The authors have declared that no competing interests exist.
While significant medical breakthroughs have been achieved through using animal models, our experience shows that often there is surplus material remaining that is frequently never revisited but could be put to good use by other scientists. Recognising that most scientists are willing to share this material on a collaborative basis, it makes economic, ethical, and academic sense to explore the option to utilise this precious resource before generating new/additional animal models and associated samples. To bring together those requiring animal tissue and those holding this type of archival material, we have devised a framework called Sharing Experimental Animal Resources, Coordinating Holdings (SEARCH) with the aim of making remaining material derived from animal studies in biomedical research more visible and accessible to the scientific community. We encourage journals, funding bodies, and scientists to unite in promoting a new way of approaching animal research by adopting the SEARCH framework.
The contributions that animal models have made to biomedical research and their translation to human health and welfare are undisputed. Many significant medical milestones, including the development of antibiotics (e.g., penicillin [
In order to gauge trends in the use of animals in biomedical research, we devised a Bristol Online Survey (BOS), which was distributed electronically to over 100 biomedical researchers representing more than 90 research institutes or groups. The questions posed aimed to establish how frequently animals were used in oncology, neuroscience, cardiology, and other disciplines and, if not used, if there was a benefit in pursuing this and the motivation behind that choice. Key questions and responses are illustrated in
Participants were asked (A) if they used animals in research, their type (GEM, genetically engineered mouse; PDX, patient-derived xenograft), and if their work would benefit from an in vivo component; (B) if they stored surplus material; (C) if they would be willing to share this material; and (D) if they would support a SEARCH initiative.
Of 135 respondents, 70% currently use animals in their research, and of those who do not, 70% wished to; thus, 90% of the total population surveyed would benefit from using animals in their research. Our survey therefore suggests that the number of animals used in research could rise over the coming years. This reflects the findings from the European Commission that the net number of animals used across Europe in cardiovascular and oncology research rose to 115,000 and 250,000, respectively, between 2008 and 2011 [
For survey respondents who do not use animals currently but wished to do so, the main reasons listed included lack of access to tissue and/or technical expertise and financial constraints. As most of those using animal models had surplus archival material, this led us to consider if it might be possible to find a way to facilitate the removal of these barriers without the need to use more animals.
Clearly many researchers are already addressing the 3Rs; however, concerns have been raised in the community about their appropriate implementation. For example, in some instances, the number of animals has been reduced to such an extent that there are concerns this may be compromising the robustness of experimental data being generated [
Our BOS highlighted that there is significant surplus material generated from most animal studies. This is often archived in-house, stored indefinitely, and frequently never revisited, hence representing a considerable untapped resource; in our survey, two-thirds of respondents currently using animals in their research reported storing excess tissue upon study completion. Encouragingly, 95% of the researchers surveyed were willing to share this material on a collaborative basis. It would make economic, ethical, and academic sense to explore the option to utilise this precious resource in the first instance, before generating new/additional animal models and associated samples, if they already exist elsewhere [
To bring together those requiring animal tissue and those holding this type of archival material, we have devised a framework called Sharing Experimental Animal Resources, Coordinating Holdings (SEARCH) with the aim of making leftover material derived from animal studies in biomedical research more visible and accessible to the scientific community [
The SEARCH framework contains four policies, described below and illustrated in
Crucial to the success of the SEARCH framework is willingness of researchers to share their archived reserves with the research community. Sharing resources has the potential to foster new collaborations, which may not be immediately apparent through current networks, hence increasing research impact. These benefits come at minimum cost to the group requiring the material and mean that the investment by the person who has developed the material has added value.
SEARCH facilitates a web-based database through which scientists can search and identify models that best fit their planned research. This echoes Shared Ageing Research Models (ShARM;
An important part of the SEARCH ethos is the generation of networks of researchers who are willing to connect and share their expertise and resources. SEARCH promotes collaboration over competition, and by doing so, we believe scientific discoveries will be accelerated. Furthermore, in a funding-restricted climate, the ability to bypass the high cost of generating complex animal models will save researchers time and money. These collaborations are also enabling researchers to complete pilot data studies in a short space of time with minimal resources that are being used to strengthen grant applications, aiding early-career researchers in accelerating their careers.
At the crux of the SEARCH initiative is the belief that high-quality research can still be achieved whilst reducing the number of animals used in biomedical research. This can be achieved by preferentially using suitable archived material, without the need to generate unnecessary new models, also helping to maintain the robustness of results whilst using fewer animals.
The SEARCH initiative requires two crucial pieces of infrastructure: a secure database whereby researchers can enter information about material available to share and an accompanying website to host the database. We have developed this for our prototype, SEARCHBreast (
The SEARCHBreast portfolio contains tissues from xenograft and genetically engineered models and also includes an extensive range of patient-derived xenograft (PDX) models. In light of the recent decision by the US National Cancer Institute (NCI) to replace their NCI-60 cell line screen with PDX models [
Using SEARCHBreast as an example, the benefits of using the SEARCH framework are shown in
(A) Traditional experimental workflow for in vivo research. (B) Advantages of utilising the faster and more streamlined SEARCH workflow. (C) Case study 1 showing the advantages of using models deposited in SEARCHBreast. (D) Case study 2 with an example of how SEARCH can accelerate tissue sharing and experimental discovery.
Since its inception in 2014, SEARCHBreast has facilitated the sharing of hundreds of tissue samples from a range of different types of models. This has resulted in sparing 400 animals, including those which would have been culled during transgenic breeding programmes. We predict this number will rise significantly with increasing awareness and adoption of the resource by the scientific community. Consequently, not only does SEARCHBreast have the capacity to save researchers time and money whilst increasing the potential impact of their research, but it addresses the reduction arm of the 3Rs. Similarly, it also aims to address replacement and refinement. The SEARCHBreast website contains links to resources, designed to ensure that researchers are well versed on the latest refinement measures in animal studies. Additionally, SEARCHBreast is committed to promoting the uptake of emerging technologies, e.g., through 3D tissue modelling [
With a perception that animal models are required to generate high-impact publications, a mind-set that will be challenging to reverse and which is beyond the scope of this article, the SEARCH framework offers a fresh way of thinking about in vivo research. As exemplified by those who have signed up to SEARCHBreast and in our BOS that addressed wider disease types, there is support by the scientific community for the complementary approach to animal research we are proposing. Furthermore, the platform we have developed for SEARCHBreast is sufficiently generic to allow easy expansion into other disease types.
Despite recognised benefits in sharing resources, there is often some reluctance by scientists to adopt this fully [
On the back of the success of SEARCHBreast and akin to the ARRIVE guidelines, which encourage comprehensive reporting on experiments and outcomes of research in animals [
(DOCX)
(DOCX)
Replacement, Reduction and Refinement
Bristol Online Survey
genetically engineered mouse
National Cancer Institute
patient-derived xenograft
Sharing Experimental Animal Resources, Coordinating Holdings
Shared Ageing Research Models