Published online May 28, 2009. doi: 10.3748/wjg.15.2506
Revised: April 29, 2009
Accepted: May 6, 2009
Published online: May 28, 2009
AIM: To explore the association between hepatocellular carcinoma (HCC) and type 2 diabetes mellitus, describe the temporal relations between the onset of diabetes and the development of HCC and evaluate the possible effects of antidiabetic therapy on HCC risk.
METHODS: We recruited 465 HCC patients, 618 with cirrhosis and 490 control subjects. We evaluated the odds ratio (OR) for HCC by univariate and multivariate analysis. Moreover, OR for HCC in diabetic subjects treated with insulin or sulphanylureas and with metformin were calculated.
RESULTS: The prevalence of diabetes mellitus was 31.2% in HCC, 23.3% in cirrhotic patients and 12.7% in the Control group. By univariate and multivariate analysis, the OR for HCC in diabetic patients were respectively 3.12 (CI 2.2-4.4, P < 0.001) and 2.2 (CI 1.2-4.4, P = 0.01). In 84.9% of cases, type 2 diabetes mellitus was present before the diagnosis of HCC. Moreover, we report an OR for HCC of 2.99 (CI 1.34-6.65, P = 0.007) in diabetic patients treated with insulin or sulphanylureas, and an OR of 0.33 (CI 0.1-0.7, P = 0.006) in diabetic patients treated with metformin.
CONCLUSION: Our study confirms that type 2 diabetes mellitus is an independent risk factor for HCC and pre-exists in the majority of HCC patients. Moreover, in male patients with type 2 diabetes mellitus, our data shows a direct association of HCC with insulin and sulphanylureas treatment and an inverse relationship with metformin therapy.
- Citation: Donadon V, Balbi M, Ghersetti M, Grazioli S, Perciaccante A, Della Valentina G, Gardenal R, Dal Mas M, Casarin P, Zanette G, Miranda C. Antidiabetic therapy and increased risk of hepatocellular carcinoma in chronic liver disease. World J Gastroenterol 2009; 15(20): 2506-2511
- URL: https://www.wjgnet.com/1007-9327/full/v15/i20/2506.htm
- DOI: http://dx.doi.org/10.3748/wjg.15.2506
Hepatocellular carcinoma (HCC) is the fifth most common malignancy worldwide and the third leading cause of cancer-related deaths. In recent years, a significant increase in HCC incidence and mortality rates has been observed in developed countries, but the causes of this growth are only partially understood. Hepatitis C virus (HCV) epidemics certainly play a role, due to the cohort effect of individuals infected in pre-serological age. Although the main risk factors for HCC are HCV, hepatitis B virus (HBV) and chronic alcohol abuse, at least 25% of HCC cases do not have any known etiology, suggesting that further risk factors could be responsible for the increasing incidence of HCC. Diabetes mellitus has recently been proposed as a risk factor for HCC. During the past two decades, the prevalence of diabetes mellitus, and in particular of type 2 diabetes mellitus, has dramatically increased in many countries, including Italy. Sedentary lifestyles, excessive food consumption and obesity appear to be the main causes of the current diabetes mellitus epidemic in western world.
Previous studies on the association between diabetes mellitus and liver diseases showed that type 2 diabetes mellitus appears to be a cause of non-alcoholic fatty liver disease (NAFLD) and that cirrhosis and HCV infection increase the susceptibility to diabetes mellitus.
Moreover, conflicting results were reported on the association between diabetes mellitus and solid tumors, in particular HCC[38–12]. While earlier investigations did not report any association between diabetes mellitus and HCC, recent data clearly indicate that diabetes mellitus is a risk factor of HCC[13–15].
However, the precise relation between diabetes mellitus and chronic liver diseases still needs to be further investigated. Therefore, the aims of this study are to explore the association between HCC and diabetes mellitus in a large cohort of patients with HCC and to describe the temporal relationship between the onset of diabetes and the development of HCC. We also considered the clinical and metabolic characteristics of the patients with type 2 diabetes mellitus and HCC, as well their antidiabetic therapy.
This work has been carried out in accordance with the Declaration of Helsinki (2000) of the World Medical Association.
We performed a retrospective, case-control study on three groups of Caucasian individuals, attending the Liver Unit and Diabetic Clinic of 3rd Internal Medicine in the Pordenone General Hospital (Pordenone, Italy) from January 1994 to June 2006. The 3rd Internal Medicine of Pordenone Hospital is a tertiary referral centre for liver disease and diabetes mellitus. This study is a single centre investigation and all patients of the three groups studied were all afferent, directly diagnosed and followed-up in the 3rd Internal Medicine of Pordenone Hospital.
A series of patients with HCC was compared with two different groups: one consisted of patients with liver cirrhosis and the other one, the controls, included individuals who were treated in our Hospital for a wide spectrum of acute conditions.
This group comprised 465 consecutive patients with HCC, of which 398 cases (85.6%) were diagnosed by means of cytological or histological examination of hepatic focal lesions. The others (14.4%) were diagnosed according to the following acknowledged criteria: ultrasound examination (also by using micro-bubbles of sulphur hexafluoride as contrast dye in the last three years), α fetoprotein (AFP) > 400 ng/mL, computerized tomography scan and/or magnetic resonance imaging of the upper abdomen. Clinical data, biochemical parameters and the antidiabetic treatment were considered at the time of HCC diagnosis.
Patients with HCC were further divided in two subgroups: the follow-up (FU) and the clinically overt group (CO). The FU group comprised 305 patients with small, single hepatic tumors who received the diagnosis during a surveillance program of HCC in cirrhotic patients based on ultrasound examinations and AFP determinations every 3-6 mo. The clinically overt group (CO) comprised 160 cases with advanced, large size and symptomatic HCC at diagnosis.
We enrolled 618 patients with liver cirrhosis (LC), matched with HCC cases by age (± 5 years), gender, body mass index (BMI), transaminases, history of diabetes, prevalence of HBV and HCV infections, alcohol consumption and time of admission. These patients were admitted to our Hospital for diagnosis, staging or therapy of liver cirrhosis. Clinical data, biochemical parameters and antidiabetic treatment were considered at the time of recruitment.
According to Child’s classification of cirrhosis, patients were classified as follows: class A: 55.5%; B: 24.3% and C: 20.2%. In the cirrhotic patients, the presence of HCC was ruled out through ultrasound examinations, CT or MRI of the upper abdomen and AFP checks.
From 28 740 in-patients of our region, 490 subjects matched with HCC and LC patients by age (± 5 years), gender, BMI, history of diabetes and time of admission were recruited. Those who were admitted for malignancies, alcohol-related disease, viral liver disease and diabetes mellitus were excluded from our study, although comorbidity of these conditions was not considered as an exclusion criterion. As previously reported, the selected control group represented our region’s general population as to HCV and HBV infections, alcohol consumption and diabetes mellitus prevalence in the age group over 65 years.
The demographic, clinical and biochemical data of each patient were collected in a computerized database. Biochemical parameters were determined at the Pordenone Hospital central laboratory using standardized and validated methods.
Hepatitis B surface antigen (HBsAg), anti-HBV surface antigen (anti-HBs), anti-HBV core antigen (anti-HBc), and hepatitis B e antigen (HBeAg) were determined using commercial assays (Abbott Diagnostic Division, Wiesbaden; Germany). Sera were also screened for antibodies against HCV (anti-HCV) using a third-generation micro particle enzyme immunoassay (AxSYM HCV version 3.0, Abbott Diagnostic Division). Positive samples were tested for anti-HCV using a third-generation line immunoassay (Immunogenetics, Gent, Belgium) and for serum HCV-RNA using the Roche Amplicor version 2.0 (Roche Molecular System, Pleasanton, CA).
The diagnosis and clinical classification of diabetes mellitus were based on the guidelines of the American Diabetes Association. In particular, the distinction between type 1 and 2 diabetes mellitus, was made according to the following clinical characteristics: age and modality of glucose intolerance onset, previous use of antidiabetes medications, occurrence of ketoacidosis, obesity and body fat distribution, concomitant autoimmunity, positive family history and HLA association, presence of micro- and macro-vascular complications when diabetes was diagnosed.
Alcohol abuse was defined as a daily consumption over 30 g for males and over 20 g for females. Average alcohol content was estimated in 5% for beer, 12% for wine and 40% for super alcoholics.
Normality tests were performed on all data. Parametric data are expressed as mean ± SD. Data with multiple time points variables were analysed by the general model ANOVA. Post hoc multiple comparisons were performed using an LSD test when ANOVA testing was significant (P ≤ 0.05). To establish univariate associations among variables, the odds ratio (OR) with a confidence interval of 95% was calculated, using the simple analysis of the logistic regression. All statistical analyses were performed using SPSS software 13.0 for Windows (SPSS Inc, Chicago, IL).
Each subject with diabetes mellitus in the HCC and LC groups showed the clinical and metabolic characteristics of type 2 diabetes mellitus. Of note, none of our HCC or liver cirrhosis patients had type 1 diabetes mellitus.
As shown in Table 1, the prevalence of type 2 diabetes mellitus was 31.2% in the HCC group, 23.3% in the LC group and 12.7% in the controls. The prevalences of type 2 diabetes mellitus in the three groups were statistically different, with an OR of 3.12 (CI 2.2-4.4) in HCC group vs controls and an OR of 2.09 (CI 1.5-2.9) in LC vs controls group.
|Subjects (n)||DM2 -ve n (%)||DM2 +ve n (%)||OR (95% CI)||P|
|HCC(465)||320 (68.8)||145 (31.2)||3.12 (2.2-4.4)a||< 0.001|
|Controls (490)||428 (87.3)||62 (12.7)||2.09 (1.5-2.9)c||< 0.001|
|LC (618)||474 (76.7)||144 (23.3)|
|HCC (364)||246 (67.6)||118 (32.4)||3.14 (2.1-4.6)b||< 0.00001|
|Controls (385)||334 (86.7)||51 (13.3)||1.99 (1.3-2.9)d||0.000|
|LC (450)||341 (75.8)||109 (24.2)|
|HCC (101)||74 (73.3)||27 (26.7)||3.11 (1.3-7.4)e||0.002|
|Controls (105)||94 (89.5)||11 (10.5)||2.59 (1.2-5.9)f||0.008|
|LC (168)||133 (79.2)||35 (20.8)|
Etiology of the chronic liver disease, mean age at HCC diagnosis and type 2 diabetes mellitus prevalence in the subgroups are summarized in Table 2.
|Etiology||HCC n (%)||Age (yr)||Prevalence of DM2 (%)|
|HBV||20 (4.3)||63.3 ± 10.3a||3 (15.0)|
|HCV||177 (38.1)||71.5 ± 7.3a||47 (26.5)bc|
|Alcohol||141 (30.4)||66.7 ± 8.5a||52 (36.9)bd|
|HBV + HCV||8 (1.7)||60.8 ± 12.8a||2 (25.0)|
|HBV + alcohol||9 (1.9)||62.9 ± 9.3a||2 (22.2)|
|HCV + alcohol||81 (17.4)||67.7 ± 9.3a||27 (33.3)|
|HBV + HCV + alcohol||2 (0.4)||68.4 ± 10.3||0|
|Cryptogenic||27 (5.8)||68.6 ± 9.3||19 (70.3)cd|
The highest prevalence (70.3%) of type 2 diabetes mellitus was found in HCC patients with cryptogenic chronic liver disease.
We also calculated that the male/female ratio in patients with cryptogenic etiology of HCC was 9.5/1, compared to a male excess of 4.4 in all diabetic patients and of 3.3 in the entire HCC group.
The multivariate analysis in HCC group vs Controls shows that type 2 diabetes mellitus is associated with an increased risk of HCC occurrence (OR = 2.2; CI 1.2-4.4; P = 0.01), independent of age, gender, BMI, alcohol abuse, HBV and HCV infections.
The data collected in the records of our Diabetes Clinic show that in 122 patients (84.9%), type 2 diabetes mellitus was diagnosed at least 6 mo before the onset of HCC, while in only 23 patients (15.1%) diabetes mellitus was recognized after the diagnosis of HCC.
The time interval between diabetes mellitus diagnosis and HCC onset was exactly calculated: diabetes was found to be present prior to the HCC diagnosis for a mean time of 141.5 ± 9.4 mo.
Moreover, in patients with HCC of cryptogenic etiology, diabetes mellitus was present before HCC occurrence for a mean period of 150.5 ± 10.1 mo.
In the subgroup with pre-existing type 2 diabetes mellitus, the time interval until the diagnosis of HCC was longer in insulin treated patients than in those treated with antidiabetic oral agents (171.5 ± 87.6 mo vs 118.7 ± 95.2 mo; P = 0.05).
The percentage of type 2 diabetes mellitus patients treated with insulin was similar in the HCC (39.6%) and LC groups (43%), but it was significantly lower in the controls (20.9%; P < 0.025 and P = 0.007, respectively) (Table 3). Based on the records of our Diabetic Clinic, the mean duration of insulin treatment in HCC insulin-treated patients was 83.1 ± 63.4 mo; before insulin therapy the patients were treated only with diet.
We observed that a significant number of HCC patients treated with antidiabetic oral agents took sulphanylureas rather than metformin, while most LC patients used metformin (Table 3). The mean period of treatment with the two antidiabetic oral agents was 125.2 ± 11.5 mo.
Univariate analysis in diabetic HCC patients vs diabetic controls shows that the OR for hepatocarcinoma in subjects taking insulin or sulphanylureas was 2.99 (CI 1.34-6.65, P = 0.007) while the OR dropped to 0.33 (CI 0.1-0.7, P = 0.006) in individuals treated with metformin.
The FU and CO groups showed a similar prevalence of diabetes (30.2% and 33.1%; P = 0.62, respectively). Mean glycated A1c hemoglobin was higher in the FU group (8.2% ± 2.8% vs 7.1% ± 2.1%), although this difference is not statistically significant (P = 0.08). The mean time interval between type 2 diabetes mellitus diagnosis and HCC onset was greater in CO than in FU patients (167.1 ± 114.3 mo vs 127.8 ± 80.1 mo; P = 0.03). Insulin treatment was more frequent in FU cases with type 2 diabetes mellitus than in diabetic CO patients (48.9% vs 28.3%; P = 0.01, respectively).
Our study shows that type 2 diabetes mellitus is an independent risk factor for hepatocarcinoma and precedes the onset of HCC. Moreover, we found that insulin or secretagogues antidiabetic oral agents treatment are associated with an increased risk for HCC, while in metformin treated patients the risk of HCC was reduced.
In this study, every HCC and cirrhotic patient with abnormal glucose tolerance showed clinical and pathophysiological characteristics of type 2 diabetes mellitus. This observation confirms the results of two recent prospective studies on the association between diabetes and HCC. El Seragh et al in a wide study carried out in the US reported that almost all (99%) diabetic patients who developed HCC had type 2 diabetes mellitus, and Lai et al in a survey in Taiwan, observed that type 2 diabetes mellitus, classified according to age at onset of diabetes mellitus, is associated with HCC.
The results of our study are consistent with the theory of the biological mechanisms underlying the epidemiological association between diabetes mellitus and cancer. In fact, this hypothesis postulates that the diabetes-cancer association is likely to be related to insulin resistance and consequent hyperinsulinemia, which are typical features in the majority of type 2 diabetes mellitus patients. Ten years ago, McKeown-Eyssen and Giovannucci observed that risk factors for cancer and insulin resistance in developed countries are almost the same. To explain this analogy, they suggested that protracted exposure to hyperinsulinemia increases the levels of IGF-1, which plays a pivotal role in carcinogenesis (insulin-cancer hypothesis). In addition, the predictive value of hyperinsulinemia on total cancer mortality and fatal liver tumor incidence has been demonstrated in non-diabetic subjects by two recent prospective analysis.
Liver cirrhosis is a significant cause of death in type 2 diabetes mellitus patients, being even more relevant than cardiovascular diseases. However, patients with type 2 diabetes mellitus often suffer from liver disease as well, and diabetes is a recognized cause of NAFLD and cryptogenic cirrhosis. In fact, it is well-known that the natural history of NAFLD might progress, over a period of many years, from steatosis to steatohepatitis, cirrhosis and, sometimes, to HCC. On the other hand, 20% of patients with cirrhosis have overt diabetes (hepatogenous diabetes) and 60% have impaired glucose tolerance. Thus, the association between diabetes and cirrhosis is complex and reciprocal.
To evaluate the relations between type 2 diabetes mellitus and HCC, regardless of cirrhosis and other risk factors, we performed a single centre, retrospective case-control study on HCC patients comparing them, not only to a group of Control subjects without liver diseases and diabetes mellitus, but also to a series of cirrhotic patients.
Our study shows that the prevalence of type 2 diabetes mellitus in the LC group is intermediate between those of the HCC and controls, indicating that the underlying liver cirrhosis is not the only cause of diabetes in HCC patients.
The similar prevalence of type 2 diabetes mellitus reported in FU and CO subgroups of HCC patients, suggests that the prevalence of diabetes mellitus is not dependent on the size of the liver tumor.
The evidence that type 2 diabetes mellitus is a risk factor for HCC in our patients is obtainable by univariate and multivariate analyses, which show an OR for HCC of 3.12 and 2.2 respectively, similar to those recently reported in U.S. and Asian populations. Moreover, like the results of a recent Japanese investigation, the patients in our study with cryptogenic HCC have the highest percentage of type 2 diabetes mellitus. Interestingly, most patients with cryptogenic HCC are male, suggesting that this sex prevalence could be related to the effect of diabetes, because the prevalence of type 2 diabetes mellitus, as shown by the current diabetes estimates based on age and sex in the populations of our region and western countries, is higher in men than in women in the age groups in which HCC develops.
The precise temporal relation between the onset of type 2 diabetes mellitus and diagnosis of HCC is only partially understood. A previous prospective study, conducted in a large cohort of males with and without diabetes mellitus, investigated, for the first time, the temporal relationship between diabetes and HCC, showing a two-fold increase of HCC incidence among patients with diabetes.
Our study showed that type 2 diabetes mellitus was present 141 mo before the diagnosis of HCC. Interestingly enough, we observed that in cryptogenic HCC patients, type 2 diabetes mellitus is the only recognized risk factor that these patients have 150 mo before the diagnosis. As already mentioned, patients with small and single hepatocarcinomas and those with advanced tumors have a similar prevalence of type 2 diabetes mellitus. Associated with this, the higher frequency of insulin treatment in FU cases with type 2 diabetes mellitus than in CO diabetic cases, suggests that type 2 diabetes mellitus is more likely to be a cause rather than merely a consequence of the liver cancer.
Our data demonstrates that exogenous insulin treatment is significantly more frequent in diabetic HCC patients than in Control diabetic cases, while cirrhotic patients are more frequently treated with metformin than HCC and Controls individuals. Moreover, in our study, insulin or sulphanylureas therapy are associated with an increased risk of HCC, conversely metformin therapy is associated with a reduced risk of HCC.
These findings are in agreement with previous studies on the relation between antidiabetic therapy and cancer-related mortality, which showed that diabetic patients treated with insulin or sulphanylureas have a significantly high mortality for cirrhosis and HCC, while treatment with insulin-sensitizer drugs that ameliorate insulin action, such as metformin, might have a protective effect on cancer risk.
The effects of insulin therapy in type 2 diabetes mellitus patients with chronic liver disease could be explained by the mitogenic action of exogenous insulin added to those of endogenous hyperinsulinemia. The sulphanylureas might cause an increase both of endogenous insulin secretion and of its precursors, that seem to have mitogenic effects by themselves. Conversely, metformin treatment can decrease insulin resistance, reducing the consequent hyperinsulinemia and its effects in these patients. Moreover, it is known that patients with type 2 diabetes mellitus treated with insulin frequently have more severe insulin resistance, hyperinsulinemia and diabetic complications. Therefore, insulin treatment might be a marker of long lasting and more severe diabetes.
The results of our study, therefore, might have important implications in the clinical management of diabetes mellitus, particularly in males with type 2 diabetes mellitus and chronic liver diseases, because we found that they are at high-risk for HCC. This observation is of primary relevance to the implementation of prevention policies and to encourage the most adequate and cost-effective programs of surveillance in cirrhotic patients. In fact, metformin treatment might have a protective effect and therefore might be recommended as a first-line therapy in diabetic patients with compensated liver cirrhosis, because its use is not associated with the risk of hypoglycaemia and body weight gain.
Thus, our data suggests that patients with type 2 diabetes and chronic liver disease should first control their diabetes mellitus through diet and changes in lifestyle, to decrease their weight and increase physical activity. Subsequently they should take metformin or other insulin sensitizers, to counteract insulin resistance and consequent hyperinsulinemia.
Our study did have some limitations. It is a retrospective study drawn from a clinical series and not from the community; however, in this case-control survey, matching by age, gender, history of diabetes, BMI and time of hospital admission, we selected a control group and we made sure that it represented the general population of our Region. A potential bias in a case-control study like this is discerning temporal relationships between exposure and outcomes, due to the complex relationships between type 2 diabetes mellitus and cirrhosis. To avoid this error, our study was conducted on a large cohort of HCC patients comparing them with both a control group and with a cohort of cirrhotic patients.
Furthermore, a retrospective study like this must be based upon complete and accurate information on the time interval between type 2 diabetes mellitus onset and the diagnosis of HCC. To achieve this, we reviewed all medical documentation kept at the Diabetes Clinic where all the Regional records are filed from the onset of the disease. Therefore, we could review the detailed clinical history of HCC diabetic patients to calculate exactly the individual time interval from the onset of diabetes to the HCC diagnosis. Prospective studies are required to demonstrate that insulin resistance and hyperinsulinemia are the biological mechanisms that explain the association between type 2 diabetes mellitus and HCC. In conclusion, our survey confirmed that type 2 diabetes mellitus is an independent risk factor for HCC and that it precedes HCC diagnosis. Moreover, in male patients with type 2 diabetes mellitus, our data show a direct association of HCC with insulin and sulphanylureas treatment and an inverse relationship with metformin therapy.
Type 2 diabetes mellitus has been associated with hepatocellular carcinoma (HCC). However, the relationship between type 2 diabetes mellitus and the underlying liver cirrhosis, and the effects of antidiabetic therapy on HCC risk have not yet been fully evaluated.
This study demonstrates that type 2 diabetes mellitus is an independent risk factor for HCC and pre-exists in the majority of HCC patients. In male HCC patients with type 2 diabetes mellitus, their data shows a direct association of HCC risk with insulin and sulphanylureas treatment and an inverse relationship with metformin therapy.
This is an interesting and detailed retrospective single center study reviewing patients with diabetes mellitus and chronic liver disease.
|1.||World Health Organization. Mortality database. //www.who.int/whosis/whostat/EN_WHS08_Part1.pdf. Accessed 4 December: Available from: URL: http 2008; .|
|2.||Armstrong GL, Alter MJ, McQuillan GM, Margolis HS. The past incidence of hepatitis C virus infection: implications for the future burden of chronic liver disease in the United States. Hepatology. 2000;31:777-782.|
|3.||Davila JA, Morgan RO, Shaib Y, McGlynn KA, El-Serag HB. Diabetes increases the risk of hepatocellular carcinoma in the United States: a population based case control study. Gut. 2005;54:533-539.|
|4.||Bonadonna RC, Cucinotta D, Fedele D, Riccardi G, Tiengo A. The metabolic syndrome is a risk indicator of microvascular and macrovascular complications in diabetes: results from Metascreen, a multicenter diabetes clinic-based survey. Diabetes Care. 2006;29:2701-2707.|
|5.||Sloan FA, Bethel MA, Ruiz D Jr, Shea AM, Feinglos MN. The growing burden of diabetes mellitus in the US elderly population. Arch Intern Med. 2008;168:192-199; discussion 199.|
|6.||Bugianesi E, Leone N, Vanni E, Marchesini G, Brunello F, Carucci P, Musso A, De Paolis P, Capussotti L, Salizzoni M. Expanding the natural history of nonalcoholic steatohepatitis: from cryptogenic cirrhosis to hepatocellular carcinoma. Gastroenterology. 2002;123:134-140.|
|7.||Moscatiello S, Manini R, Marchesini G. Diabetes and liver disease: an ominous association. Nutr Metab Cardiovasc Dis. 2007;17:63-70.|
|8.||El-Serag HB, Richardson PA, Everhart JE. The role of diabetes in hepatocellular carcinoma: a case-control study among United States Veterans. Am J Gastroenterol. 2001;96:2462-2467.|
|9.||Verlato G, Zoppini G, Bonora E, Muggeo M. Mortality from site-specific malignancies in type 2 diabetic patients from Verona. Diabetes Care. 2003;26:1047-1051.|
|10.||Coughlin SS, Calle EE, Teras LR, Petrelli J, Thun MJ. Diabetes mellitus as a predictor of cancer mortality in a large cohort of US adults. Am J Epidemiol. 2004;159:1160-1167.|
|11.||El-Serag HB, Tran T, Everhart JE. Diabetes increases the risk of chronic liver disease and hepatocellular carcinoma. Gastroenterology. 2004;126:460-468.|
|12.||Chen CL, Yang HI, Yang WS, Liu CJ, Chen PJ, You SL, Wang LY, Sun CA, Lu SN, Chen DS. Metabolic factors and risk of hepatocellular carcinoma by chronic hepatitis B/C infection: a follow-up study in Taiwan. Gastroenterology. 2008;135:111-121.|
|13.||Inoue M, Iwasaki M, Otani T, Sasazuki S, Noda M, Tsugane S. Diabetes mellitus and the risk of cancer: results from a large-scale population-based cohort study in Japan. Arch Intern Med. 2006;166:1871-1877.|
|14.||Lai MS, Hsieh MS, Chiu YH, Chen TH. Type 2 diabetes and hepatocellular carcinoma: A cohort study in high prevalence area of hepatitis virus infection. Hepatology. 2006;43:1295-1302.|
|15.||Donadon V, Balbi M, Casarin P, Vario A, Alberti A. Association between hepatocellular carcinoma and type 2 diabetes mellitus in Italy: potential role of insulin. World J Gastroenterol. 2008;14:5695-5700.|
|16.||Llovet JM, Beaugrand M. Hepatocellular carcinoma: present status and future prospects. J Hepatol. 2003;38 Suppl 1:S136-S149.|
|17.||National Diabetes Data Group. Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance. National Diabetes Data Group. Diabetes. 1979;28:1039-1057.|
|18.||American Diabetes Association. Report of the expert committee on the diagnosis and classification of diabetes mellitus. Diabetes Care. 2003;26 Suppl 1:S5-S20.|
|19.||McKeown-Eyssen G. Epidemiology of colorectal cancer revisited: are serum triglycerides and/or plasma glucose associated with risk? Cancer Epidemiol Biomarkers Prev. 1994;3:687-695.|
|20.||Giovannucci E. Insulin and colon cancer. Cancer Causes Control. 1995;6:164-179.|
|21.||Giovannucci E. Nutrition, insulin, insulin-like growth factors and cancer. Horm Metab Res. 2003;35:694-704.|
|22.||Balkau B, Kahn HS, Courbon D, Eschwege E, Ducimetiere P. Hyperinsulinemia predicts fatal liver cancer but is inversely associated with fatal cancer at some other sites: the Paris Prospective Study. Diabetes Care. 2001;24:843-849.|
|23.||de Marco R, Locatelli F, Zoppini G, Verlato G, Bonora E, Muggeo M. Cause-specific mortality in type 2 diabetes. The Verona Diabetes Study. Diabetes Care. 1999;22:756-761.|
|24.||Tolman KG, Fonseca V, Dalpiaz A, Tan MH. Spectrum of liver disease in type 2 diabetes and management of patients with diabetes and liver disease. Diabetes Care. 2007;30:734-743.|
|25.||Takamatsu S, Noguchi N, Kudoh A, Nakamura N, Kawamura T, Teramoto K, Igari T, Arii S. Influence of risk factors for metabolic syndrome and non-alcoholic fatty liver disease on the progression and prognosis of hepatocellular carcinoma. Hepatogastroenterology. 2008;55:609-614.|
|26.||Pilotto L, Gaggioli A, Lo Noce C, Dima F, Palmieri L, Uguccioni M, Pede S, Giampaoli S, Vanuzzo D. [Diabetes in Italy: a public health problem]. Ital Heart J Suppl. 2004;5:480-486.|
|27.||Wild S, Roglic G, Green A, Sicree R, King H. Global prevalence of diabetes: estimates for the year 2000 and projections for 2030. Diabetes Care. 2004;27:1047-1053.|
|28.||Bowker SL, Majumdar SR, Veugelers P, Johnson JA. Increased cancer-related mortality for patients with type 2 diabetes who use sulfonylureas or insulin: Response to Farooki and Schneider. Diabetes Care. 2006;29:1990-1991.|
|29.||Evans JM, Donnelly LA, Emslie-Smith AM, Alessi DR, Morris AD. Metformin and reduced risk of cancer in diabetic patients. BMJ. 2005;330:1304-1305.|
|30.||Vazquez-Martin A, Oliveras-Ferraros C, Menendez JA. The antidiabetic drug metformin suppresses HER2 (erbB-2) oncoprotein overexpression via inhibition of the mTOR effector p70S6K1 in human breast carcinoma cells. Cell Cycle. 2009;8:88-96.|
|31.||Inchiostro S, Bertoli G, Zanette G, Donadon V. Evidence of higher insulin resistance in NIDDM patients with ischaemic heart disease. Diabetologia. 1994;37:597-603.|