The Institute for In Vitro Sciences (IIVS) has sponsored a series of workshops with the aim of developing recommendations for scientific and technical approaches to conducting in vitro assays to assess potential toxicity within and across tobacco and nicotine products. As well as providing a background overview of oral tobacco and nicotine products, this report focuses on the methods used to generate extracts from these products for in vitro testing, the recommended best testing practices, and suggested areas for future research. This publication was developed by a working group of workshop members, in conjunction with the ninth IIVS in vitro workshop entitled 'In Vitro Models for Testing Oral Tobacco and Nicotine Products and Continued Discussion of In Vitro Models of Toxicity and Disease', which was held in January 2024.
The development of antidepressants has traditionally relied on vertebrate models. Drosophila melanogaster offers a promising alternative approach, yet evidence of antidepressant-like effects is fragmented. The aim of this systematic review was to analyse the literature on the effects of antidepressants on D. melanogaster behaviours. A comprehensive search of a number of bibliographic databases (i.e. PubMed, Web of Science, Scopus and EMBASE) was followed by a further screening of the retrieved articles (i.e. inclusion of studies on any D. melanogaster strain, sex, age, dose or outcome, tested in any behavioural assay; exclusion of reviews, studies on co-treatments and/or on non-prototypical antidepressants, and studies lacking control groups). This resulted in the final selection of 12 articles, comprising 146 studies. Most studies involved adult flies (k = 131), males (k = 56), fluoxetine treatment (k = 109) and courtship outcomes (k = 85). Some studies involved larvae (k = 15), females (k = 39), treatment with other antidepressants (citalopram, k = 12; desipramine, k = 6; amitriptyline, k = 2; imipramine, k = 2) and other outcomes (locomotion, k = 11; sucrose preference, k = 5; light preference, k = 2; aggression, k = 2; climbing, k = 1). Studies on larvae had, on average, larger sample sizes than studies on adult flies, resulting in more conclusive effect sizes. The effect sizes indicated that most studies presented inconclusive effects. However, sucrose preference, courtship and light preference were outcomes with conclusive findings, depending on the conditions (sex, treatment, stress), and thus these may represent more reliable indicators of antidepressant effects. However, the lack of consistency among studies precluded drawing any definitive conclusions on reliable indicators. In addition, incomplete reporting precluded an accurate risk-of-bias assessment. These findings highlight the need for standardisation and transparent reporting, in order to establish flies as a reproducible, ethically acceptable alternative to vertebrates in antidepressant research.
The non-publication of negative or null results contributes to increased animal use in biomedical research and experimentation, by encouraging the unnecessary repetition of studies. This practice also delays the development, validation and uptake of alternative methods, including New Approach Methodologies (NAMs). This Comment article discusses that recognising and disseminating negative results is important, in terms of accelerating the implementation of alternative methods and supporting replacement, reduction and refinement (the Three Rs).
The preclinical evaluation of novel imaging agents, such as contrast agents for computed tomography (CT), still largely relies on rodent models or simplified in vitro systems, each carrying limitations with regard to ethical burden, physiological relevance or translational value. To address this gap, we propose Galleria mellonella larvae as a scalable, ethically favourable and physiologically relevant in vivo model for use in the early-stage screening of novel CT contrast agents. We established a standardised imaging protocol utilising isoflurane-based sedation and longitudinal micro-CT scanning, and used it to evaluate three commercially available contrast agents: one clinically approved molecular agent and two nanoparticulate formulations that are used for preclinical studies. All agents were systemically distributed, well-tolerated and enabled consistent visualisation of internal anatomy over eight days. While differences in clearance kinetics and route related to particle size or formulation were less pronounced compared to vertebrate models, this larval system reliably reflected biodistribution trends and allowed non-invasive tracking of agent functionality and safety. The use of G. mellonella in this context enables early elimination of less promising candidates and reduces reliance on vertebrates in downstream (pre)clinical development, aligning with the Three Rs principles. Our findings support the use of G. mellonella as a practical intermediate model, bridging in vitro cell-based assays and in vivo rodent studies in the development of next-generation CT and imaging contrast agents.
Risk decision-making has evolved from a vernacular focused on risk assessment and management to fully integrated approaches designed to inform risk acumen. Consequently, there has been a renewed interest, particularly in areas with significant paradigm shifts, to understand the complex nature of the underlying intersections between data, ethics and risk. For example, building more awareness of the risk and ethical implications for applying artificial intelligence for generative (content creation) and agentic (decision-making) purposes or relying on next generation risk assessments grounded in models reflective of the Three Rs principles (i.e. the replacement, reduction and refinement of animal studies). Global thinkers in risk science and analysis have also developed frameworks and models, such as the Projector Model, to show the complex nature of these intersections relevant to public health and regulatory risk decision-making. This Comment article builds on this work by sharing real-life examples from an expert panel discussion, which occurred during the Chief Data & Analytics Officer (CDAO) Canada Public Sector 2025 meeting. These panel members relied on the Projector Model to navigate the discussion in a session titled 'Data-driven science - Transforming risk assessment and regulation in the public sector'. The examples showed how institutional values and norms serve as foundational elements for risk decision-making, and highlighted that data-driven science, especially that based on novel approaches and technologies, needs careful consideration from an ethical and risk perspective.
