Tracing laboratory animals and their data – 3Rs and science reproducibility implications

In laboratory animal research, animal welfare and data reproducibility are considered the two common key points to any study.

On the animal welfare side, the global 3Rs centers placed focus their main activities on new technologies (e.g. organ on a chip) that would aid with replacement or reduction of animal for different type of research and governments, e.g. in the Netherlands, phasing out animal research by 2025 is being proposed (https://www.maastrichtuniversity.nl/will-netherlands-ever-be-free-animal-testing). However, animal research is still needed. Only in a “fully working” and evolutionary constructed compartment such as an integer subject or animal, key biological questions can be addressed. In fact, in support of Covid19 a group of Dutch scientists wrote a comprehensive review regarding why animals are needed (Genzel et al., 2020).

Additionally, several associations such as European Animal Research Association (EARA) and others local or international appeared to communicate to the “non-science” audience, i.e. citizens, the importance animals are playing in the development and testing of new drugs, such vaccine with the current Covid-19 pandemic and to discover mechanisms underlying important human pathologies such as cancer, diabetes or neuro-degenerative diseases.

On the reproducibility side, several aspects have been experimentally proposed to improve the reproducibility within and between experiment (von Kortzfleisch et al., 2020) in a lab or multicentric level (Richter et al., 2010). Different approaches have advantages and limitations, however the push for greater impact factor, for open accessibility (with associated publication costs) and to publish outstanding results might bias the results (mainly only positive). The lack of methodological study information dramatically impacts the possibility of replicating findings.

A recent article published in Science:

precisely touches the point and quotes:

“Most animals used in biomedical experiments are not accounted for in published papers, a first-of-its-kind study suggests”.

A follow up article made by the Medical University of Hannover suggests very similar findings (Wieschowski et al., 2019).

What are the implications?

When animals used in research studies are not reported in scientific manuscript, we might lack important negative findings that would:

• Avoid the need for researchers to repeat experiment that are not working, and potentially reduce the number of animals to be used
• Potentially reduce bias towards positive findings only.

What are the responses to it?

The authors of the original paper (https://preclinicaltrials.eu/) as well also the German Centre for the Protection of Laboratory Animals started databases in which scientists can discuss their protocols and procedures (https://www.animalstudyregistry.org/asr_web/index.action). Basically, once a study series is registered, the registration receives a DOI which is similar to a patent to preserve the intelligence properties. However, since this is not yet legally required to register those protocols neither by funding agencies nor by local animal welfare offices, scientists might still want to focus their time on the research. Therefore, the compliance is still low.

To increase such compliance, the Berlin institue of quest started an award of 1.000 Euros for scientists from Berlin centres if they preregister they animals (https://www.bihealth.org/en/research/quest-center/calls-and-awards/quest-calls-and-awards/preregistration).

A totally different approach started in the UK called Octopus, tries to cover not only the registration of the studies but also involves funding agencies. This is to ensure that authorship are meritocratically assigned and data is transparent (https://www.sciencemag.org/careers/2018/11/meet-octopus-new-vision-scientific-publishing). The near future will tell which of such activities will work to improve actual reporting of animal’s research.

How are digital technologies helping with animal traceability?

Systems such Animal Management Systems (AMS) software are mainly used by facility managers to manage activities like breeding, track colonies, document animal health records, handle animal movements, schedule tasks for staff, and account for animal per-diem costs. AMS systems are replacing cumbersome excel or paper-based documentation that is prone to data entry errors, can easily be lost, and most of all, is cumbersome to search and extract valuable information from. Data and metrics from AMS reports can easily be made available to scientists, that have to more and more be acquainted with such systems to request animals authorized on their protocols. Tracing of animals has also benefitted from integration with electronic animal and cage tags (RFID, barcode, QR code or other) that helps identify with a quick scan where animals are and check on their health status. Therefore, in case of sickness, animal history such can easily be reconstructed in a precise manner. Besides obvious animal welfare benefits, this is very helpful at times of inspections from the authorities of the animal facility and experiment compliance. Scientists can demonstrate what was applied for is what was actually performed and reported.

Cage manufacturers have developed technologies to integrate with AMS systems for cage tracking resulting in a better overview on the animal facility activities. These technologies also track animal activities to assess welfare and phenotype the animals while in the home cage where animals are born.

Several studies have been conducted using this digital home cage technology (e.g. DVC? from Tecniplast S.p.A. (link below)

to obtain in a seamless way 24/7 a full phenotype in different fields of research of circadian rhythms, aging, neuroscience and cancer research. It is enabling more and more the researchers to conduct scientific experiments with those systems. Such systems, in order to be used, require a cage/animal registration so that for instance activity calculation is accounted by the numerosity within cage (for data normalisation). Those data can alternatively be automatically inserted in most AMS systems and are transferred directly in the DVC?. This allows tracking of all animals as well as their wellbeing, helping the facility manager or veterinarian reduce potential animal distress and maintain the highest standards of care. This option could increase the compliance of reporting of animals that are in a study and help publishing scientific findings.

Lastly, the opportunity of automatically recognising and tracking cages (and its animal content) with the DVC? between difference locations in a facility (cage and animal census) aids to addressing several problems in the lean management process as stated below:

Determining animal census numbers can aid in calculating per diem charges, assigning work tasks, projecting the budget, allocating resources, allocating space for projects, and calculating the amount of assigned and unassigned space. It is also an important tool to monitor numbers of animals approved on protocols"

Source: https://www.ncbi.nlm.nih.gov/books/NBK500430/

Should such systems become part of the welfare officer system to check remotely animal welfare (e.g. in Covid19 time where all the care take are not present) and part of the science to be conducted, it would be easier to publish transparent data, or at least to report all valuable “negative” data. The benefit will be that animals could be used in a more precise and senseful way. It potentially could help create digital libraries for specific experiments that could be shared among scientists (e.g. control group) to improve data reproducibility across labs.

More and more digital technologies will be implemented to scalable home cage systems. This will ensure that one-spot experiments run with other technologies correlate with what happen in the other 99% of the animal’s time, namely, when they are in their home cage. The reward will be improved transability, increased animal welfare and reproducibility.

Acknowledgements

I would like to thank Eric Rieux, Vicepresident of Sales and Marketing at BSI, for the AMS contribution part.

References

Genzel, L., Adan, R., Berns, A., Van Den Beucken, J., Blokland, A., Boddeke, E., Bogers, W.M., Bontrop, R., Bulthuis, R., Bousema, T., Clevers, H., Coenen, T., Van Dam, A.M., Deen, P.M.T., Van Dijk, K.W., Eggen, B.J.L., Elgersma, Y., Erdogan, I., Englitz, B., Fentener Van Vlissingen, J.M., La Fleur, S., Fouchier, R., Fitzsimons, C.P., Frieling, W., Haagmans, B., Heesters, B.A., Henckens, M., Herfst, S., Hol, E., Van Den Hove, D., De Jonge, M.I., Jonkers, J., Joosten, L.a.B., Kalsbeek, A., Kamermans, M., Kampinga, H.H., Kas, M.J., Keijer, J.A., Kersten, S., Kiliaan, A.J., Kooij, T.W.A., Kooijman, S., Koopman, W.J.H., Korosi, A., Krugers, H.J., Kuiken, T., Kushner, S.A., Langermans, J.a.M., Lesscher, H.M.B., Lucassen, P.J., Lutgens, E., Netea, M.G., Noldus, L., Van Der Meer, J.W.M., Meye, F.J., Mul, J.D., Van Oers, K., Olivier, J.D.A., Pasterkamp, R.J., Philippens, I., Prickaerts, J., Pollux, B.J.A., Rensen, P.C.N., Van Rheenen, J., Van Rij, R.P., Ritsma, L., Rockx, B.H.G., Roozendaal, B., Van Schothorst, E.M., Stittelaar, K., Stockhofe, N., Swaab, D.F., De Swart, R.L., Vanderschuren, L., De Vries, T.J., De Vrij, F., Van Wezel, R., Wierenga, C.J., Wiesmann, M., Willuhn, I., De Zeeuw, C.I., and Homberg, J.R. (2020). How the COVID-19 pandemic highlights the necessity of animal research. Curr Biol 30, R1014-R1018.

Richter, S.H., Garner, J.P., Auer, C., Kunert, J., and Wurbel, H. (2010). Systematic variation improves reproducibility of animal experiments. Nat Methods 7, 167-168.

Von Kortzfleisch, V.T., Karp, N.A., Palme, R., Kaiser, S., Sachser, N., and Richter, S.H. (2020). Improving reproducibility in animal research by splitting the study population into several 'mini-experiments'. Sci Rep 10, 16579.

Wieschowski, S., Biernot, S., Deutsch, S., Glage, S., Bleich, A., Tolba, R., and Strech, D. (2019). Publication rates in animal research. Extent and characteristics of published and non-published animal studies followed up at two German university medical centres. PLoS One 14, e0223758.