SEARCHbreast: A virtual resource to facilitate sharing breast cancer materials and knowledge to benefit the 3Rs

We will develop a searchable database of animal resources related to breast cancer that scientists can share to potentially minimise repetition, at the same time providing access to an under-used resource of archival material. Whilst a virtual resource we anticipate valuable cross collaborations between network partners and new researchers in sharing materials rather than duplicating models to generate such. We recognise there may be concerns given the sensitive nature of the information provided by such an on-line resource hence this would be achieved via secure password protection. This will be complemented by on-line technical expertise accompanied by detailed SOPs describing e.g. how to estimate the number of animals needed in genetically engineered mouse (GEM) and transplantation models of breast cancer and how best to collect and preserve material thereby reducing waste, refining ongoing work and impacting positively on future studies. NC3Rs has already improved the reporting of animal experiments (ARRIVE guidelines). We aim to take this a step further by developing a web based bioinformatics search portal to allow researchers to mine published in vivo studies related to breast cancer to minimise repetition of previous work and access data placed in public repositories. This differs from other resources e.g. DFCI-MGI, KOMP which focus on collection/ storage of single gene/allele knockouts across a broad range of disease types rather than a comprehensive catalogue of the types of model available for study in a single disease. We will hold a series of workshops bringing researchers together to share ideas and resources. These will be hosted by each centre in turn, allowing attendance by scientists throughout the UK. The first will take place within the first 6 months of the project to introduce the UK breast cancer research community to our SEARCHbreast vision. Workshop format will include 4-6 presentations, followed by breakout discussion groups. Subsequent workshops will be themed and could include how to select the best in vivo model to address a specific research question, use of the bioinformatics portal, information on how to establish 3D cell culture models as animal alternatives etc. We will invite one international guest speaker to at least one of these workshops, broadening the SEARCHbreast impact. We will organise a joint NC3Rs-British Association of Cancer Research (BACR) one day workshop to promote the SEARCHbreast blueprint to the wider cancer research community. Dr Holen sits on the BACR Meetings and Training Committee which directs and encourages this sort of initiative. Additionally, we will offer a joint workshop with the Breast Cancer Campaign Tissue Bank, providing advice/training on how to develop and use alternative 3D multicellular tissue culture models developed from human breast tissue to study breast biology. A series of webinars will promote SEARCHbreast beyond the UK. Workshops and webinars will be advertised widely. Once established in breast cancer, in collaboration with NC3Rs, we will arrange meetings with other funding bodies e.g. MRC, CRUK, Prostate Cancer UK, LLR, Heart Research UK, Parkinsons UK etc, to demonstrate the benefits our resource could have impacting on the 3Rs across other disease types. These will be attended by one of the applicants and the coordinator. While most funding agencies now implement data sharing as a routine condition of funding, we will work with these agencies/journals aiming to embed our improved data sharing concept into their application/submission process. Our blueprint is sufficiently generic such that it could be adopted easily for other diseases with the acronym adapted to reflect this while retaining uniformity with the original e.g. SEARCHcardiovascular, SEARCHlung etc. SEARCHlung may be particularly attractive to the NC3Rs given their substantial investment in asthma research. We will actively engage with this group of researchers to promote this.

We will develop a searchable secure database of animal resources related to breast cancer and a specialist bioinformatics resource. The database architecture will have an access layer with a web server and a storage layer with a MySQL database server. It will be populated by information obtained from collaborators. This will host technical information and detailed SOPs, developed by the scientific coordinator, the applicants and their collaborators to unify procedures, collect and preserve animal materials and provide information on how to estimate the fewest numbers of animals needed in various breast cancer models. Access from external users will be password protected to prevent inappropriate use. The SEARCHbreast bioinformatics portal will be online enabling researchers to mine and integrate published data on animal studies in breast disease from multiple different sources. This will be based on existing structure developed by Dr Chelala (http://bioinformatics.breastcancertissuebank.org/ and www.pancreasexpression.org).This will allow scientists to perform fast meta-analysis on publicly available data, overcome possible data analysis obstacles and avoid duplication of work, not only being more time and cost-effective but also impacting on the 3Rs. The resource will be established using Open Source BioMart technology (www.biomart.org) which enables the resource to be easily integrated with complementary resources e.g. Ensembl (www.ensembl.org), international efforts in cancer research e.g. International Cancer Genome Consortium (ICGC; www.icgc.org) and the Cancer Biomedical Bioinformatics Grid caBIG (http://cabig.cancer.gov/). Interoperability with ICGC and caBIG will promote the SEARCHbreast bioinformatics resource not only within the UK, but also internationally. Our Institutes have facilities for professional hosting of planned workshops and webinars. We will manage the project by fortnightly teleconferences and 6 monthly face to face meetings.

According to Pub Med (May 2013), in the past 5 years there were over 200 original research articles published which used animal models of breast cancer in the UK. We expect SEARCHbreast will have a substantial impact on reducing the number of animals without impacting on advancing science, and refinement through sharing methodology As evidenced from the 18 letters of support from leading UK breast cancer researchers, some of whom are current (Prof Watson) or previous (Dr Eccles; Dr Grabowska) recipients of NC3Rs funding, there is an enthusiastic need for this type of resource by the scientific community. Not only is the project sufficiently generic but our project acronym, SEARCHbreast, is such that it could be adapted to reflect this while retaining uniformity with the original e.g. SEARCHprostate, SEARCHcardiovascular, SEARCHlung, SEARCHneurological etc meaning the potential impact could be enormous, having broad and significant impact across multiple disease types. Our secondary aim of promoting the use of alternative humanised models of breast disease will not only impact on the 3Rs but will also provide added value to SEARCHbreast through linking with a complementary national resource of breast cancer materials collected from patients, the Breast Cancer Campaign Tissue Bank. This means SEARCHbreast effectively and imaginatively addresses all 3 values of the NC3Rs. Finally, SEARCHbreast will introduce a completely new way of working with in vivo models and rejuvenate the potential of historically collected tissues from these models. Currently most researchers have little option than to initiate their own in vivo work to generate high impact publications, as there is no easy way to establish whether appropriate material already exists and if so where. We predict that once this resource is up and running it will open up a whole new world of opportunities for cancer researchers, and not only to those specifically working in breast cancer, providing information on the appropriate material, collaborator, experimental guidance etc. In addition, the holders of current collections of archival material will have the opportunity to engage with new projects and collaborators, adding considerable value to existing holdings of animal materials which may otherwise remain unused or potentially destroyed.