The Short-Term Incidence of Hepatocellular Carcinoma Is Not Increased After Hepatitis C Treatment with Direct-Acting Antivirals: An ERCHIVES Study DK Li, YJ Ren, DS Fierer, S Rutledge, OS Shaikh, V Lo Re III, T Simon, A Abou-Samra, RT Chung, AA Butt Massachusetts General Hospital, Boston VA Pittsburgh Healthcare System, Pittsburgh Icahn School of Medicine, Mt Sinai, New York Perelman School of Medicine, University of Pennsylvania, Philadelphia Hamad Medical Corporation, Doha, Qatar Weill Cornell Medical College, Doha, Qatar & New York Accepted for Publication in HEPATOLOGY 2018 HKASLD Bi-monthly Scientific Meeting Journal Review James Y.Y. Fung 18 th Jan 2018
Introduction • IFN-induced SVR has been shown to reduce HCC risk by about 4-fold, regardless of stage of liver disease • The extrapolation of this benefit to DAAs and the expectation of greater reduction has not been consistent • Paradoxical increased risk of HCC in patients treated with DAA therapy
7/16 had been treated initially with • resection and 9/16 with ablation Median time from HCC treatment to • DAA was 11.2m (25%-75%: 3.6-23.2) Median time from CR to DAA was 1.7 • and 1.3 for HCC recurrent patient Median time from CR to recurrence • was 3.5 months (1.1-8) Recurrence rate is higher than • observed in historical non DAA treated controls Reig et al. J Hepatol 2016;65:719-726
Retrospective cohort study Median interval between HCC • • treatment and DAA 376 days 344 cirrhotics without HCC at • (range, 45-2706) time of DAAs Treated with DAAs and FU for 24 • weeks SVR 91% • 28.8% 59 with HCC history • – 19 resections – 2 resections + RFA 7.6% – 2 resections + TACE 3.2% – 18 RFA – 4 RFA + TACE 9/285 17/59 26/344 – 6 PEI – 5 TACE – 3 Data not complete Conti F et al. J Hepatol 2016;65:727-733
ANRS C012 Cirvir Cohort ANRS C022 HEPATHER Cohort 13% 21% 1/13 (8%) 31/66 (47%) ANRS C023 CUPILT Cohort No increase risk of recurrence of HCC with DAA-treated patients compared with non-treated patients 2.2% The ANRS collaborative study group on HCC. J Hepatol 2016
Study Aim • Whether DAA treatment is associated with higher rates of incident HCC using a large, well-established national cohort of HCV-infected United States Veterans without a prior diagnosis of HCC
Design & Data Source (I) • Retrospective cohort study of HCV-infected persons in the Electronically Retrieved Cohort of HCV Infected Veterans (ERCHIVES) database • All HCV-infected Veterans at any US-wide Dept of Veteran Affairs medical facilities with anti-HCV+ between 2002 and 2016 were identified • Demographic, clinical, lab data were obtained from the National Patient Care Database and the Corporate Data Warehouse
Design & Data Source (II) • Pharmacy information, including all prescriptions written, doses, duration, number of pills, number of refills, and date of refills, was retrieved from the Pharmacy Benefits Management database. • Data merged on established algorithms
Study Participants • Inclusion criteria – Anti-HCV+ • Exclusion criteria – HBV (HBsAg+) or HIV co-infection – Prior to baseline • Diagnosis of HCC • Missing HCV RNA or FIB-4 score • Incomplete data for FIB-4 calculation at least 12 weeks after baseline • Missing HCV RNA data for calculation of SVR12
Cohort Groups • Group A (IFN): – Received PEG + RBV for ≥28 days – If multiple courses of PR – 1 st treatment course used as baseline – Standard IFN excluded • Group B (DAA) – Received SOF/SMV±RBV; SOF/LED±RBV; SOF/ DCL±RBV; PRoD±RBV for ≥28 days – Other newer DAAs not included (small numbers) • Group C – No HCV treatment for ≥28 days
Baseline Definitions • For treated groups – Baseline was the date of HCV treatment initiation • For untreated groups – Baseline equal to the duration of HCV infection prior to treatment initiation in the corresponding treated person • Determining the average time between the first anti- HCV+ date and treatment initiation date for the treated group • Adding the same duration to the time between the first anti-HCV+ date in each untreated person and assigning that date as baseline
Definition of Cirrhosis • FIB-4 score >3.5 – FIB-4 = (age[years] x AST [IU/L]/(platelet count [platelets x 10 9 /L] x ALT ½ [IU/L]) • Lab data obtained at yearly intervals – FIB-4 score recalculated at each interval
Study Outcomes • Primary outcome – Development of incident HCC • Defined as presence of at least one inpatient or two outpatient ICD-9/10 codes for HCC made ≥3 months after baseline • Time to development of HCC determined relative from baseline • Given that advanced fibrosis is one of the strongest predictors of HCC development, primary analysis was performed on persons with baseline cirrhosis (FIB-4 >3.5)
Patient Characteristics Age (yrs) Sex, % male Race, % White Black Hispanic Others Diabetes, % BMI, median Alcohol, % Smoking, % Current Former Never
HCV GT, % 1 2 3 4,5,6 Missing Baseline HCVRNA log PPI use during treatment Statin use prior to baseline FIB-4 baseline, % <1.45 no fibrosis >3.5 cirrhosis 1.45-3.5 Median AFP RBV use, % Treatment, % <8 weeks 8 weeks 12 weeks 24 weeks SVR, % Incident HCC, %
Incidence Rate of HCC by Treatment Group and SVR Status (Entire Cohort)
Incidence Rate of HCC by Treatment Group and SVR Status (Cirrhotics Only) Treatment Group All cirrhotics IFN DAA regimens Untreated controls SVR12 subgroup IFN DAA regimens Non SVR12 subgroup IFN DAA regimens
Probability of HCC Development in Cirrhotic Person who achieved SVR
Baseline Cohort Characteristics from FIB-4 Sensitivity Analysis (Limiting definition of baseline FIB-4 to within 12 months prior to baseline) Overall 1933 FIB-4 values excluded 209 IFN 82 DAA 1642 untreated
Incidence Rate of HCC in Persons with Cirrhosis, by Treatment Group and SVR Status in FIB-4 Sensitivity Analysis
Baseline Characteristics of DAA Subgroups
Incidence Rate of HCC in Persons with Cirrhosis, by DAA Treatment and SVR Status Treatment Group Cirrhotics IFN SOF/SMV SOF/LED SVR12 subgroup IFN SOF/SMV p=0.07 SOF/LED Non SVR12 subgroup IFN SOF/SMV SOF/LED
Age, per 10 year increase Predictors for the Male sex Race Development of White (comparator) HCC in Persons Black Hispanic with Cirrhosis Other/Missing Diabetes based on BMI (per 1 unit increase) Multivariate, Cox Alcohol abuse history Smoking history Proportional Non-smoker (comparator) Current smoker Hazards Analysis Former smoker Missing HCV genotype, % 1 (comparator) 2 3 4,5,6 Missing HCV RNA, per 1 log increase PPI use (baseline onwards) Statin use AFP >20 (vs 20 or below) Treatment regimen Peg/RBV (comparator) Any DAA Attainment of SVR
Predictors for the Development of HCC in Cirrhotics Based on Multivariate Cox Proportional Hazards Analysis in FIB-4 Sensitivity Analysis
Discussion • DAA-treated persons did have a significantly higher HCC rate than IFN-treated persons (overall group) – Higher rate of known risk factors • Cirrhosis, older age, higher baseline AFP • Analyzing cirrhotic patients alone showed no difference in HCC-free survival between DAA and IFN groups – Lower HCC incidence in sensitivity analysis • Pre-treatment risk of HCC is the main driver of post- treatment risk of HCC
Discussion • Higher prevalence of underlying HCC risks among DAA patients due to “warehousing”? • Treatment with earliest DAA regimens may be surrogate for higher baseline HCC risk • Failed therapy, cirrhotics • Persons treated with SIM/SOF had higher rates of cirrhosis (c/w IFN/later DAAs) • Previous studies have not provided baseline data in combination with DAA regimens
Discussion • In contrast to previous studies, DAA treatment is not associated with higher risk of HCC compared to IFN treatment when controlling for differences in baseline HCC risk factors • Achievement of SVR with antiviral regimen is the most important determinant of a lower HCC risk
Study Limitations • Relatively short follow-up time in DAA cohort • Potential for limited generalizability – Veterans, males, higher smoking & alcohol use • Retrospective cohort study – Prospective studies unlikely • Lack of liver histology for diagnosing cirrhosis • Lack of info regarding HCC surveillance practices • Excluded those with prior HCC
Study Conclusion • In cirrhotics with chronic HCV infection, there was no association between HCC incidence rates and DAA treatment compared to IFN • Insufficient controlling for baseline risks for HCC and selection bias likely drove previously reported associations between DAA treatment and increased HCC risk
Recommend
More recommend
