J Leukoc Biol. 2026 Aug 4:qiag108. doi: 10.1093/jleuko/qiag108. Online ahead of print.
ABSTRACT
Critical care research focuses on life-threatening conditions such as sepsis, trauma, hemorrhage, and burn injury, which account for millions of hospitalizations and hundreds of thousands of deaths annually in the United States alone. Recent policy initiatives by the U.S. Food and Drug Administration (FDA) and the National Institutes of Health (NIH) have promoted New Approach Methodologies (NAMs), including organoids, organ-on-chip platforms, and computational models, as alternatives to animal research. While NAMs offer valuable tools for mechanistic investigation and screening applications, this review examines whether current NAM technologies can adequately replace animal models in critical care research. Critical illness involves the whole organism, including dynamic organ-organ interactions, immune-microbiome crosstalk, and adaptive systemic feedback loops. By examining major domains in critical care research and targeted organ injuries, it becomes clear that while NAMs excel at interrogating isolated subsystems, they cannot currently replicate integrated physiological responses. Animal models remain essential for questions requiring assessment of multi-organ dysfunction, therapeutic safety evaluation, and clinically relevant disease trajectories. Premature policy shifts away from animal research will impede advances in critical care medicine. This review proposes a “methodological pluralism” approach that integrates NAMs with appropriately designed animal studies through harmonized endpoints and reverse-translation frameworks. Recommendations include continued refinement of animal models to better represent the heterogeneity of human populations, adoption of quality standards for preclinical research, and strategic deployment of both NAMs and animal models based on fit-for-purpose criteria. We believe this approach will satisfy ethical considerations, scientific rigor, and public health needs in critical care research.
PMID:42547063 | DOI:10.1093/jleuko/qiag108