BACKGROUND AND AIMS: Diabetes mellitus is associated with various cancers; however, little is known of the relationship between cancer and diabetes in chronic liver disease (CLD) patients. The aim of this study is to investigate the morbidity and associated factors of cancer, including the use of anti-diabetics, in CLD patients with diabetes. PATIENTS AND METHODS: We performed a multicenter survey in 2012 and 478 CLD patients with diabetes were enrolled (age 64.3 ± 12.1 years, female/male 187/291). A frequency analysis of cancer and antidiabetic use was performed. Independent factors for cancer were analyzed using logistic regression and decision-tree analysis. RESULTS: The morbidity of cancer was 33.3%. Hepatocellular carcinoma (HCC) and extra-hepatic cancer were diagnosed in 24.7 and 11.3% of enrolled patients, respectively. The frequency of antidiabetic use was 66.5%. Of prescribed antidiabetics, 39% were dipeptidyl-peptidase 4 inhibitors; however, their use was not significantly associated with cancer. In contrast, the use of exogenous insulin (OR 2.21; 95% CI 1.16-4.21, P = 0.0165) and sulfonylurea (OR 2.08; 95% CI 1.05-3.97, P = 0.0353) were independently associated with HCC and extra-hepatic cancer, respectively. In decision-tree analysis, exogenous insulin and sulfonylurea were also identified as a divergence factor for HCC and extra-hepatic cancer, respectively. CONCLUSIONS: We found a high morbidity of not only HCC, but also extra-hepatic cancer in CLD patients with diabetes. We also showed a possible association between the use of antidiabetics and the morbidity of cancer. Thus, a large-scale cohort study is needed to establish a therapeutic strategy for diabetes to suppress carcinogenesis in CLD patients.
BACKGROUND AND AIMS: Diabetes mellitus is associated with various cancers; however, little is known of the relationship between cancer and diabetes in chronic liver disease (CLD) patients. The aim of this study is to investigate the morbidity and associated factors of cancer, including the use of anti-diabetics, in CLD patients with diabetes. PATIENTS AND METHODS: We performed a multicenter survey in 2012 and 478 CLD patients with diabetes were enrolled (age 64.3 ± 12.1 years, female/male 187/291). A frequency analysis of cancer and antidiabetic use was performed. Independent factors for cancer were analyzed using logistic regression and decision-tree analysis. RESULTS: The morbidity of cancer was 33.3%. Hepatocellular carcinoma (HCC) and extra-hepatic cancer were diagnosed in 24.7 and 11.3% of enrolled patients, respectively. The frequency of antidiabetic use was 66.5%. Of prescribed antidiabetics, 39% were dipeptidyl-peptidase 4 inhibitors; however, their use was not significantly associated with cancer. In contrast, the use of exogenous insulin (OR 2.21; 95% CI 1.16-4.21, P = 0.0165) and sulfonylurea (OR 2.08; 95% CI 1.05-3.97, P = 0.0353) were independently associated with HCC and extra-hepatic cancer, respectively. In decision-tree analysis, exogenous insulin and sulfonylurea were also identified as a divergence factor for HCC and extra-hepatic cancer, respectively. CONCLUSIONS: We found a high morbidity of not only HCC, but also extra-hepatic cancer in CLD patients with diabetes. We also showed a possible association between the use of antidiabetics and the morbidity of cancer. Thus, a large-scale cohort study is needed to establish a therapeutic strategy for diabetes to suppress carcinogenesis in CLD patients.