Screening of Susceptibility Genes for APAP Induced Drug Induced LIver Injury in ChiNese Population: a Case-control Study (PAIN)

I. Study Object

(i) Into the group strategy: continuous into group.

(ii) Grouping strategy：

1. Case group.

   Inclusion criteria：

   • A clear history of acetaminophen (or acetaminophen-containing drugs) ingestion.
   • Plasma and/or urine testing for acetaminophen components if history of ingestion is unclear.
   • Monitoring of Alanine aminotransferase (ALT) or Aspartate Aminotransferase (AST) ≥ 1000 IU/L at any time after APAP administration and Roussel Causality Assessment Method(RUCAM) score > 6
   • Age ≥ 14 years old.
   • The subject or guardian agrees to participate in this project and signs an informed consent form.

   Exclusion criteria： •The use of drugs for which frequency of adverse reactions to liver damage is defined as "common or very common" (≥1%) in the instructions.

   • Concurrent use of herbs that are clearly susceptible to liver damage (see list of definitions in the Annex).
   • Have a known definite cause of liver damage: active viral hepatitis; alcoholic liver disease; autoimmune liver disease; primary or secondary liver tumors; and other underlying liver disease that has affected liver function.
   • Those who fail to provide complete general information and clinical information.
   • Subjects or guardians who do not agree to see this project do not sign the informed consent form.

2. Control group

   Inclusion criteria：

   • A clear history of acetaminophen (or acetaminophen-containing drugs) ingestion.
   • Plasma and/or urine testing for acetaminophen components if history of APAP ingestion is unclear.
   • Age: ≥ 14 years old.
   • The subject or guardian agrees to participate in this project and signs an informed consent form.

   Exclusion criteria： •The use of drugs for which frequency of adverse reactions to liver damage is defined as "common or very common" (≥1%) in the instructions.

   •Concurrent use of herbs that are clearly susceptible to liver damage (see list of definitions in the Annex).

   •There are known definite causes of liver damage (see attached list of definitions): active viral hepatitis; alcoholic liver disease; autoimmune liver disease; primary or secondary liver tumors; and other underlying liver disease that has affected liver function.

   •Those who fail to provide complete general information and clinical information.

   •Subjects or guardians who do not agree to see this project do not sign the informed consent form.

   (iii) Matching strategy.

   Matching principle：

   •Case and control participants were matched according to ingested dose, duration of antidote administration, and duration of gastric lavage.

   •1:2 matching.

   Confounding factors：

   •Dose intake: patient report (primary) + blood concentration test (secondary)

   •Antidote use time: <4 hours, 4-24 hours, >24 hours

   •Gastric lavage time: <1 hour, ≥1 hour

   (iv) Estimation of sample size

   Parameter source：

   • Focus on drug metabolizing enzymes, immunogenic loci
   • The distribution of APAP-dependent DILI susceptibility allele frequencies in Asian populations based on literature findings.

2. Parameter value：

• Hypothesis: Gene-Environment.
• Outcome Model: Baseline risk P0=0.10.
• Genetic Effect =1.2,
• Power=0.8, Type I error rate=0.05 (two-sided).

Results： The case sample size should be 113 with a control sample size of 226.

II. Exposure/risk factors.

(i) Definition. Exposure factors refer to susceptibility genes for DILI caused by APAP, including：

• 2,036,060 marker sites
• Human Leukocyte Antigen (HLA) :HLA-A and HLA-B immune loci.
• Genotype-Imputation.

(ii) Measurement methods.

1. Adopt candidate gene strategy. The candidate gene strategy is a flexible approach with low cost, relatively simple quality control and statistical analysis, and a relatively easy biological explanation behind the association because the genes selected are all important for the disease.
2. Using the hypothesis-driven approach. Considering the hypothesis that some specific genes may be associated with the study outcome, a sequence-based approach was chosen for the analysis of potentially functional single nucleotide polymorphism (SNPs), selecting SNPs that are more likely to be of functional value associated with drug metabolizing enzymes, immunogenic loci, drug transport proteins, drug-acting receptors and other links, especially those located in candidate genes coding regions.
3. Using custom chips. The Expanded Multi-Ethnic Genotyping Array (MEGAEX), a microarray built by consortia with enhanced capabilities to understand complex diseases in a variety of populations. This commercially available microarray determines individual drug response through the detection of 2,036,060 marker loci. It also complements HLA-A and HLA-B immune loci information to complete Genotype-Imputation.

III. Quality control The study has a strict standard test operation procedure, and relevant training is conducted for the personnel involved in the test before the start of the experiment, and the test can only be conducted after passing the training. We have a quality control center with dedicated personnel responsible for subject progress and data quality control. An online digital randomization platform and information entry management system is set up based on the web server terminal, which is capable of timely case randomization grouping and electronic clinical case observation form information entry, and can effectively and automatically check and correct errors during the information entry process. That is, it helps to ensure the accuracy and reliability of the entered information, and can provide real-time data monitoring for the supervisors. In addition, this study will monitor the conduct of experiments and data entry.

IV. Data Management Plan

• Data Center: The data management center established by the national data management standard undertakes data management work.
• Data management system: REDCap is used to establish electronic case report forms for online data collection and management.
• Data collection: Data entry by independent data registrants not involved in the study in the study lead unit.
• Data consistency monitoring: Data consistency monitoring by an independent third-party monitor.
• Data audit and locking: Data is audited, cleaned and locked by a dedicated data manager.

V. Statistical analysis plan

(i) Hypothesis testing. H0: No difference in genes between subjects with and without liver damage after ingestion of APAP H1: Significant genetic differences between subjects with and without liver damage after ingestion of APAP

(ii) Analytical strategies.

• Continuous variables were described as mean ± standard deviation (SD) or median of interquartile range (IQR), and differences between groups were analyzed by two-factor ANOVA or nonparametric tests. Categorical variables were expressed as numbers and percentages and analyzed by logistic regression models.
• Hardy-Weinberg equilibrium (HWE) between control groups was tested using the goodness-of-fit χ2 test. Using haplotype analysis view（Haploview） 4.2 software, the haplotype region group was selected considering the linkage disequilibrium between tags of single nucleotide polymorphisms (tagSNPs).
• Multivariate conditional logistic regression analysis was performed to estimate the association between genotype and risk of DILI from APAP by dominance ratios (ORs) and 95% confidence intervals (CIs) with gastric lavage and antidote use as covariates.
• Three different genetic models were used to synthesize the role of tagSNPs. Software package for Hardy-weinberg analysis(SHEsis) online program was used for haplotype analysis under logistic regression model.
• All analyses were performed using Statistics is a powerful statistical software platform(SPSS) for Windows (version 26.0 USA). Two-tailed P values <0.05 were considered statistically significant.