Development and Application of an Index to Adjust for Concomitant Hepatotoxic Medication Exposure in Hepatic Safety Studies

Tuesday, August 27, 2024

9:15 AM – 9:30 AM CEST

Location: Conv Hall I D

Presenting Author(s)

Jessie Torgersen, MD, MHS, MSCE (she/her/hers)

Assistant Professor of Medicine
University of Pennsylvania Perelman School of Medicine, United States

Slides

Co-Author(s)

CN

Craig W. Newcomb, MS

Department of Biostatistics, Epidemiology, and Informatics, Center for Clinical Epidemiology and Biostatistics, Center for Real-World Effectiveness and Safety of Therapeutics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA, United States
DC

Dena M. Carbonari, MS

Department of Biostatistics, Epidemiology, and Informatics, Center for Clinical Epidemiology and Biostatistics, Center for Real-World Effectiveness and Safety of Therapeutics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA
Philadelphia, United States
Sean Hennessy, PharmD, PhD

Professor of Epidemiology
Center for Real-World Effectiveness and Safety of Therapeutics, University of Pennsylvania
Philadelphia, United States
Christopher T. Rentsch, PhD, FISPE

Associate Professor
London School of Hygiene and Tropical Medicine / Yale School of Medicine
London, United Kingdom
JT

Janet P. Tate, ScD

Associate Professor
VA Connecticut Healthcare System, US Department of Veterans Affairs, West Haven, CT, USA, United States
Jason A. Roy, PhD

Professor of Biostatistics and Epidemiology
Rutgers School of Public Health, United States
TT

Tamar H. Taddei, MD

VA Connecticut Healthcare System, US Department of Veterans Affairs, West Haven, CT, USA, United States
Amy C. Justice, MD, PhD

Yale School of Medicine
West Haven, United States
Vin Lo Re, III, MD, MSCE (he/him/his)

Associate Professor of Medicine and Epidemiology
University of Pennsylvania
Philadelphia, United States

Background: Studies evaluating the hepatic safety of medications have been limited by their inability to control for confounding due to concomitant receipt of other potentially hepatotoxic drugs and may lead to inaccurate estimates of associations between a medication of interest and severe acute liver injury (SALI).

Objectives: Develop an index to adjust for hepatotoxic medication exposure within hepatic safety studies.

Methods: We created 195 cohorts of persons without liver/biliary disease who initiated a suspected hepatotoxic medication (defined by ≥4 published case reports in the US NIH LiverTox website) as outpatients in the Veterans Health Administration from 2000-2021. Patients were considered exposed to a suspected hepatotoxic drug from initiation through 30 days after discontinuation or up to one year. The main outcome was hospitalization for SALI, defined by either inpatient: 1) alanine aminotransferase >120 U/L + total bilirubin >2.0 mg/dL, or 2) INR ≥1.5 + total bilirubin >2.0 mg/dL, with both abnormalities present within 2 days of admission. We censored at dispensed oral anticoagulants or liver/biliary disease diagnoses recorded during follow-up or as hospital discharge diagnosis. We calculated age-/sex-adjusted incidence rates (IRs) with 95% confidence intervals (CI) of SALI. Twenty-two medications had the highest IR of SALI (≥3.0 events/10,000 person-years). IRs of these 22 drugs served as weights in a risk index, termed the Hepatotoxicity Adjustment Index (HAI). To demonstrate real-world use of this index, we then calculated the HAI for initiators of proton pump inhibitors (PPIs). We summed the weights of the hepatotoxic drugs included in the HAI that were dispensed within 90 days prior to PPI initiation to account for most proximal concomitant medication exposure. Age-/sex-adjusted hazard ratios (HRs) of SALI with 95% CIs were determined with and without adjustment for HAI.

Results: Among initiators of omeprazole (n=2,022,861), pantoprazole (n=324,130), and lansoprazole (n=115,423), IRs of hospitalization for SALI were 2.3 (2.0-2.6), 1.5 (0.9-2.3) and 4.2 (2.6-6.9) events/10,000 person-years, respectively. Among all PPI initiators, higher HAI category was associated with greater HRs of SALI (HAI 3-4: HR=2.31 [1.42-3.76]; HAI 5-14: HR=2.69 [1.86-3.91]; HAI 15+: HR=5.82 [2.57-13.15] compared to HAI=0). After age/sex adjustment, risk of SALI was increased with lansoprazole (HR=1.70 [1.02-2.84]) but not pantoprazole (HR=0.65 [0.40-1.04]) compared to omeprazole. After additional adjustment for HAI, the association between lansoprazole and SALI was no longer significant (HR=1.55 [0.93-2.59]).

Conclusions: The HAI provides a tool to adjust for potential confounding due to hepatotoxic drugs within hepatic safety studies.