1. f. 정 회 훈 Am J Gastroenterol 2012;107:46-52 Risk of Hepatocellular Carcinoma in Diabetic Patients and Risk Reduction Associated With Anti-Diabetic Therapy: A Population-Based Cohort Study Shih-Wei Lai, MD 1, 2, 7, Pei-Chun Chen, PhD, MSPH 3, 4, 7, Kuan-Fu Liao, MD, MS 5, Chih- Hsin Muo, MS 3, Cheng-Chieh Lin, MD, PhD 1, 2, 6 and Fung-Chang Sung, PhD, MPH 3, 4

2. Introduction HCC etiological factors –Hepatitis B and hepatitis C –Positive for HBS Ag and / or HCV Ab : more than 90 % Non-viral causes of HCC –Cirrhosis, fatty liver disease –Diabetes mellitus (DM), obesity –Hereditary hemochromatosis –Alcohol, smoking –Dietary and environmental exposures

3. Patients with DM are more prone to cancer Gut 2005;54:533-9

4. A study of site-specific cancer mortality in Asian Ann Oncol 2011;22:730-8

5. HCC in Taiwan –Most lethal cancer for decades –Mortality rate of 33.6 per 100,000 persons ( n = 7759) –2 nd leading cause of cancer deaths after lung cancer in 2009 DM –Prevalence increased from 11.3 % in 1987 to 24.9 % in 2005 –5 th leading cause of death in 2009

6. Southern Taiwanese community with dual hepatitis B and C –Cross-sectional and case control studies –DM is not a risk factor for HCC Gastroenterol 2010;105:624-31 Type 2 DM is a strong independent predictor of HCC –In southern Taiwan –Hazard ratio of 2.7 (95% CI = 1.7– 4.3) Cancer Epidemiol Biomarkers Prev 2009;18:2054-60 Lifestyle changes or anti-diabetic drugs –Prevent and reverse DM may reduce the risk of cancer

7. Metformin –Widely used anti-diabetic drug –Recently great attention for lowering cancer risk –Inhibit cancer cell growth in vitro and in vivo –Reduced risks of breast cancer and HCC –Odds ratio for HCC in diabetic dropped to 0.3 –Metformin or thiazolidinediones 70 % reduction in HCC risk among diabetic patients Conducted in the United States also Limitation : small clinical samples of HCC –DM medication in reducing risk of HCC among Asian populations Only one cohort study, also from Taiwan, Only metformin was studied

8. Cancer 2010;116:1938-46

9. Aims To clarify the role of diabetes in risk of developing HCC –Population-based cohort study –Large-size data set available from National Health Insurance program in Taiwan Risk of HCC –Presence of hepatitis B and / or hepatitis C HCC risk is reduced with DM therapies –Metformin and thiazolidinediones

10. METHODS

11. Data sources National Health Insurance program in Taiwan –Covering more than 99 % of population Data –Available since 1996 –1 million subjects randomly selected in 1996 – 2000 –Registration of insurance –Inpatient and outpatient claims –Prescribed drugs –Basic sociodemographic information Sex and birth date

12. Design Retrospective cohort study –Diabetic patients group and a matched non-diabetic group Diabetic group –DM who had been prescrived anti-diabetic drugs in 2000 – 2005 –ICD-9 code 250.xx and Acode A-181 Exclusion –Younger than 20 years old –Diagnosed with cancer before the index date

13. Index date for the diabetic patients –Date of diagnosis –Same index month as their matched diabetic patients Other comorbidities presented before the index date –Obesity –Cirrhosis –Alcoholic liver damage –Nonalcoholic fatty liver disease –Hereditary hemochromatosis –Hepatitis B, hepatitis C

14. Both diabetic and non-diabetic groups were followed up to determine the incidence of HCC End point –Until the end of 2008 –Death –Withdrawal from the insurance program –Loss to follow-up consultations Confirmed the occurrence of HCC –To reduce the likelihood of misclassification of the disease –Pathological reports –Laboratory and image studies

15. Statistical analysis χ 2 -Tests and Student ’ s t -tests –Compare the differences between DM and non-diabetic group Incidence rate –As the number of HCC cases Cox proportional hazard models –Estimate HR with 95 % CI –Association between diabetes and the risk of developing HCC –Risk of developing HCC associated with comorbidities –HCC risks associated with DM medications SAS soft ware version 9.1 Statistical significance level : P < 0.05.

16. RESULTS

17. *Incidence of HCC by subject characteristics 2000~2005

19. *HCC associated with comorbidities

20. *Interaction between diabetes and comorbidities

21. *Anti-diabetic drugs and HCC risk reduction

22. Conclusion

23. DM have a higher risk of developing HCC Cirrhosis, hepatitis B, hepatitis C –Significantly aggravate the risk of developing HCC Hepatitis C and cirrhosis –Synergism –Markedly elevated risk of developing HCC –New insights into HCC prevention –Screening of high-risk patients

24. Metformin or thiazolidinediones –Reduced risks of developing HCC Further investigations –These insulin sensitizers have similar effect on HCC development among diabetic subgroups of liver comorbidities.