The 15th Animal Science Meeting (ASM), co-organised by the Royal Society of Biology (RSB) and the Animals in Science Regulation Unit (ASRU), was held on Friday 8th December 2017. Over 70 representatives from academic, industry and CRO establishments around the country participated in a series of talks, discussion and workshops.
The Head of ASRU opened the event and highlighted how technological and methodological innovations are advancing animal research and animal welfare outcomes, hand in hand. ASRU’s operations to implement the Animals (Scientific Procedures) Act is therefore also under constant development to enable efficient and effective regulation of the sector.
The keynote lecture was given by Dr Matthew Leach from the University of Newcastle on Scientific Advances in the Study of Animal Welfare. The talk focused on the assessment of pain in non-human animals and, in particular, the type of experiments aimed at addressing the emotional component of pain in animals. A definition of pain in humans takes into consideration both a sensory and emotional component. In the case of non-human animals, we cannot assess the emotional component through a common, shared language that in humans gives access to the subjective individual experience. Therefore, scientists infer an emotional component of pain in non-human animals based on the assessment of similarities and differences in the anatomy, physiology, biochemistry and behaviour between humans and other animals. The talk considered these different dimensions of pain processing, focusing particularly on species of laboratory animals. Species-specific behaviours associated with pain can be identified because they are modulated by the administration of analgesia. However, they usually do not easily generalise across procedures, species, and sometimes even strains. A more comprehensive analysis of behaviour can be obtained by the use of automated algorithms for behavioural classification. So far, however, only scores of general behaviours have been obtained with further validation required to single out behaviours more specifically linked to pain. The study of complex natural behaviours (e.g. in rodents: explorative behaviour, nest building, grooming, and burrowing) and the use of learning paradigms (e.g. operant conditioning, such as in self-administration experiments, or conditioned perceptual discrimination) can provide a framework to investigate the emotional component of pain, because these behaviours cannot be simply explained by a “stimulus-response” reflex but require cerebral processing, and may carry an emotional valence attached to them. The study of cognitive biases is also being adapted to study pain in animals. These experiments can reveal how information previously associated with potentially painful stimuli can alter the motivational state and the way in which animals make choices under ambiguous situations. Finally, the scoring of facial expression using grimace scales has provided an accurate and reliable way to assess pain in non-human animals, as well as in non-verbal humans, in a way that might directly reflect emotional processing of negative experiences. In summary, to understand pain in animals there is a need to use a combination of different indices and experimental paradigms, tailored to the context and type of study undertaken. The adaptation of these methods to individual species and strains will be necessary to test their validity and generality.
The second talk of the morning focused on the importance of communication between tiers of the animal research licensing system, in delivering a complete and correct project license application that is fit for purpose. Anne-Marie Farmer (Cambridge University and President of the Laboratory Animal Science Association) gave an overview of the different actors involved in project license applications at establishments and their corresponding roles. Anne-Marie examined what is entailed in applying for a project licence (PPL), and provided a set of solutions to the challenges applicants frequently face, some of which have been tested at Cambridge University. Senior inspectors from ASRU then presented the regulator’s viewpoint on PPL applications. They listed a set of expectations, such as the submission of a high quality draft, and sketched potential routes for applicants to meet them, such as: ongoing and active communication with named people at establishments; involvement of the inspector during the application drafting phase; sharing advice between PPL holders about similar procedures; and adhering to the guidance provided by the annotated PPL form in addition to other Home Office advice notes and published resources by NC3Rs, RSPCA, IMPROVE etc.
A discussion and Q&A followed, where attendees contributed their views about the major challenges faced by applicants. Many of these had been described during the 2017 PPL workshop series organised by ASRU. For example, attendees pointed to constraints in the current licensing system, which may in some cases prompt applicants to write very long PPLs and later apply for amendments. Some attendees also voiced concern over the potential for variability in the assessment of applications. ASRU stressed the importance of devoting time to write high-quality first drafts for submission and the need to include details of the experimental design in the applications, and confirmed that the new e-licensing system will aim to further support applicants through the application process, to overcome many of these challenges.
In the final talk of the morning, Dr James Bussell (Wellcome Trust Sanger Institute) and Dr Sara Wells (Mary Lyon Centre, MRC Harwell) looked at whether genome editing with Crispr can provide opportunities for reduction in the creation of genetically-modified animals (GMA). Crispr has already led to a net reduction of animals used for certain GMA projects, e.g. for the creation of knock-out models, because of the efficiency of the system and higher rates of germ-line transmission. However, faster times in the creation of GMAs can enable the completion of more research projects in any given year. Both speakers stressed the paramount importance of quality controls in order to identify unwanted artefacts introduced during the editing process. To this end, all F0 animals must be properly screened and all F1 animals used to establish founder lines must be thoroughly sequenced. International collaborations and consortia set up with the goal to create, share and archive new GMA lines should prevent unnecessary duplication of effort and resource. The great versatility and cost-effectiveness of Crispr will expand the portfolio of targeting applications and the number of species in which projects will be carried out. This will lead to refinements in GMA creation, if not necessarily a reduction in overall numbers. The most interesting avenues of research include: the study of a much wider set of genes and more sophisticated alleles, including modelling human mutations in animal models at a much faster time scale; and the use of strains with different genetic backgrounds, in order to introduce controlled heterogeneity in the experiments and potentially increase the predictability and translatability of animal models. Better GMA models will make experiments more refined – including ameliorating genetic conditions in laboratory animals which also leads to improved animal welfare – and provide more accurate answers to current and future research questions.
The afternoon sessions were organised as roundtable discussions considering the 3Rs (Reduction, Refinement and Replacement).
The first roundtable considered how increased robustness and reproducibility in animal research can be achieved and effectively encouraged. The group reported the reasons why researchers might exclude data from their analysis, and what guidelines should be given to promote the reporting of results in a transparent and useful way. The role of confounders in animal research (e.g. husbandry; sex of the animal/experimenter; timing of experiments; handling of animals; influence of microbiome) was discussed and also the type of errors and cognitive biases most frequently affecting researchers (e.g. from low statistical power; P-hacking and data dredging; confirmation and hindsight biases, to more cultural factors such as potential lack of proper understanding of instrumentation used to obtain data, or a Chinese-whispers type effect in laboratory practices and training). The group went on to propose ways to better educate students and researchers to identify and avoid these pitfalls, which will require several stakeholders to support the right environment for more reproducible science.
The second roundtable discussed systematic review and meta-analysis of animal studies. The group distinguished between different types of review and the quantitative methods involved. The sheer volume of data available in relation to animal research requires a systematic analysis that takes into account the heterogeneity of data/methods reported in the literature. Despite the impact of meta-analysis in the field of clinical sciences and translational research, more needs to be done to support its uptake in pre-clinical studies involving animals, where it could spur improvement in experimental design and avoid unnecessary replications.
The third roundtable looked at how the 3Rs are taken into account in the peer-review process for publication. The group highlighted the need for transparency and easy access to the results of animal research - even and perhaps especially if studies report negative outcomes, which may not lead to a published paper, or do not have the necessary power to reach a definite conclusion. Metadata and experimental details must also be included, to enable research methodologies to be reproduced and refined. Workshop attendees advised that incentives for greater transparency in reporting of research must be provided by funders and research institutions, who, together with publishers, should also consider carefully how the 3Rs have been applied to each study design, both in application and review. Finally, assessment of the quality of review processes and the process of open review was discussed.
The fourth roundtable was devoted to refinement of in-vivo experiments. The group listed obstacles to refinements and the solutions that are available to overcome them. Attendees suggested solutions including knowledge exchange tools allowing researchers to share best practices and experimental tips, greater impetus for the publishing of experimental refinements, and an online repository of experimental details on refinements. The group also highlighted the importance of competence and establishing named experts for refinement, together with an easy process for international experts to come to the UK to demonstrate refinements. Finally, the case of refinement for research involving non-human primates (NHPs) was discussed, including challenges and opportunities for the development of novel refined techniques/procedures for NHP experiments. The establishment of a UK NHP expert group, which supports the sharing of methodological advances and best practices, is enabling the process of refinement on an ongoing basis.
If the event presentations are publicly available, they will be available to download after the meeting:
- Scientific advances in the study of animal welfare
- Communication between tiers of licensing system
- Genome editing and GMA creation: opportunities for reduction
- Workshop on 3Rs practices (round table discussions with facilitators)
To read past reports visit the Animal Science Meetings webpage.
For more information about the Animal Science Meetings, please contact: firstname.lastname@example.org