean size of total HCC lesion or lesions was 5.05 ± 4.86 cm in those who used propranolol and 5.23 ± 4.32 cm in the nonuser group. We observed no statistically significant differences between the 2 groups (P =.352). The average time from diagnosis of cirrhosis to development of HCC was 6.29 ± 3.6 years for propranolol users and 5.6 ± 3 years for nonusers. Again, we observed no statistically significant differences between the 2 groups (P = .65).At the end of the study, 75/231 study patients (32.5%) died. Of these, 24 patients (66.76%) with HCC development died, whereas 51 patients (26.15%) who did not develop HCC died. Numbers of deaths for propranolol user and nonuser groups were 42 (27.5%) and 33 (42.3%), respectively (P =.02).
In this study, we analyzed the effect of propranolol treatment on the incidence of HCC in 231 patients waiting for liver transplant who had cirrhosis due to various causes. Of 153 who used propranolol and 78 who did not use propranolol, the incidence of HCC was found to be 5.22 times lower in propranolol users than in nonusers (P < .001). This result shows that propranolol has a protective effect on HCC in cirrhotic patients waiting for liver transplant.
Hepatocellular carcinoma is the third most common cause of mortality due to all cancers after lung cancer and stomach cancer.3 Hepatocellular carcinoma is responsible for 90% of all malignant tumors of the liver, and its annual incidence is similar to the annual mortality rate.21 Every year, more than 600 000 people all over the world die because of HCC.22 In the past 20 years, HCC incidence has increased, and it is expected to continue to increase until 2030.4 One of the most important reasons for the increased incidence of HCC is that patients with cirrhosis remain on liver transplant wait lists for extended periods due to the insufficient number of deceased donors. Given the increased incidence of HCC, it is necessary to seek and implement preventive measures to reduce the burden of HCC and its cost to health care.
The use of the HBV vaccine, the universal measures taken to prevent HBV and HCV transmission by blood, and the effective antiviral treatments used in HBV and HCV infections have reduced the risk of HCC associated with chronic viral hepatitis; however, these measures have not eliminated it.23-26 Early diagnosis of alcohol liver disease and encouraging discontinuation of alcohol are the most important methods to prevent progression in hepatic fibrosis and HCC development.27 Weight loss and physical activity can reduce the risk of NASH progression and development of HCC.28 In addition, early diagnosis of hereditary hemochromatosis and treatment of iron overload in these patients can reduce HCC risk,29 and reducing aflatoxin exposure can decrease HCC risk.30 As a result, preventative measures against HCC have significantly improved over the past decade, especially as a result of effective primary measures and treatments of chronic viral liver diseases. However, these preventive measures do not completely remove risk of HCC, especially after cirrhosis has developed. Therefore, new effective strategies should be developed to prevent HCC development in patients with cirrhosis.
Nonselective beta blockers are routinely used in clinical practice for primary and/or secondary prophylaxis to prevent esophageal variceal bleeding in cirrhotic patients with moderate to large esophageal varices by decreasing the hepatic venous pressure gradient.31 It has also been reported that the effect of NSBBs on the hepatic venous pressure gradient may reduce the risk of developing acid and hepatorenal syndrome.32 Nonselective beta blockers have been reported to reduce the risk of spontaneous bacterial peritonitis, although this is not fully elucidated by the hemodynamic changes that NSBBs cause in cirrhotic patients.33 Perhaps most importantly, in a limited number of recent studies, NSBBs have been reported to reduce the risk of developing HCC in patients with cirrhosis.17-19 However, in addition to all of these benefits, NSBBs have recently been reported to increase risk of hepatorenal syndrome and shorten survival time, especially in decompensated cirrhotic patients with refractory ascites.34
There are many biochemical processes that are responsible for the development of HCC and at the same time may be targets for the prevention and/or stopping of HCC. It has been suggested that some intracellular biochemical and signaling pathways responsible for cancer formation are affected by NSBBs. These included the following: (1) prevention of lipid peroxidation of membranes; (2) weakening of mitochondrial dysfunction; (3) inhibition of Bax protein-mediated cytochrome C release, NADPH oxidase, and protein kinase C activity; (4) reduction of vascular endothelial growth factor/basic fibroblast growth factor gene expression; (5) antagonization of cell migration associated with catecholamines, angiogenesis, invasiveness, and proliferation of tumors; (6) decreased bacterial translocation and reduced partial burden of proinflammatory bacterial products; and (7) inhibition of cAMP responsive element binding protein, nuclear factor κB, and activator protein, induction of apoptosis, and reduction of matrix metalloproteinase activation and tumor angiogenesis.5-10
In the literature, 4 studies have evaluated the effects of NSBBs on HCC development in cirrhotic patients.17-20 In 2012, Nkontchou and associates reported for the first time in a retrospective cohort study of 291 cirrhotic patients with HCV that propranolol users had a lower incidence of HCC than those who did not use it.17 Patel and colleagues reported that NSBBs were associated with inhibition of HCC in a retrospective study of 182 patients with cirrhosis and portal hypertension associated with various causes.18 Herrera and associates reported that the likelihood of developing HCC was lower in NSBBs-treated patients than in nontreated patients in a study conducted in 2016 with 173 cirrhotic patients due to various causes.19 However, in 2012, Kim and associates reported that there was no relationship between NSBBs and the incidence of HCC in a study of 273 patients with cirrhosis of various causes.20 In summary, except for the study by Kim and associates, our study agreed with the 3 studies that reported that incidence of HCC was lower in NSSB-treated cirrhotic patients than in nontreated patients with cirrhosis.17-19 In addition, in a 2015 meta-analysis involving 23 randomized studies and 2618 patients with cirrhosis, Thiele and colleagues reported that NSBBs have the potential to reduce HCC incidence.35 The results of our study and similar recent studies support the use of NSBBs, which are easily available, affordable, and well tolerated by most patients, to prevent HCC in patients with cirrhosis.
The use of NSSBs for primary and/or secondary prophylaxis to prevent esophageal variceal bleeding in patients with cirrhosis has been shown to increase overall survival.36 In our study, the mortality rate was 27.5% in the propranolol group and 33% for nonusers (P = .02). The effect of propranolol use on prognosis and survival in patients who developed HCC could not be evaluated because of the low number of HCC patients in the propranolol group.
One of the most important limitations of our study is that it is retrospective. However, our center is a liver transplant center, with patient data fully protected in the computer environment within the HCC Surveillance Program. Another important limitation is that the effects of other treatments that reduce HCC risk were not included in our study. Furthermore, although most patients who received propranolol in our study used a daily dose of 40 mg, the effective dose of propranolol to prevent HCC could not be determined. Similarly, because the number of patients using propranolol and developing HCC was low, the role of propranolol in preventing the development of HCC could not be analyzed.
Although the cause of cirrhosis and stage of cirrhosis at start of analyses were similar in both groups, the incidence of HCC at the 6.5-year follow-up was significantly lower in the propranolol group than in the group not using it. This result shows that propranolol has an effect on preventing HCC formation in patients with cirrhosis waiting for liver transplant. We suggest that the role of NSBBs in HCC development is an important area for further investigations. More prospective and randomized controlled clinical trials are needed to determine the effective dose and duration of use of NSSBs for the prevention of HCC in patients with cirrhosis. In conclusion, we recommend the use of NSBBs to prevent HCC, which is likely to develop independently of the presence and stage of esophageal and/or gastric varices in cirrhotic patients with portal hypertension who are waiting for liver transplant.
DOI : 10.6002/ect.2018.0321
From the Department of Gastroenterology, Faculty of Medicine, Başkent
University, Ankara, Turkey
Acknowledgements: The authors have no sources of funding for this study and have no conflicts of interest to declare.
Corresponding author: Nuretdin Suna, Faculty of Medicine, Başkent University, Mareşal Fevzi Çakmak Cd. No: 45, Çankaya/Ankara 06490, Turkey
Phone: +90 312 2036868
Table 1. Cirrhosis Causes and Rates
Table 2. Demographic and Clinical Features
Table 3. Rates of Hepatocellular Carcinoma and Death