DILI control compounds list: An analysis of FAERS data

The analysis of the DILI (drug-induced liver injury) control compounds list using FAERS (FDA Adverse Event Reporting System) data aimed to validate the consensus-driven list of DILI-positive and negative control drugs proposed by Segovia-Zafra et al. Here is a detailed breakdown of the study:

### **Study Aim**

The primary goal of the analysis was to provide real-world validation of the proposed DILI control compounds list. This list was designed to categorize drugs based on their potential to cause liver injury (DILI-positive) or their lack of association with liver injury (DILI-negative). The study used FAERS data to assess the hepatotoxic potential of these drugs and to confirm the reliability of the list for use in in vitro model validation.

---

### **Methodology**

1. **Data Source**: Researchers utilized FAERS, a robust pharmacovigilance database containing reports of adverse drug reactions submitted between 2004 and 2024.

2. **Algorithms Used**: Multiple pharmacovigilance algorithms were applied to detect signals of liver injury:

• **Reporting Odds Ratio (ROR)**
• **Proportional Reporting Ratio (PRR)**
• **Information Component (IC)**
• **Empirical Bayesian Geometric Mean (EBGM)**

3. **Analysis Focus**: The study analyzed the frequency and strength of liver injury signals for the drugs on the DILI control list, comparing DILI-positive and DILI-negative compounds.

---

### **Key Findings**

1. **Validation of DILI-Positive Drugs**:

• Most DILI-positive drugs exhibited strong liver injury signals across all metrics, confirming their hepatotoxic potential.
• **Isoniazid** showed the strongest association with liver injury, with a Reporting Odds Ratio (ROR) of 43.3 and an IC025 of 3.7, indicating a very high likelihood of DILI.

2. **Validation of DILI-Negative Drugs**:

• Negative control compounds, such as **diphenhydramine** and **lidocaine**, had few or no reports of liver injury, supporting their classification as non-hepatotoxic.

3. **Data Gaps**:

• Certain drugs, such as **troglitazone** and **isoproterenol**, had no available FAERS data. This was attributed to factors such as market withdrawal or limited clinical usage, which reduced the number of adverse event reports for these drugs.
• These gaps highlight a limitation of spontaneous reporting systems like FAERS, as they rely on voluntary submissions and may not capture data for all drugs.

---

### **Conclusion**

The FAERS-based analysis largely supports the proposed DILI control compounds list. The findings enhance confidence in the list's validity and its potential utility for in vitro model validation. However, the study also acknowledges limitations in FAERS data, such as underreporting and data gaps for certain drugs. To address these challenges and further strengthen the validation process, the authors recommend complementary studies using structured real-world datasets, such as electronic health records or clinical trial data.

---

### **Implications**

The study reinforces the utility of pharmacovigilance databases like FAERS in evaluating drug safety. The validated DILI control compounds list can serve as a reliable reference for researchers developing predictive models of hepatotoxicity, ultimately improving drug safety assessments and reducing the risk of liver injury in patients.