Background In vitro-in vivo extrapolation (IVIVE) is an approach utilizing in vitro experimental data to predict in vivo phenomena. It is a promising tool for chemical risk assessment.

Objective To learn the hotspots, evolution path, and trend of IVIVE in risk assessment by literature search and bibliometric analysis, and provide reference and data support for subsequent research.

Methods PubMed and Web of Science Core Collection were selected as foreign databases to search for literature about IVIVE applied in risk assessment published by December 31, 2023. The number of relevant documents in CNKI and Wanfang database was too small, so the Chinese databases were not included in this study. This study employed bibilometric analysis using VOSviewer and CiteSpace for visualizing networks categorized by author, institution, country, journal, keyword, and co-citation.

Results A total of 189 articles were included in this study. The first article was published in 2006, and since then the number of publications overall showed an upward trend and increased significantly after 2016. The institution with the most publications was the United States Environmental Protection Agency (28 articles). The United States was the most productive country (87 articles), and had a close cooperation with the United Kingdom. The journal with the most publications and the highest number of citations per article was Archives of Toxicology (19 articles). The keyword co-occurrence analysis suggested that research on IVIVE in risk assessment mainly studied the methods and models of IVIVE and prediction of chemical toxicity, and toxicity, in vitro, and models were the research hotspots in this field. Keyword timeline cluster analysis suggested that the assessment objects gradually expanded from drugs to environmental chemicals, organic chemicals and food additives. The co-citation analysis suggested that articles about IVIVE in risk assessment mostly cited journals in the environment, food, and drug fields, and these articles were mainly methodological studies followed by literature reviews.

Conclusion The research of IVIVE in risk assessment has developed rapidly. With the improvement of prediction models and the expansion of application scope, animal experiments in risk assessment may be greatly reduced and the efficiency of risk assessment can be increased. At present, the United States has a leading position in this field, while China has few relevant studies and needs to actively carry out international cooperation to improve the level of applied research of IVIVE. In the future, it is hoped that the IVIVE method can be further refined to improve its application and expand its research fields.