HCC in patients with NAFLD Introduction NAFLD Definition

HCC in patients with NAFLD Introduction • NAFLD • Definition • Presence of fat in the liver (hepatic steatosis) either on imaging or on liver histology after the exclusion of secondary causes of fat accumulation in the liver Term Description NAFLD Indicates fatty infiltration of the liver Defined as a fat concentration > 5 -10% of the liver weight NASH Hepatic steatosis with inflammation, ballooned hepatocytes, and/or fibrosis (which may progress to cirrhosis) Type of NAFLD 1 Simple steatosis (no inflammation or fibrosis) 2 Steatosis with lobular inflammation but no fibrosis or balloon cells 3 Steatosis, inflammation and fibrosis of varying degrees (NASH) 4 Steatosis, inflammation, ballooned cells, and Mallory’s hyaline or fibrosis (NASH) Ballooning and Mallory-Denk bodies perisinusoidal/pericellular fibrosis Puneet Puri et al. , Clinical liver disease, 2012 Yoshihisa Takahashi et al. , World J Gastroenterol. 2014

HCC in patients with NAFLD Introduction • NAFLD • Prevalence • • US (US) : 34% , US (CT) : 17% Japan (US) : 21. 8% Global : 25. 24% Metabolic comorbidities • • Obesity (51. 34%; 95% CI: 41. 38 -61. 20) Type 2 DM (22. 51%; 95% CI: 17. 92 -27. 89) HL (69. 16%; 95% CI: 49. 91 -83. 46%) HTN (39. 34%; 95% CI: 33. 15 -45. 88) Younossi ZM et al. , Hepatology, 2016 K Omagari et al. , Journal of gastroenterology and hepatology, 2002 Elizabeth K. Speliotes et al. , Hepatology, 2010 Kim D et al. , Hepatology, 2013

HCC in patients with NAFLD Introduction • NAFLD • Clinical course • Liver cirrhosis • • • Progression to cirrhosis in NASH Pts : 10% Median survival of decompensated NASH cirrhosis Pts : approx. 2 yr HCC • • Progression to cirrhosis in Pts with simple steatosis : 0 -4% NASH was found to be third most common risk factor for HCC in a U. S. veterans Non-liver related disease and death • NAFLD may be independently associated with predictors of CVD including endothelial dysfunction, arterial stiffness and myocardial dysfunction • NASH were twice as likely to die from CVD compared to the reference general population Greater proportions of patients with HCC without evidence of cirrhosis had NAFLD Proportion of HCC patients with or without evidence of cirrhosis by risk factor Matteoni C. A. et al. , Gastroenterology, 1999 Mittal S. et al. , Clin. Gastroenterol. Hepatol. 2015 Younossi, Z. M et al. , Hepatology, 2011 Luis Calzadilla Bertot et al. , Int. J. Mol. Sci, 2016

HCC in patients with NAFLD

Risk of HCC in Patients with NAFLD Methods Results Discussions • Data source • National VHA Corporate Data Warehouse (CDW) and Central Cancer Registry (CCR) • CDW : ICD codes, AUDIT-C, laboratory results • CCR : date of diagnosis, primary tumor site, histology

Risk of HCC in Patients with NAFLD Methods Results Discussions • Study cohorts • ≥ 18 yr • At least 1 visit to VHA hospitals (2003. 1. 1 ~ 2008. 12. 31) • NAFLD cohorts • NAFLD definition • ≥ 2 elevated ALT, ≥ 6 months apart • HBs. Ag (-), HCV RNA (-) • Exclusion • Excess alcohol use • Alcohol related ICD-9 codes • AUDIT-C score (+) • Evidence of rare chronic hepatitis Ø Index date (1 st elevated ALT) 2004. 1. 1 ~ 2008. 12. 31 • Control cohorts ALT (2004. 1. 1 ~ 2008. 12. 31) Persistently normal ALT No NAFLD HBs. Ag (-), HCV RNA (-) No evidence of excess alcohol use No any other documented liver related risk factor Ø Random sampling to select a control cohort matched to NAFLD cases on § Gender § Age at 1 st ALT § Duration from 1 st VA visit to 1 st ALT test date • • •

Risk of HCC in Patients with NAFLD Methods • Results Discussions ICD-9 code ICD-O 3 rd edition, 2001 Diagnostic Code Descriptions (ICD-9) - Province of British Columbia Shah AG et al. , Clin Gastroenterol Hepatol, 2009

Risk of HCC in Patients with NAFLD Methods Results Discussions • Statistical analysis • Primary analysis • Annual & cumulative incidence rates for HCC in NAFLD & control cohorts • Multivariable model : adjusting for variables of race/ethnicity, BMI, DM, HTN • Sensitivity analysis • Misclassification of HCV or alcohol abuse Pts • Limited to cohorts : ≥ 1 HCV test and ≥ 1 AUDIT-C test & all negative • Multiple imputation of missing HCV and AUDIT-C data for cases and controls • Differences in healthcare utilization • Limited to cohorts : ≥ 2 visits in 2 sequential yrs • Subgroup and secondary analysis • Annual HCC incidence in NAFLD patients including subgroups (gender, age, race/ethnicity, cirrhosis status, and FIB-4) • HCC in NAFLD Pts without cirrhosis : EMR review

Risk of HCC in Patients with NAFLD Methods • Patients characteristics Results Discussions Characteristic, N (%) NAFLD (n=296, 707) Non-NAFLD (n=296, 707) 55. 4 (13. 2) 55. 4 (13. 1) 280, 177 (94. 4) 16, 530 (5. 6) White 203, 814 (68. 7) 175, 487 (59. 1) African American 33, 755 (11. 4) 46, 408 (15. 6) Hispanic 15, 839 (5. 3) 11, 446 (3. 9) Other races 7, 803 (2. 6) 6, 243 (2. 1) 35, 496 (12. 0) 57, 123 (19. 3) 30. 7 (7. 0) 28. 7 (6. 9) HTN 210, 194 (70. 8) 178, 483 (60. 2) DM 88, 593 (29. 9) 69, 072 (23. 3) LC 1, 084 (0. 4) 534 (0. 2) Mean duration of FU (yr) 9. 0 (SD 2. 2) 8. 67 (SD 2. 60) Age, mean (SD) Gender Men Women Race/ethnicity Missing BMI median, IQR

Risk of HCC in Patients with NAFLD Methods Results Discussions • Results of Chart validation • High accuracy of modified NAFLD algorithm • PPV 89% (95%CI = 84 -94%) • NPV 98% (95%CI = 93 -99%) • HCC case ascertainment • 367 Pts were diagnosed with HCC during FU in CCR • 17 Pts were classified as possible HCC base on histology in CCR EMR review 5 Pts had HCC confirmed in EMR • 406 Pts had ICD codes for HCC but not diagnosed EMR review 173 Pts had HCC confirmed in EMR Ø Total 545 Pts

Risk of HCC in Patients with NAFLD • Risk of HCC in NAFLD Methods Results Discussions NAFLD Non-NAFLD Annual incidence rate 490 Pts, 0. 21 per 1000 PY (95%CI 0. 19 -0. 22 per 1000 PY) 55 Pts, 0. 02 per 1000 PY (95%CI 0. 02 -0. 03 per 1000 PY) 5 -yr cumulative incidence rate 0. 8 per 1000 PY 0. 09 per 1000 PY 10 -yr cumulative incidence rate 1. 7 per 1000 PY 0. 18 per 1000 PY Characteristic Adjusted HR (95% CI) P-value 7. 62 (5. 76 -10. 09) <0. 0001 ≤ 45 yrs 0. 05 (0. 02 -0. 13) <0. 0001 ≥ 65 yrs 1. 83 (1. 53 -2. 18) <0. 0001 Hispanic 1. 59 (1. 14 -2. 20) 0. 005 African American 0. 74 (0. 53 -1. 03) 0. 07 Other race 1. 03 (0. 59 -1. 78) 0. 92 Missing 1. 16 (0. 90 -1. 49) 0. 25 1. 18 (0. 99 -1. 42) 0. 06 3. 03 (2. 52 -3. 64) <0. 0001 1. 10 (0. 84 -1. 43) 0. 50 NAFLD (ref no) Yes Age (ref 46 -64 yrs) Race/ethnicity (ref White) BMI (ref <30) ≥ 30 DM (ref no) Yes HTN (ref no) Yes

Risk of HCC in Patients with NAFLD Methods Results Discussions • Risk of HCC in NAFLD Association between NAFLD and HCC cancer (sensitivity analysis) HCC incidence group HCC case Total person-year (PY) of FU IR (95% CI) (per 1000 PY) Adjusted HR (95% CI) ① Limiting to patients with complete rule out for HCV and alcohol NAFLD (n=164, 259) 309 1, 364, 656 0. 23 (0. 20 -0. 25) 5. 86 (4. 29 -7. 99) Non-NAFLD (n=164, 259) 46 1, 342, 817 0. 03 (0. 03 -0. 05) Ref ② Using all patients with imputation of missing data on HCV and alcohol NAFLD (n=232, 056) 395 1, 894, 084 0. 21 (0. 19 -0. 23) 6. 5 (4. 8 -8. 6) Non-NAFLD (n=232, 056) xx 1, 865, 435 0. 03 (0. 02 -0. 04) Ref ③ Limiting to patients with at least 1 visit in each of the 2 consecutive follow up years following index date NAFLD (n=254, 100) 455 2, 033, 812 0. 22 (0. 20 -0. 25) 6. 81 (5. 21 -8. 91) Non-NAFLD (n=254, 105) 61 1, 993, 340 0. 03 (0. 02 -0. 04) Ref

Risk of HCC in Patients with NAFLD Annual incidence rate of HCC in subgroups of NAFLD Pts Methods Results Group HCC incidence HCC case Total person-year (PY) of FU IR (95% CI) (per 1000 PY) 4 563, 905 0. 01 (0. 00 -0. 02) 46 -64 281 1, 312, 276 0. 21 (0. 19 -0. 24) ≥ 65 205 506, 108 0. 41 (0. 35 -0. 46) 484 2, 242, 258 0. 22 (0. 20 -0. 24) 6 140, 030 0. 04 (0. 02 -0. 09) White 345 1, 638, 140 0. 21 (0. 19 -0. 23) African American 32 273, 494 0. 12 (0. 08 -0. 17) Hispanic 38 129, 745 0. 29 (0. 21 -0. 40) White 199 1, 259, 727 0. 16 (0. 14 -0. 18) African American 24 242, 465 0. 10 (0. 06 -0. 15) Hispanic 20 110, 350 0. 18 (0. 11 -0. 28) White 146 378, 413 0. 39 (0. 33 -0. 45) African American 8 31, 029 0. 26 (0. 11 -0. 51) Hispanic 18 19, 395 0. 93 (0. 55 -1. 47) 165 433, 378 0. 45 (0. 39 -0. 54) Age (yr) ≤ 45 Discussions Gender Men women Race/ethnicity Age and race/ethnicity <65 yr ≥ 65 yr Hepatic steatosis in image Yes

Risk of HCC in Patients with NAFLD Methods • Incidence of HCC in NAFLD Pts with vs without cirrhosis Results Annual incidence per 1000 PY Cirrhosis Discussions • yes 10. 6 No 0. 08 Annual HCC risk in subgroups of patients with cirrhosis Annual incidence per 1000 PY (95%CI) Overall cirrhosis 10. 63 (9. 46 -11. 91) Gender Men women Annual incidence per 1000 PY (95%CI) Age by race <65 yr 11. 05 (9. 83 -12. 39) 1. 62 (0. 20 -5. 85) Race White 9. 85 (8. 55 -11. 29) African American 7. 43 (3. 96 -12. 71) Hispanic 15. 12(10. 28 -21. 47) White 11. 94 (9. 11 -15. 37) ≥ 65 yr African American 11. 94 (3. 25 -30. 56) White 9. 18 (7. 76 -10. 80) Hispanic 23. 76 (12. 27 -41. 50) African American 6. 37 (2. 91 -12. 08) Hispanic 12. 30 (7. 41 -19. 21) Yes 12. 36 (10. 67 -14. 24) no 8. 51 (6. 96 -10. 29) Age <65 yr 9. 74 (8. 46 -11. 17) ≥ 65 yr 13. 43 (10. 82 -16. 49) DM

Risk of HCC in Patients with NAFLD Methods Results Discussions • Incidence of HCC in NAFLD Pts without cirrhosis • Total 193 Pts • • 98 62 20 13 (50. 7% of 193; 20. 0% of 490) Pts : evidence of absence of cirrhosis (32. 1% of 193) Pts : evidence of cirrhosis (10. 3% of 193) Pts : possible cirrhosis during F/U (6. 7% of 193) Pts : undetermined

Risk of HCC in Patients with NAFLD Methods Results Discussions • Conclusion • Risk of HCC in NAFLD Pts was significantly higher than that in age and gender matched controls. • Risk of HCC in NAFLD Pts was different by age, gender, and race/ethnicity. • Cirrhosis was the most common as well the strongest risk factor for HCC in NAFLD. • NAFLD-related HCC cases sometimes occurred in the absence of cirrhosis. • Limitation • Limited to veterans. • Possibility of underestimation & misclassification

HCC in patients with NAFLD HEPATOLOGY, December 2015

Association of NAFLD with HCC in US from 2004 to 2009 Methods Results Discussions • Surveillance, Epidemiology and End Results (SEER) • Medicare linked database • 2004 -2009 • Population • HCC Pts group • SEER • Diagnosed with their 1 st primary liver cancer during 2001 -2009 • Control group • SUMDENOM • The data files : PEDSF, MEDPAR (inpatients), NCH (physician/supplier), Outpatient claims

Association of NAFLD with HCC in US from 2004 to 2009 Methods ICD C 22. 0 (liver) & morphology codes 8170 -8175 Results Discussions Chronic liver disease (-) N=2502(16. 7%)

Association of NAFLD with HCC in US from 2004 to 2009 Methods Results • Identifying comorbidities • Modified CCI • ICD-9 -CM codes • Heart disease comorbidity Discussions • acute myocardial infarction (410) • old myocardial infarction (412) • congestive heart failure (428) • Definition of different chronic liver disease • ICD-9 codes • HCV infection : (070. 41, 070. 44, 070. 51, 070. 54, 070. 7, V 02. 62) • HBV infection : (070. 2, 070. 3, 070. 42, 070. 52, V 02. 61) • Alcoholic liver disease (ALD) • alcoholic fatty liver disease (571. 0) • alcoholic hepatitis (571. 1) • alcoholic cirrhosis of the liver (571. 2) • alcoholic liver damage (571. 3) • other alcohol-related disorders (303, 305. 0, V 11. 3, V 79. 1, 291) • NAFLD • (571. 8, 571. 9) • Cryptogenic without mention of alcohol or other liver diseases (571. 5) + DM (250), obesity (278) • Other liver disease • autoimmune hepatitis (571. 4), biliary cirrhosis (571. 6), hemochromatosis (275. 0), and Wilson’s disease (275. 1) • Treatment After Diagnosis of HCC • hospital procedure codes (ICD 9 -CM procedure code) • liver transplant (LT) 50. 5

Association of NAFLD with HCC in US from 2004 to 2009 • Description of study cohort Methods Results Discussions NAFLD-HCC vs. HCV/HBVHCC Older, more likely to be white, more likely to be unstaged (p<0. 05)

Association of NAFLD with HCC in US from 2004 to 2009 Methods Results Discussions

Association of NAFLD with HCC in US from 2004 to 2009 Methods Results Discussions NAFLD-HCC vs. NAFLD-non. HCC More likely to be male, more sicker (p<0. 05)

Association of NAFLD with HCC in US from 2004 to 2009 Methods Results • Temporal Trends in the Proportion of HCC From 2004 to 2009 • Factors associated with the diagnosis of HCC Discussions Annual increase 11% 13% 9%

Association of NAFLD with HCC in US from 2004 to 2009 Methods Results Discussions • Mortality Trends Between 2004 and 2009 • Mortality rate within 1 yr of diagnosis NAFLD-HCC vs. HCV/HBV-HCC 61% vs. 50% (p<0. 0001) • Factors Associated With 1 -Year Mortality.

Association of NAFLD with HCC in US from 2004 to 2009 Methods Results Discussions • The rate of HCC is increasing. • HCV is the most common diagnosis associated with HCC, followed by ALD and NAFLD. • NAFLD-HCC • Older at the time of diagnosis, more likely to be white, shorter survival time than HCV/HBC hepatitis-HCC • Cause of poor prognosis of NAFLD-HCC Pts compared with HBV/HCV-HCC Pts • NALFD Pts may get less HCC screening test than viral hepatitis Pts (Esp. early stage of NAFLD) • US may fail to detect small tumors d/t visceral obesity and intrahepatic fat. • NAFLD-HCC Pt’s metabolic comorbidities make aggressive treatment difficult. • Limits • Limitation of SEER data and Medicare • True increasing rate of HCC prevalence vs. increasing rate of HCC screening

HCC in patients with NAFLD Further discussion • Criteria of NAFLD Pts groups don’t match definition of NAFLD • NAFLD Pts could have normal ALT • Not all cryptogenic liver disease with DM or obesity is NAFLD Kanwal F et al. , Gastroenterology, 2018 Zobar M. Younossi et al. , Hepatology, 2015

HCC in patients with NAFLD Further discussion • Possibility of misclassification • Alcohol : AUDIT-C identifies alcohol abuse or dependency • HCC staging : What is Unstaged tumor? ≥ 4 in men, ≥ 3 in women Risk of alcohol misuse

HCC in patients with NAFLD Further discussion • Cause & Pathology of NAFLD-HCC w/o cirrhosis • Obesity, T 2 DM • Insulin resistance, hepatic steatosis adipose tissue-derived inflammation, adipokines (hormone) effect, oxidative stress • Diet, gut microbiome • Generic risk factor : MBOAT 7, PNPLA 3, TERT Ramy Younes et al. , Journal of hepatology, 2017 B Donati et al. , nature, 2017 Zámbó V et al. , World J. Hepatol, 2013

HCC in patients with NAFLD Further discussion • HCC surveillance for NAFLD Pts • HCC surveillance is more important in NASH Pts than in early NAFL Pts. • Surveillance by US has a suboptimal performance in NAFLD Pts. • HCC surveillance may be needed to some non-cirrhotic NAFLD Pts. (Cost-effectiveness) Julei K Heimbach et al. , Hepatology, 2018 Jorge Fonseca et al. , Journal of hepatology, 2016 www. cancer. go. kr