Drug-Induced Liver Injury Network Causality Assessment: Criteria and Experience in the United States

Background & Aims: The United States Food and Drug Administration (FDA) regulates a broad range of consumer products, which account for about 25% of the United States market. The FDA regulatory activities often involve producing and reading of a large number of documents, which is time consuming and labor intensive. To support regulatory science at FDA, we evaluated artificial intelligence (AI)-based natural language processing (NLP) of regulatory documents for text classification and compared deep learning-based models with a conventional keywords-based model.Methods: FDA drug labeling documents were used as a representative regulatory data source to classify drug-induced liver injury (DILI) risk by employing the state-of-the-art language model BERT. The resulting NLP-DILI classification model was statistically validated with both internal and external validation procedures and applied to the labeling data from the European Medicines Agency (EMA) for cross-agency application.Results: The NLP-DILI model developed using FDA labeling documents and evaluated by cross-validations in this study showed remarkable performance in DILI classification with a recall of 1 and a precision of 0.78. When cross-agency data were used to validate the model, the performance remained comparable, demonstrating that the model was portable across agencies. Results also suggested that the model was able to capture the semantic meanings of sentences in drug labeling.Conclusion: Deep learning-based NLP models performed well in DILI classification of drug labeling documents and learned the meanings of complex text in drug labeling. This proof-of-concept work demonstrated that using AI technologies to assist regulatory activities is a promising approach to modernize and advance regulatory science.

Download Full-text

Liver transplantation for acute liver failure from drug induced liver injury in the United States

Liver Transplantation ◽

10.1002/lt.20204 ◽

2004 ◽

Vol 10 (8) ◽

pp. 1018-1023 ◽

Cited By ~ 300

Author(s):

Mark W. Russo ◽

Joseph A. Galanko ◽

Roshan Shrestha ◽

Michael W. Fried ◽

Paul Watkins

Keyword(s):

United States ◽

Liver Transplantation ◽

Liver Injury ◽

Liver Failure ◽

Acute Liver Failure ◽

The United States ◽

Drug Induced ◽

Drug Induced Liver Injury

Download Full-text

Circulatory Inflammatory Mediators in the Prediction of Anti-Tuberculous Drug-Induced Liver Injury Using RUCAM for Causality Assessment

Biomedicines ◽

10.3390/biomedicines9080891 ◽

2021 ◽

Vol 9 (8) ◽

pp. 891

Author(s):

Cheng-Maw Ho ◽

Chi-Ling Chen ◽

Chia-Hao Chang ◽

Meng-Rui Lee ◽

Jann-Yuan Wang ◽

...

Keyword(s):

Liver Injury ◽

Inflammatory Mediators ◽

Cox Regression ◽

Causality Assessment ◽

Area Under The Curve ◽

Assessment Method ◽

Drug Induced ◽

Drug Induced Liver Injury ◽

Tb Treatment ◽

Pre Treatment

Background: Anti-tuberculous (TB) medications are common causes of drug-induced liver injury (DILI). Limited data are available on systemic inflammatory mediators as biomarkers for predicting DILI before treatment. We aimed to select predictive markers among potential candidates and to formulate a predictive model of DILI for TB patients. Methods: Adult active TB patients from a prospective cohort were enrolled, and all participants received standard anti-tuberculous treatment. Development of DILI, defined as ≥5× ULN for alanine transaminase or ≥2.6× ULN of total bilirubin with causality assessment (RUCAM, Roussel Uclaf causality assessment method), was regularly monitored. Pre-treatment plasma was assayed for 15 candidates, and a set of risk prediction scores was established using Cox regression and receiver-operating characteristic analyses. Results: A total of 19 (7.9%) in 240 patients developed DILI (including six carriers of hepatitis B virus) following anti-TB treatment. Interleukin (IL)-22 binding protein (BP), interferon gamma-induced protein 1 (IP-10), soluble CD163 (sCD163), IL-6, and CD206 were significant univariable factors associated with DILI development, and the former three were backward selected as multivariable factors, with adjusted hazards of 0.20 (0.07–0.58), 3.71 (1.35–10.21), and 3.28 (1.07–10.06), respectively. A score set composed of IL-22BP, IP-10, and sCD163 had an improved area under the curve of 0.744 (p < 0.001). Conclusions: Pre-treatment IL-22BP was a protective biomarker against DILI development under anti-TB treatment, and a score set by additional risk factors of IP-10 and sCD163 employed an adequate DILI prediction.

Download Full-text