Impact case study database

The reverse genetics technology covered by the patent to Fodor, Brownlee and colleagues [1] is used by AstraZeneca to manufacture its seasonal flu vaccine each year, ensuring that it contains the most relevant and up-to-date virus strains [A]. The technology was licensed to MedImmune Inc. (now part of AstraZeneca) to improve production of the initially licensed trivalent formulation vaccine, branded as FluMist in the U.S. and Fluenz in Europe, which is a live attenuated influenza vaccine (LAIV) and given intranasally. The technology was first applied to generate FluMist as trivalent (LAIV) vaccine for the 2008 and following seasons. In 2012, it was used to generate the quadrivalent vaccine (FluMist Quadrivalent, LAIV4) protecting against four flu virus strains. FluMist Quadrivalent and Fluenz Tetra, licensed in the US and Europe respectively, have been in use worldwide since 2013.

Prior to 2008, LAIV vaccine virus strains were generated by an inefficient empirical process of natural genetic reassortment between the wild-type influenza virus and the LAIV Master Donor Virus. The helper-free reverse genetics technology in the patent granted to the University of Oxford researchers [1] was described by AstraZeneca as vital to the success of FluMist and Fluenz:

*“This technology has been of critical importance to the success of Flumist/Fluenz as it is a quicker, simpler and more robust way of generating the vaccine strains than previous technology. This is important as the timelines to manufacture seasonal influenza vaccines are incredibly challenging. This technology has increased the speed of the vaccine to market, ensuring that the maximum number of patients are able to receive the vaccine before the start of the Flu season.*” [Ai]

The REF2014 impact case study described the health benefits from use of FluMist in the U.S. from 2008 to July 2013, the economic benefits to MedImmune from FluMist sales and the impact of the reverse genetics technology for the production of vaccines against the 2009 swine-origin pandemic influenza virus. This impact case study describes wider health benefits to populations from use of FluMist and Fluenz in national childhood vaccination programmes the UK and other countries from August 2013 to 2020.

In 2012, the Joint Committee on Vaccination and Immunisation (JCVI) recommended the stepwise roll-out of a universal childhood influenza vaccine programme with the trivalent vaccine Fluenz [B]. The programme was recommended to cover all children 2-17 years of age because children experience some of the highest rates of infection during seasonal epidemics and also play a key role in the spread of the virus. Recommendation of the programme was based on its predicted direct and indirect impact, with the aim to provide individual protection to children themselves and, by reducing transmission across all age groups, protection to vulnerable members of the population. The commonly used trivalent inactivated influenza vaccine was documented to have poor immunogenicity in children and therefore the intranasally administered LAIV, which early efficacy studies suggested provides excellent protection for children, was recommended. The LAIV vaccine is suitable for all children except those who are clinically immunosuppressed.

Flu vaccination has been offered to children in a phased roll-out since 2013. The programme was initiated with the trivalent LAIV and then changed to LAIV4 following European approval in 2013. Fluenz was introduced in the 2013-14 flu season, UK-wide, to all children aged 2 and 3 years [C]. The vaccine was subsequently rolled out to include older children, starting in 2014/15 with school year 1 and expanding by a single year group each season up to the current 2020/21 season when it has been offered to children who have just started secondary school in year 7. The programme combined stepped roll-out nationally with geographical school-based pilots covering the full 4-11 age range and additionally, in some cases, secondary school children (11-13 years). Over 20,000,000 children in the UK aged 2-17 have been vaccinated since the programme was rolled out in 2013 [A]. Vaccine uptake has generally increased over the period of introduction of the vaccine and ranges from 30% to 80% [D]. The intranasal formulation may have helped to make vaccination more acceptable in this age group by overcoming needle phobia, which is estimated to affect at least 10% of children.

Since its introduction in 2013/14, the programme has been shown to positively impact influenza-related outcomes in children and the community as a whole, with the greatest impact observed in pilot areas where children in **all primary school years have been offered vaccination. Early results from the school-based pilot in England in the 2013/14 season followed by an evaluation of the impact of the programme and continuation of the pilots in the 2014/15 season informed the programme’s national roll-out. The early results indicated that the programme in the pilot areas had a significant impact on flu in multiple age groups [E]. Evaluation of the 2014/15 season confirmed this [F]; vaccinating primary school-aged children resulted in direct protection in children aged 5–10 years, with significant reductions in cumulative GP influenza-like illness (ILI) consultations (94% reduction), emergency department respiratory attendances (74% reduction) and confirmed influenza hospital admissions (93%) as well as a reduction in excess respiratory mortality. These changes were also associated with indirect protection of other age groups. Information captured by the Royal College of General Practitioners network for the 2014/15 season showed that cumulative GP ILI consultation rates in individuals aged 50-70 years were lower in pilot areas that vaccinated primary and secondary (children aged 11–13 years) school children compared with non-pilot areas (consultation rate: 3.4 per 100,000 versus 17.4 per 100,000, respectively); similar results were also seen for swab positivity (7.7% positive versus 29.5% positive, respectively) [D]. These changes were accompanied by a significant reduction in cumulative GP ILI consultations at a population level i.e. in individuals aged 17 years or more (59% reduction) [F].

A review of clinical impact of the programme across all UK nations over six influenza seasons provided further data about vaccine effectiveness and influenza-related outcomes in different age groups [D]. Vaccine effectiveness against laboratory-confirmed influenza in children aged 2-17 in primary care in the UK varied from year to year. Whilst the data were not available for the first two seasons, the vaccine prevented 58% of flu cases in 2015/16, 66% of flu cases in 2016/17, 27% of flu cases in 2017/18, 49% of flu cases in 2018/19 and 45% of flu cases in 2019/20 [D,G]. In the 2015/16 season, vaccinating primary school-aged children was associated with lower ILI GP consultations and swab positivity in children of primary school age and those aged <5 years for pilot versus non-pilot areas. In addition, these changes were associated with significant indirect protection of the wider population, with reductions in the adjusted cumulative primary care indicators for ILI (63% reduction) and swab positivity (estimated 48% reduction) in individuals aged 17 years or more from pilot areas compared with those from non-pilot areas. In Northern Ireland, significant reductions were observed in GP in-hours consultations and out-of-hour calls for ILI both overall (61% reduction in GP in-hours; 51% reduction in out-of-hours) and in children aged <14 years (38% reduction in GP in-hours; 39% reduction in out-of-hours), in seasons with full implementation of the programme (2014/15 to 2016/17) compared with pre-programme seasons (2010/11 to 2012/13).

There is also evidence that vaccination in young children through the programme may contribute to a reduction in prescribing of amoxicillin in children. Amoxicillin is one of the most prescribed antibiotics in primary care and is indicated for two common complications of influenza: community-acquired pneumonia and acute otitis media. Whilst further studies are needed to confirm the size of the effect, a study in 2017 found a 12.8% to 14.5% reduced rate of amoxicillin prescribing during periods of LAIV-induced immunity in preschool children [H].

As the UK’s programme expanded beyond the pilot areas and to older children, it has become more challenging to fully capture the benefits of the vaccination programme especially those on the wider community. Nevertheless, the data available to date demonstrate a positive impact of the UK’s programme for a range of key surveillance indicators. In addition, an economic analysis of influenza vaccination in England found that vaccination (across all age groups) helped to avert between 5,678 and 8,800 premature deaths per year [l].

Cost-effectiveness studies were amongst those that provided evidence supporting extending the UK influenza vaccination programme to children in 2012. These studies continue to support the vaccination of children to optimise the benefits of influenza vaccination programmes. A recent cost-benefit assessment of influenza vaccination in England concluded that “ public policy should focus on improving coverage rates among young people aged 15 years or under, as the most efficient approach to improving overall benefits relative to costs” [I].

Influenza vaccination of healthy children with Flumist/Fluenz has been introduced through national vaccination programmes in the United States, Canada, Finland and other European countries, and South Korea. Up to 22,000,000 doses of Flumist/Fluenz are manufactured in Liverpool, UK each season and used to vaccinate people in the US, Europe, Israel, Canada and South Korea [A].

FluMist has been used in the US since before 2013 in children from the age of 2 upwards. FluMist Quadrivalent is approved for use in healthy individuals aged 2-49 and was first introduced in the 2013/14 season. The US Advisory Committee on Immunization Practices recommends annual influenza vaccination for everyone aged 6 months and older with any licensed, influenza vaccine that is appropriate for the recipient’s age and health status. FluMist has been used in most seasons since 2013/14 and from 2018/19 onwards [J].

In Finland, influenza vaccine has been given free of charge to all eligible children (6-35 months) since 2007. To enhance vaccine uptake, Fluenz (Fluenz Tetra) was introduced in the 2015/16 season. Since then, all 2-year-old children have been eligible for vaccination with either Fluenz or trivalent inactivated influenza vaccine, without a recommended preference. Vaccination coverage has increased steadily following Fluenz introduction – from 11% of children aged 6–35 months in 2014/15 to 24% in 2017/18 [K]. Vaccine effectiveness of Fluenz measured over three seasons (2015/16 to 2017/18) was comparable with that in the UK – between 20% and 54%. Since the 2018/19 season, children aged between 2 and 6 years have been offered the Fluenz vaccine free of charge as part of the national vaccination programme. For the 2020/21 season, 116,000 doses of Fluenz were procured for the programme which is sufficient for four out of ten children in this age range [L].

With COVID-19 putting pressure on hospitals in winter 2020/21, Ireland offered Fluenz for the first time to children aged between 2 and 12 years from October to December 2020 [M].

Whilst a preliminary prototype of BRAHMS had existed previously, the research on Leucaena [1] drove a step-change enhancement in functionality [2,3,4] that led during the REF2021 impact period to BRAHMS becoming a world-leading data management system for managing complex datasets from natural collections and surveys. BRAHMS database software now manages data from natural history collections, botanic gardens, seed banks, field surveys, taxonomic research and biogeographic study.

BRAHMS facilitates data management and integration, which then enables environmental policy implementation and monitoring. BRAHMS has been adopted by governments, national and international institutions, businesses, NGOs, educators and local peoples to support science-led policy-making on the management and use of biodiversity. At December 2020 it was enabling the management of data resources including more than 12,000,000 botanical records in over 60 countries. This represents about 5% of the total number of botanical records held in herbaria, globally. The most recent major release of the software (version 8) was published in May 2019 with a wide range of features [A]. The previous version remains available without needing a licence, enabling the widest usage.

Effective policy application and monitoring requires accurate species names. The widespread proliferation of different names for the same cultivated plant is increasingly causing confusion for plant breeders, horticultural traders, gardeners and the general public worldwide: it makes it more difficult not only to provide accurate horticultural information but also to enforce plant breeding rights. Furthermore, a standardised approach to plant names is critical to satisfy the increased requirement to document and track the movement of cultivated plants for legal and biosecurity issues. In 2019, BRAHMS enabled researchers at the Royal Horticultural Society (RHS) to develop a new approach to determine cultivated plant names, by ranking the various names a plant may have. In partnership with the RHS, the functionality of BRAHMS was extended to manage data on names, providing a global naming standard promoted throughout the horticultural world, and at October 2020 was being used by the RHS for 358,532 species [C(i)]. Similarly, the National Herbarium of the Netherlands were using BRAHMS to manage 659,878 species names in 2019 [C(ii)].

In 2020, BRAHMS was formally adopted by the RHS for managing their living plant collections, not only through the transfer of taxonomic data, but also the tracing of plant records across their five gardens to help limit the spread of pests and diseases [B]. The Head of Horticultural Information described other benefits soon after licencing the new version:

“Already we have been able to use BRAHMS to replace a legacy system that was serving information to the RHS website – a move which has sped up the rate at which we are able to add details about new plants and make those available for free to the public to aid gardeners to choose the correct plant for them. This is a significant part of the RHS’s charitable remit.” [B(ii)]

The Royal Botanic Garden Kew's Millennium Seed Bank Partnership (MSBP) is the largest ex situ plant conservation project in the world. This network of 120 seed banks covers 95 countries and is a fundamental resource for the more than 150 active collaborating partners. It set a first target of securing seed collections from 25% of the world’s flora by 2020. In 2017, BRAHMS software integrated data from across this network into a single, online resource, the MSBP Data Warehouse [D(i),(ii)]. For the first time, data on native plants conserved in these seed banks were published as a single, globally accessible online resource and provides an opportunity for MSBP partners to query the status of their seed collections alongside other collections from across the MSBP. Partners frequently compare seed accession location data with known distribution of a species to establish if a new collection is likely to represent new genetic diversity in the collection, while germination test data enables seed bank managers the best insight into the treatments and conditions required to achieve optimal seed germination. Moreover, the MSBP Data Warehouse provides the means to monitor collection targets, thus providing management tools to monitor progress towards the 2020 target [D(ii)]. The MSBP’s Africa Programme Coordinator wrote in 2020 that:

“I have been working with the BRAHMS software for many years and we have made some strong steps in achieving the integration of data from the collection event such as linking the seed collection with the herbarium voucher, or being able to monitor the management of the living plants generated from our seed.”

MSBP Africa Programme Coordinator, RBG Kew [E(i)]

At November 2020, the MBSP’s BRAHMS database held records of over 228,000 seed collections from 204 countries and territories and 96 different seed banks, providing information on over 48,800 different species [E(ii)]. The database also held data on 215,700 germination tests and over 9,870 x-ray images. More than half of these records were by then available online [E(ii)].

Large-scale international and national biodiversity networks have used BRAHMS to gather and integrate primary botanical data from multiple institutions.

Data stored in BRAHMS, from around 2,000,000 specimens across 10 institutions in South Africa, Namibia, Botswana and Malawi contribute directly to the botanical infrastructure of this region. Specimen-level records from the three herbaria at the South African National Biodiversity Institute (SANBI), the National Herbarium in Pretoria, the Compton Herbarium in Cape Town, and the KwaZulu-Natal Herbarium in Durban, have since 2013 been digitised, stored, curated and improved using BRAHMS. This data is used for threatened species assessments, conservation planning, management and policy advice; is disseminated to other international projects; and helps to fulfil commitments under the 2020 targets of the Global Strategy for Plant Conservation [F]. The Deputy Director of Biodiversity Information and Planning at SANBI wrote that

“BRAHMS is thus crucial for SANBI to be able to fulfil its mandate… BRAHMS helps to raise South Africa’s international profile as a valuable source of biodiversity data for our unique flora. This would not be possible without the technical and training support provided by the University of Oxford as well as the extensive source of user documentation ... “ [F]

BRAHMS is used extensively in most regions of Brazil managing data in over 80 academic and non-academic institutions. In 2016, a review by the Central Environmental Resources Institute (CRIA) highlighted the key role of BRAHMS in strengthening the botanical infrastructure across Brazil. Approximately half of CRIA’s botanical data (around 3,500,000 records) have been contributed by 82 separate BRAHMS databases. In 2019, approximately 1,500,000 plant and fungi records (including 11,000 images) were accessed daily. These data and images have been used for a wide range of purposes, including improving species identification, monitoring species distribution, and setting conservation priorities [G].

The National Herbarium & Botanic Gardens of Malawi have digitised over 32,000 specimens, recording data in BRAHMS, and report uses including species conservation (Malawi National Redlist Project), relating pollinator biodiversity to agro-ecological interventions, and inventories of threatened, invasive and alien plants [H]. In many of these, BRAHMS’ compatibility with the Global Biodiversity Information Facility (GBIF) data standards [I] enabled biodiversity information to be made more accessible.

The National Herbarium of the Netherlands (Naturalis) uses BRAHMS to enable access to botanical data from approximately 7,000,000 specimens collected from Europe, South East Asia, West Africa and South America. Other non-academic institutions, such as the National Museums of Kenya and the New Zealand National Forest Service, use BRAHMS to manage fine-scale botanical data.

Fine-scale mapping of species’ distributions using BRAHMS (e.g. [3]) has enabled areas with high concentrations of rare species to be identified, prioritised and monitored. In 2015, BRAHMS enabled the Forestry Research Institute of Malaysia (FRIM) to assemble evidence to underpin proposals of High Conservation Value Forests (HCVFs). HCVFs are defined by the Forestry Stewardship Council to be of outstanding importance due to their biodiversity, environmental, landscape or socio-economic attributes. To identify HCVFs, it was essential to understand and integrate the distributions and rarity of all plants living in the forests of the Malay Peninsula. Using BRAHMS, FRIM assembled data from herbaria, published literature and unpublished field-records from around the world. Using BRAHMS’ high-functionality mapping and analysis tools, areas of Malaysia containing high concentrations of rare species were identified. At least 33 HCVFs for a range of Critically Endangered and Endangered species were approved by the Malaysian government as a result of these data and contribute to the implementation of a Sustainable Forest Management protocol in Malaysia [J].

BRAHMS is licensed worldwide through Oxford University Innovation (OUI). A number of users of earlier versions have decided to adopt BRAHMS as a paid and supported product to ensure the longevity of their high-value datasets and management systems, reflecting its importance to their operations. Since 2015, BRAHMS has had 23 paying licensees worldwide, including the USA, New Zealand, Malaysia, Africa and Europe, spanning herbaria, arboreta, botanic gardens and seed banks. Since 2015, 122 organisations have commenced free 60-day trials of BRAHMS, which has facilitated discussions about longer-term usage and customisation [K].

The integrated techniques developed by WildCRU for modelling ecological processes and assessing their impacts on wild felids have had widespread impact on decisions and implementation by policy-makers, conservation practitioners, and planners worldwide.

Ulu Muda, Malaysia

In Malaysia, WildCRU research had a crucial role in delivering protection of Ulu Muda Forest, which is one of the most critical pristine forests remaining in the region, rich in biodiversity and providing essential habitat for vulnerable clouded leopards and many other species. Ulu Muda covers approximately 160,000ha and is also the source of water for 4,200,000 people and at least 40% of the country’s rice production. WildCRU studies of rare leopards in Ulu Muda in 2015 and integration of data into Malaysian national planning for clouded leopards in 2017 (using tools exemplified in [2, 4, 5]) attracted significant interest from the Department of Wildlife and National Parks Peninsular Malaysia (PERHILITAN) [A]. Their camera-traps documented illegal logging and poaching at Ulu Muda, which WildCRU and a local conservation group communicated in 2017 in a video “Saving Ulu Muda” [Bi] that was broadcast on national television as well as online [A], and WildCRU work was extensively highlighted in press coverage [Bii]. These videos were coupled with a petition that gained over 102,000 signatures [A, Biii]; one month after the petition was sent the government, all logging permits at Ulu Muda were revoked in September 2018 [A, Biv].

Myanmar

In 2018 in Myanmar, the results of WildCRU analyses (including [6]) of the impact of many major past, ongoing and potential future developments on clouded leopard populations were presented to senior government officials including the Director General of the Myanmar Forest Department, and international non-governmental organisations [C]. After meeting D. Macdonald, Aung San Suu Kyi (State Counsellor of Myanmar and Minister of Foreign Affairs) assigned people from her personal charity to facilitate application of the WildCRU methods in Myanmar [C]. Based on this research and using the WildCRU ‘toolkit’, the Forest Department modelled important wildlife corridors for felids at the country-scale and applied this information to assess the environmental feasibility for all development projects in the country [C]. These development assessments included proposed mining exploration in northern Myanmar and proposed highway development project in central Myanmar.

Wider South-East Asia

Similarly, in Borneo from 2017, WildCRU used their multispecies modelling tools (e.g. [5, 6]) to assess the likely impact of 59 components of the infrastructure and land-use plans in Sabah state (the Sabah Structure Plan 2033) on the vulnerable Sunda clouded leopard; notably, Sabah has one of the highest deforestation rates in the world. This analysis influenced senior government officials, exemplified by a meeting with the Sabah Deputy Chief Minister in 2018 [A, Di]. WildCRU multi-species and multi-scale analyses (exemplified by [5, 6]) have also helped several other governments in South-East Asia to identify important areas for conservation, such as core habitats and corridors. For example, in Thailand, a key breeding area for tigers was identified as threatened by a proposed major roadway and dam project, contributing to UNESCO strongly advising against the development in 2017 [Dii]. In Bhutan, WildCRU has worked closely with the government to develop a toolkit identifying key areas for conservation [Diii]. In Sumatra, Indonesia, in 2020 the government planned designation of high conservation value areas based on WildCRU methods, and adopted the framework for identifying core areas and corridor networks between protected areas [Div].

KAZA Transfrontier Conservation Area

The WildCRU landscape conservation planning toolkits and analyses (based on research including [1, 3]), including approximately 6,000,000 camera-trap images, have directly guided development and conservation decisions in the Kavango Zambezi Transfrontier Conservation Area (KAZA TFCA, spanning Botswana, Zimbabwe, Zambia, Namibia, Angola), an area of approximately 520,000km² with exceptionally high global biodiversity value. WildCRU has worked closely with KAZA officials [E] and led workshops in Botswana and Zimbabwe in October 2018, attended by senior government and conservation officials [E, F]. Through these close interactions, WildCRU research has directly influenced incorporation of wildlife corridors and landscape connectivity into large-scale development and conservation strategies including: Master Integrated Development Plan for KAZA TFCA (2015-2020); Wildlife Dispersal Areas for KAZA TFCA (2013-2015); and KAZA Carnivore Conservation Coalition Strategy (2018-2022) [E, F]. The Executive Director of the KAZA Secretariat stated that the impact of the WildCRU research has been “ hugely significant and at a landscape-level scale” [E]. Within KAZA, the research has also shaped national policies, including in Zimbabwe. The Zimbabwe National Parks and Wildlife Management Authority National Coordinator for Transfrontier Conservation Areas stated that WildCRU research was important for landscape-level scale integrated land-use planning that evaluated the compatibility of activities such as mining, tourism, conservation around Hwange National Park, directly influencing the Authority’s thinking on lion conservation, and providing “ pivotal” information for policies including the Hwange National Park General Management Plan (2016-2026) and Lion Conservation Strategy for Zimbabwe (2020-2025) [F].

Kenya

Similarly, in Kenya, WildCRU modelling (based on research including [1, 3]) supported the government and the Kenya Wildlife Service (KWS) to develop the national Recovery and Action Plan for Lion and Spotted Hyena 2020-2030 [G, Hi], for these two species with important ecological roles in savannah ecosystems and major population declines. Launched in August 2020, the action plan takes landscape-level approaches towards conservation and includes utilisation of the WildCRU decision support toolkit methods [Hi]. The Minister for Tourism and Wildlife acknowledged the key role of WildCRU in the action plan [G, Hi]. The Chief Executive of a landowner membership organisation (The Kenya Wildlife Conservancies Association), also representing Maasai communities, commented that “ the National Strategy, and the consultations and research that underpinned it…have already changed things here, both on the ground and at a policy level” and that the work is “ a beacon of how conservation should interact with local people and with local decision makers to deliver impact on the ground” [Hii].

International species status

WildCRU research has informed international decision-making on the status of several felid species. The International Union for Conservation of Nature (IUCN) Red List of Threatened Species is used globally as a basis for conservation and policy changes, and WildCRU data has been essential for assessments that have determined the status of felids in South-East Asia (confirmed by [A]): Sunda clouded leopard (vulnerable, 2015) [Ii]; Borneo bay cat (endangered, 2016) [Iii]; marbled cat (near threatened, 2015) [Iiii]; leopard cat (least concern, 2015) [Iiv]; and, crucially, the up-listing of the Indochinese leopard to critically endangered in 2019 [Iv], based almost entirely on WildCRU work [J]. In 2017, lions were added to the UN Convention on Migratory Species (CMS), and the Head of the Terrestrial Species Team for CMS confirmed that evidence from WildCRU research on lions ([1, 3], as examples of a large body of work) was pivotal in this decision [Ki]. Furthermore, the Chief of the UN CITES (Convention on International Trade in Endangered Species of Wild Fauna and Flora) Science Unit stated that WildCRU work on lions has been “extremely influential” on decision making, including on measures for conserving lions and other big cats that the 183 Parties to CITES agreed to in 2016 and 2019 [Kii].

National action to protect rare leopards

The work by WildCRU and their collaborators on Sunda clouded leopards in Sabah, Borneo (Malaysia) (e.g. [4, 5]), underpinned an international workshop on conservation of this species in June 2017 and the drafting of the first state-level action plan for the species [A]. This ‘Sunda Clouded Leopard Action Plan for Sabah’ was endorsed by the state government in June 2019 [A], and was instrumental in securing a donation of approximately GBP700,000 (MYR4,000,000, 06-2019) from the Sime Darby Foundation to support anti-poaching work, in June 2019 [A].

In Cambodia, camera-trap work by WildCRU (exemplified by [2]) since 2012 showed an alarming population decline in one of last remaining populations of Indochinese leopards. From 2014, this resulted in WWF-Cambodia and local authorities changing their patrolling regime to focus on hotspots of remaining leopards and hiring 6 new WWF staff in 2016 to improve efficacy of patrols [J]. The WildCRU data also led to implementation of snare-removal teams (which have removed 5,000 snares between 2016 and 2020), and identification of poacher movements, leading to arrests [J]. As an emergency measure to protect the last leopards, WWF implemented fencing of areas where the leopards occur, identified from the camera-trap data. WWF-Cambodia stated that the WildCRU research “ has led to vast improvements and increased efficiency of our law enforcement activities, approved by the Cambodian government” [J].

In Iran, WildCRU used species distribution models (e.g. based on [2, 4]) to map potentially suitable habitats for the Persian leopard and Asiatic cheetah. In 2019, these data influenced the inclusion of the Persian leopard in the Central Asian Mammals Initiative (CAMI) and the Iranian Department of Environment’s submission for future inclusion in the UN CMS [Li]. The WildCRU work directly resulted in the development of a tool kit for managing conflict between human and large carnivores, including Persian leopards, which was distributed to more than 3,000 Iranian rangers, and formally adopted and published in 2019 by the Iranian Department of Environment [Lii].

Shaping conservation actions for the Scottish wildcat

WildCRU applied their camera-trap and GPS collar approaches (exemplified by [1, 2]) to survey wildcats in Scotland from 2010 and 2015, respectively [M]. The camera-trap data and WildCRU researchers made important contributions to the 2015 Scottish Wildcat Conservation Action Plan, including identifying 6 priority areas for conservation efforts [M]. These data, and WildCRU work on wildcat genetics, led to the IUCN concluding in 2018 that the wildcat was at risk of immediate extinction in Scotland without population reinforcement [M]. In response, the European Union funded the Saving Wildcats project (2019-2015) to re-introduce and reinforce wildcat populations, with WildCRU GPS data contributing to choices of re-introduction sites [M].

Trees are fundamental to terrestrial ecosystems and the lives of people across the planet, contributing to at least 13 of the United Nations’ 17 Sustainable Development Goals. It is estimated that 420,000,000ha of forest were lost through conversion to other land uses between 1990 and 2020. Many forests have become fragmented, with nearly 20% of global forest area being in patches less than 1,000ha in size, 9% in fragments with little or no connectivity, and approximately 12,000 tree species at risk of extinction. University of Oxford research showing the importance and relevance of remnant trees in fragmented agroecosystems to conservation, forest management and associated rural livelihoods has led to changes in policy, training, and professional practice internationally.

**UN Food and Agriculture Organisation: Boshier was commissioned to write a chapter (“ Fragmentation, landscape functionalities and connectivity”) for the UN Food and Agriculture Organisation (FAO) report on the state of the world’s forest genetic resources, which drew extensively on the University of Oxford research, including [1,3,4] and was published in 2014 [A]. This report directly informed the FAO’s Global Plan of Action on Forest Genetic Resources in 2014 [B], which identified 27 strategic priorities for forest conservation.

Canadian Species at Risk Action Plan: In 2015, the Canadian Species at Risk Action Plan for the endangered cucumber tree ( Magnolia acuminata) - the only native Magnolia in Canada - was changed, directly influenced by University of Oxford research [4] [C(i)]. The previous Action Plan (since 2007) had specified a minimum of 10 mature trees for a habitat to be considered critical. Citing the research [4], the 2015 Action Plan states “single trees and small populations contribute to genetic connectivity across the landscape; fragmented sites can play a role as functioning elements of a larger population as ‘stepping stones’ between sites”. Under the revised criteria, the number of critical habitat sites increased from 7 to 15. Importantly, a 2019 assessment of Canadian Environmental Sustainability Indicators found the cucumber tree is showing progress towards population and distribution objectives [C(ii)], indicative of success of the Action Plan.

UK Woodland Trust: The Woodland Trust is a major woodland conservation charity, owning more than 1,000 woods in the UK, managing woodlands for conservation, and campaigning to protect woodlands. In 2017, the Woodland Trust adopted a policy to support conservation of single trees and small woods, as part of the wider network of woodland habitat across the UK [D(i)]. According to a Woodland Trust Regional Director [D(ii)], this policy was based on University of Oxford research [3,4] showing the importance of single trees and small woods as ‘stepping stones’ and habitats. This policy on the importance of single trees and small woods also influenced the Woodland Trust’s Emergency Tree Plan campaign and the Woodland Trust and the independent Committee on Climate Change’s recommendation to the UK Government to implement “ significant expansion of trees outside woods to achieve net zero carbon” [D(ii)].

**Forest Stewardship Council (FSC), Chile: The FSC is a global organisation setting standards for certification of sustainable forest management and sustainably managed forest products. Forestry companies must conform to FSC requirements (criteria and indicators) to achieve FSC certification, which is increasingly essential for access to international markets, and consequent economic benefits. In 2016, FSC-Chile updated their certification standards, and criteria and indicators for the “ Identification, Management and Monitoring of High Conservation Value forests”, citing University of Oxford research [3,4] with respect to important local conditions, such as temporary habitats and migration sites [E].

The University of Oxford research has directly shaped training delivered by the researchers and non-governmental organisations to forestry professionals, reaching at least 300 people from more than 25 countries, influencing their learning and practice, and contributing to Strategic Priority 21 of the FAO’s Global Plan of Action on Forest Genetic Resources to strengthen training and education [B].

**Conservation in disturbed agroecosystems, Chile: In 2015, based on their research showing importance of tree species in fragmented landscapes, particularly the value of trees in agricultural landscapes and thus potential of ‘circa situm’ conservation approaches, Boshier trained 144 participants - principally Chilean government forestry officials [F]. The Chilean Foundation for Agricultural Innovation provided USD13,140 to support dissemination and adoption of the novel land management approach for conservation in disturbed agroecosystems, including circa situm approaches [F]. The training programme and circa situm methods received media coverage [F], and Chilean organisations (Chilean Forestry Institute (INFOR), Institute of Agricultural Development, BioBio Regional Government) implemented the approach. For example, in 2016, INFOR’s Conservation and Genetic Improvement Group reported an apiculture programme, financed by the BioBio Regional Government, to increase the number of available flowers through the restoration and supplementation of floral diversity with emphasis on native forest species. The report showed dual benefits for local businesses and tree conservation, stating:

“INFOR promotes use of the circa situm model (conserve through use), popularised by David Boshier of the University of Oxford. This model has been applied in the creation of honey orchards with native species. These orchards, practical conservation banks for forest genetic resources, are established and used by bee-keeper members of BIOMIELAG, to improve availability of nectar and pollen in support of honey production by small and medium scale regional beekeeping enterprises. Thus, the honey orchards are valued and transformed into true ex situ conservation units” [G] (translated from Spanish)

Collaborating with Biodiversity International, Boshier developed free online teaching materials on forest genetic resources, including the importance of trees outside forests (based on [1-4]) published in 2014 [H(i)]. Between Oct 2014 and Oct 2016, the training guides were accessed more than 2,700 times (unique events) with over 2,000 PDF downloads [H(ii)].

The UK Centre for Ecology and Hydrology has used these resources for training in East Africa, including in Tanzania (2018) and Kenya (2019) where there were 50 participants from a wide range of sectors, including forest research, forest policy, forest managers, extension workers, national park staff, and NGO staff [I(i)]. The leader of the training courses stated that these materials have been a “ major help in communicating messages regarding the identification, protection and use of tree genetic resources…trainees are also very appreciative of the fact that you have made all of this [content] available free online” and that the training was “very influential” [I(i)]. From the course in Kenya, 14 attendees (78%) “strongly agreed” and 4 attendees (22%) “agreed” in answer to the statement “ after the course I plan to apply content from the course in my role” [I(ii)].

Since 2013, Biodiversity International have run 5 forest genetics training courses using the materials influenced by the University of Oxford research. These have taken place in Costa Rica, China, France and Malaysia, with a total of 136 participants from at least 20 countries, of whom at least 67% were from non-academic organisations (including forest and conservation managers, government officials, and non-governmental organisations) [J]. Boshier has delivered additional training courses in several countries including Cambodia (24 participants, mostly from central and local forestry administration, March 2020) [K] and Brazil (33 participants, 9 from non-academic organisations, August 2019) [L]. According to the Deputy Director of the Institute of Forest and Wildlife Research and Development in Cambodia, a participant said “ it is very useful to have learned the different strategy for conservations, particularly the circa situm which is quite new for me” [K]. The material is also being adopted by other environmental training organisations: the coordinator of the Environmental Leadership and Training Initiative for the Philippines is quoted by Biodiversity International as saying,

“*I realized how important forest genetic resources are for restoration. We run training programmes for governmental and civil society organizations and research institutes on using native tree species in restoration, but until now we haven’t covered genetic aspects in the training sessions. I can’t wait to go back and incorporate in our training programme what I learned about genetic conservation strategies during this course.*” [M].

NP-C is a rare disease (1 in 100,000 live births) with a highly variable but uniformly fatal clinical spectrum, ranging from a perinatal, rapidly progressive systemic disorder to an adult-onset chronic and life-limiting neurodegenerative form. Platt’s research has had an impact on patients with NP-C and other LSDs in three main areas: (a) Miglustat is the only approved targeted therapy for the disease and has had a transformational impact on the treatment of NP-C patients worldwide; (b) Benefits to the companies Orphazyme, IntraBio and Idorsia in reaching clinical trials of efficacy of new treatments, and in one case FDA Breakthrough Therapy status ahead of full approval; (c) development and validation of a simplified clinical severity scale that has been recommended for multiple clinical purposes including in clinical trials.

“The impact of Professor Platt’s work in the area of Lysosomal diseases and in particular NPC, has been truly immense. It has not only led to the only approved therapy for NPC, but has greatly expanded our understanding of the disease, and contributed to the development of other therapeutic targets and new ways to monitor and assess disease progression.”*

Niemann-Pick UK [A]

Multiple studies have shown that miglustat substantially slows or stabilises progression of neurological manifestations of the disease and leads to significant improvements in both life expectancy and quality of life. This is affirmed by two long-term follow-up studies of the largest cohort of miglustat-treated NP-C patients from 2009 to 2017 [B]. One of these studies described a range of improved treatment outcomes for 472 patients from 22 countries and reported that 70% of patients on continuous treatment had improved or had stable disease [Bi]. The other study found a significant reduction in the risk of mortality for 789 patients from 21 countries; compared with patients not treated, median survival of treated patients was longer by approximately 10 years from onset of the patients’ neurological manifestations [Bii].

Miglustat is authorised for use in most countries except the US and has been recommended as the standard of care for NP-C patients since 2018 [C]. There are approximately 100 NP-C patients currently in the UK, between 70-90% of whom are treated with miglustat. This proportion has increased from an estimated 50% in 2013, and has contributed to a greater proportion of patients diagnosed as children living into their teens/adulthood [A]. The number of patients diagnosed with NP-C, in particular adult-onset patients, has also increased in recent years [A]. A detailed analysis in France showed that during the period 2009-2016, half of new NP-C diagnoses were in adults compared with only one fifth during the period 2000-2008, and suggested this was due to ‘ improvement in awareness of NP-C among neuropsychiatrists after miglustat therapy became available’ [D].

Although miglustat is not approved by the FDA for treatment of NP-C in the US, it can be prescribed to patients ‘off-label’ because it is approved for the treatment of Gaucher disease. The Mayo Clinic estimates that there are approximately 500-600 NP-C patients in the US, of whom at least half are treated with miglustat [E].

Sales figures from Actelion, available up to 2016 before the company was taken over by Johnson and Johnson, reflect the increasing use of miglustat (brand name Zavesca) for NP-C [F]. Total sales for miglustat (outside the US) were strong in 2014 (CHF103,000,000), with an increase since 2013 attributed to its indication for NP-C and as a result of “ expanding diagnostic tools and heightened awareness”. Demand against NP-C indication increased in 2015 and gave ‘double-digit growth’ for 2016. The report for 2016 states that “ Globally, patients receiving Zavesca grew by 6% compared to 2015, driven by a 13% increase in the treatment of patients with NP-C”.

Following expiry of the patent for use of Zavesca in NP-C in 2019, generics became available that year in many countries. In the UK, all three regional contracts for miglustat use in NP-C patients are currently held by generics companies [G] and in France, most patients are using generics. Generics were introduced at a reduced price compared with Zavesca and the costs of both have fallen over the subsequent 18-month period. The saving varies across European countries from 20% to 50%. In Italy, where contract prices are available, the generic product was first marketed at EUR32 per capsule [text removed for publication] whereas the cost is now EUR15-20 [G]. Given that the daily dose is 6 capsules, the annual saving from using the generic is of order EUR40,000 per patient, per year. The generic is the preferred drug for insurance companies in the US because of its lower cost.

A study published in 2020 showed that miglustat may reduce aspiration risk and improve quality of life, providing further evidence of the benefits of miglustat therapy [H]. This new study, along with supporting evidence from other recent studies, is being used by patient groups and [text removed for publication] [E].

As a result of work by Platt and collaborators [2], Orphazyme chose NP-C for its first clinical trial of arimoclomol [I]. The Phase II/III trial in NP-C patients (50 in total, 34 receiving arimoclomol) reported improvement in neurological clinical symptoms and quality of life for patients and slower rates of disease progression (annual progression reduced by more than 60% compared with placebo) following 12 months of treatment [Ji, Jii, Ki]. The best outcomes were seen in patients on miglustat and arimoclomol. Based on these findings and in order to bring the product to patients as soon as possible, the US FDA granted Breakthrough Therapy and Orphan Drug designations to arimoclomol for NP-C in November 2019 [Kii]. A small but increasing number of patients in the US and Europe are being treated via this expanded access programme [I, Ki]. Orphazyme has submitted an application for marketing approval to the FDA in the US (July 2020) [Kiii] and submitted a Marketing Authorisation Application to the European Medicines Agency in November 2020 [Kiv].

There has also been substantial economic impact from the development of arimoclomol with Orphazyme’s CSO stating that “ Professor Platt has been instrumental through the development of Orphazyme from being a one-person company [in 2010] to one of Europe’s most prolific young biotechs....with >150 employees with offices in Denmark, US, France and Switzerland” [I].

In October 2020, IntraBio announced results from its trial of ALL in NP-C (NCT03759639, N=32). The compound demonstrated a statistically significant and clinically meaningful improvement in symptoms, functioning, and quality of life for paediatric and adult patients with NP-C [L].

The company Idorsia has invested in the testing of lucerastat, the second generation miglustat analogue developed and tested by Platt and colleagues [4,5], through a randomised, double-blind Phase III trial for Fabry patients across multiple international centres (MODIFY, NCT03425539) [M]. This progressive and potentially life-threatening LSD is more common than NP-C, affecting between 1 in 40,000 and 1 in 60,000 males, and has no fully effective first-line treatment.

The work carried out by Platt and colleagues on developing tools for monitoring NP-C disease progression [6] has had improved the ability to monitor treatment responses in clinical trials. Orphazyme already commenced its trial of arimoclomol before validation of the ASIS metric, but applied the metric to their initial cohort and as a result were able to recruit a larger cohort in order to be more confident of having meaningful outcomes from the study [A]. The company also used the 5-domain clinical severity scale validated by Platt and Cortina-Borja [6] as the primary outcome measure in its trial of arimoclomol [I,J] and used the ASIS metric to examine the rate of disease progression at the start and end of the trial.

Lions are one of the world’s most iconic species, but their population declined by 43% between 1993 and 2014. They remain in less than 10% of their historic range and are classed as vulnerable by the International Union for Conservation of Nature (IUCN). Only 7 countries still have more than 1,000 lions, including Tanzania and Zimbabwe. Where lion populations occur outside protected areas, they live alongside some of the world’s most impoverished people (proportion of people in multidimensional poverty is 31% and 55% in Zimbabwe and Tanzania, respectively), and co-existence with wildlife can impose substantial additional costs on these vulnerable communities.

The University of Oxford research was instrumental in developing IUCN recommendations on trophy hunting and human-wildlife conflict. The 2018 IUCN Cat Specialist Group publication ‘ Guidelines for the Conservation of Lions in Africa’ extensively references University of Oxford research, including [1], [2], [3] and [4], and includes a chapter from WildCRU researchers providing guidance to help ensure that where trophy hunting is practiced, it “minimises the risk of detriment to the population and maximises the chance of effective conservation” [Ai]. University of Oxford research contributed to the 2019 IUCN Trophy Hunting Briefing paper (citing [1] and data on lion populations from WildCRU), which provides guidance for responsible international decision-making [Aii]. The researchers also provided recommendations for conflict mitigation developed by the IUCN Human-Wildlife Conflict Task Force; this living toolkit for best practice (developed from 2015) extensively references approaches used by the Ruaha Carnivore Project (RCP) [Aiii].

WildCRU research demonstrating the negative effects of unsustainable trophy hunting [1, 2] has led to greatly improved sustainability of trophy hunting in Zimbabwe. Specifically, from October 2013, WildCRU worked with the Zimbabwe parks authority to establish an age-based quota setting system for lion trophy hunting, implemented from 2014 [Bi]. This quota system has had a major impact on lion population sustainability: for example, in 2013 approximately 60% of hunted lions were aged less than 4 years, whereas in 2015 less than 5% of hunted lions were in this age group [Bii]. The population is now stable, with benefits not only for lion conservation but also industries such as trophy hunting and photo-tourism (confirmed by the Zimbabwe Professional Guides Association (ZPGA) [Biii], which rely on lion presence and contribute to the economy, and local communities benefit from revenues derived from sustainable hunting [Biv].

The WildCRU projects, RCP and Hwange Lion Research Project, have provided practical conservation outcomes to address human-lion conflict. Their research showed that livestock depredation was a key driver of conflict in both sites [4, 5, 6], so reducing this has been a priority. Food insecurity is a major issue in both locations, affecting 38% and 15% of rural Zimbabweans and Tanzanians, respectively. Protecting livestock improves both economic and food security. Since Aug 2013, RCP has predator-proofed 207 livestock enclosures, protecting at least 17,300 livestock animals worth over USD1,000,000 from carnivore attack [Ci, ii]. Households with wire enclosures lose 56% less livestock to carnivore attack [Ci]. Additionally, as of 2019, 15 conflict officers were employed and 9 livestock guarding dogs were placed to protect over 500 households from lion attacks [Cii, iii]. RCP has trained and employed warriors from traditionally marginalised groups as ‘Lion Defenders’ (17 employed in 2019 [Ciii]) to chase away lions from households and find lost livestock, which are particularly vulnerable to attack (per year, approximately 4,000 livestock recovered, worth approximately USD750,000 annually) [Ciii]. Depredation has been reduced by over 60% in target households, improving economic security and protecting culturally valuable assets [Ci, ii, iv]. During 2019, the Lion Defenders prevented or stopped 26 lion hunts, bringing the total stopped since 2013 to 111 [Ciii]. Retaliatory killing of carnivores has been reduced by over 80% compared to the baseline in 2011: for example, only 4 lions were killed in 2018 compared to approximately 60 per year prior to the Lion Defenders programme [Cv]. Strategies and a case study from the University of Oxford research were included in the Tanzanian National Human-Wildlife Conflict Strategy, endorsed by the government in 2020, which recommends approaches used by RCP including livestock guarding and education [D].

In Hwange, since 2013 the WildCRU project has employed local men and women as ‘lion guardians’ (the Long Shields programme; 9 employed in 2013; 12 in 2020 [Biv]) to establish and improve livestock husbandry practices, including predator-proofing livestock enclosures, implementing successful non-lethal methods of deterring predators from community land, and alerting villagers of nearby lions [Ei, ii]. The programme covers an area of approximately 2,500km² and benefits over 6,800 rural households engaged in subsistence agro-pastoralism. This has led to substantial (up to 50%) reductions in losses of livestock to lions and other predators since the initiation of the programme in October 2013 [Ei, ii]. WildCRU found there to have been a 41% reduction in retaliatory killing of lions by local people since the research-based conflict mitigation programme around Hwange (2013-2017) [Fi]. Attitudes to the presence of lions in the area has also measurably improved: evaluation of farmers’ attitudes before and after introduction of the Long Shields programme showed that 91% felt the programme had helped them avoid impact from lions, and the proportion of farmers wanting lion populations to decrease fell from 93% to 65% [Fii]. Village leaders have welcomed the mitigation measures and report improved attitudes towards lions and protection for livestock [Eiii,iv,v]. Between 2014 and 2019, the project team trialled 10 mobile canvas livestock enclosures (‘bomas’) to protect livestock and simultaneously fertilise crop fields, benefiting 89 households and protecting 854 cattle from carnivore attack and improving crop yields on fertilised fields by approximately 30%, substantially improving local food security [Bi], with the local council stating that bomas bring “ soil structure regeneration…a huge positive impact being realized through increased yields of cereal crops” [Ei].

WildCRU’s Hwange programme served as a blueprint for community-based conservation programmes elsewhere, including in Namibia, Botswana, and other sites in Zimbabwe [Bi]. For example, WildCRU provided training for community programmes to reduce human-predator conflict in Chizarira and Gonarezhou National Parks (Zimbabwe) that have successfully protected livestock [Gi, ii]; informed other practitioners, e.g. through a Human-Wildlife Conflict training workshop in Gabon in 2017 organised by the World Bank [Giii]; and shaped national human-wildlife conflict strategies (e.g. Human Wildlife Conflict Strategy for Zimbabwe, 2018 [Bi]).

In both Hwange and Ruaha ecosystems, wildlife often imposes a substantial burden on vulnerable rural communities, so wildlife-related benefits need to outweigh those costs [e.g. 4, 5, 6]. The research teams employ over 100 local people (Hwange: 30 Zimbabweans employed, supporting approximately 150 family members; Ruaha: more than 80 Tanzanians employed, supporting at least 800 family members). A local village leader in Hwange confirmed that the employment is valuable to the community [Eiii]. The RCP ensures 800 primary school children are fed each day, reducing food insecurity and increasing school attendance [Ciii]. 42 secondary-school and 4 higher-education scholarships have been funded and a programme that twins 16 local schools with international schools provides the local schools with essential materials and equipment for teaching [Ciii]. In a village in Ruaha, the school pass rate increased dramatically since 2014, which the village leader attributes at least in part to “ the impact of school feeding, as well as…support materials and infrastructure” received through RCP [Civ]. A ‘community camera-trap’ programme, in which villagers monitor the wildlife and gain rewards for wildlife presence, covers 16 villages (approximately 30,000 people) and delivered USD70,000 of education, healthcare and veterinary care in 2020 [Ci]. A village leader stated “ before there was no benefit, only bad things, but now people easily see the benefits of having carnivores and we want to continue to protect them” [Civ]. Trials of this approach in northern Tanzania, Zambia and Botswana were ongoing in 2020.

The research projects have improved local awareness of conservation. In Ruaha, the researchers produced a local-language book to foster changes in attitudes towards lions, distributed free to all local primary schools [Ciii]. In Ruaha, over 40,000 local people were engaged through educational film nights and visits to the Park since 2013; in 2019, 99% of respondents from the local people who had visited the Park reported more positive feelings towards wildlife after the visit [Ciii]. The conservation and local development impacts of this work have been recognised by international awards, including Cincinnati Zoo Wildlife Conservation Award 2018 [Hi], and 2016 Saint Louis Zoo Conservation Award [Hii] to Amy Dickman. The CEO of Saint Louis Zoo is quoted in the award press release as stating: “ Amy and her team have converted lion killers into lion conservationists and in doing so, saved countless animals" [Hii].

The illegal killing on 1 July 2015 by a trophy hunter of one of long-term study-animals from the Hwange project, ‘Cecil’, sparked an international media furore, including at least 95,000 print media articles and 700,000 social media mentions in 3 months [Ii]. This reaction was based in part on University of Oxford GPS collar-tracking [1], providing data showing Cecil was wounded 10-12 hours before his death [e.g. Iii]. This changed international public perceptions of trophy hunting: for example, more than 40 airlines introduced or reaffirmed bans on transporting trophies from the ‘Big 5’ African animals [Iiii]; and a survey of 1,000 adults in the USA by National Geographic one month after the killing showed 10% had signed a petition or pledge in response [Iiv]. A WildCRU proposal in 2017 on funding conservation was reported by The Economist [Ji] and in 2018 the UN Development Programme established The Lion’s Share fund, to which companies are asked to donate 0.5% of media budgets for advertising that uses animals, raising USD2,917,873 by 31 Dec 2019 [Jii]. At the fund’s anniversary in 2019, Dickman addressed the UN General Assembly.

WildCRU research [e.g. 1, 2] has contributed to the evolution of UK government policy on trophy hunting. In a debate in the UK House of Commons in 2015 [Ki], the Minister for the Environment stated that the government department had worked with WildCRU and that “… Andrew Loveridge and David Macdonald from Oxford University have contributed a great deal on the subject. For DEFRA, trophy hunting is a serious issue…the Government will ban the importation of trophies into Britain unless we see very significant improvements in what is happening in Africa”. In 2016, WildCRU provided a commissioned report to the Minister for the Environment on trophy hunting, particularly with respect to lions [Kii], and in Oct 2019 the UK government’s legislative programme (Queen’s Speech) included proposed legislation to ban imports from trophy hunting. This started with a formal consultation [Kiii], to which WildCRU researchers contributed evidence from their research and also presented to a UK parliamentary group in June 2020. Similarly, in the US, WildCRU research informed legislative debate in 2019 on the Conserving Ecosystems by Ceasing the Importation of Large Animal Trophies Act (‘CECIL Act’), including their work (including [2]) being cited in testimony to a Sub-committee of the House of Representatives [Kiv].

The research has led to a major change in immunisation policy in the UK, and other countries, and is an exemplar of the public health benefits of publicly-available libraries of genomic information for control of bacterial and other microbial pathogens.

Invasive meningococcal disease (IMD) caused by Neisseria meningitidis represents a serious public health concern. More than 1,000,000 cases are estimated to occur worldwide every year. The bacterium causes a range of serious, life-threatening diseases including septicaemia and meningitis, particularly in children and teenagers. In the UK mortality rates are around 5%, with about 20% of survivors suffering from life-altering sequelae such as limb amputations and brain damage. Some aggressive epidemic strains have much higher mortality rates.

Historically, the incidence of serogroup W strains causing IMD had been low in the UK (1%–2% of total cases annually). However, from 2009 onwards there was a significant rise in cases of IMD caused by serogroup W strains in England from 22 in 2009-10 (2 deaths) to 55 in 2012-13 (9 deaths). Even though case numbers were low, infections were associated with severe outcome, atypical clinical presentations (e.g. septic arthritis and severe respiratory tract infections), and a case fatality rate of 12-13% [A].

By late 2013 Public Health England (PHE, at that time the Health Protection Agency) was concerned about the increase in serogroup W cases. The Oxford research played a key part in guiding the appropriate actions of PHE’s Vaccine Preventable Invasive Bacterial Diseases Forum, of which Maiden was a member [B]. An international collaboration led by the Maiden Laboratory and PHE was established to characterise the strain responsible for the increase in cases and determine whether it posed a serious public health risk. Isolates sequenced by PHE and by the Maiden group were analysed using the PubMLST Neisseria database (incorporating the BIGSdb tools [1]), and also using the MRF-MGL. This allowed the rapid identification and characterisation of the serogroup W strain (W:cc11) and its relationship to other circulating meningococcal strains.

Crucially, information on previously characterised strains contained in the databases established that the serogroup W strain was derived from hyperinvasive strains that had caused recent major outbreaks in South America with high mortality rates, and was related to the serogroup C strain which had caused global outbreaks of IMD in the 1990s, with thousands of cases and hundreds of deaths in the UK every year. The results [4] were shared with PHE in real time.

The strain identification [4] was one of the first exemplars of the application of genomic data to drive a population-level public health intervention, and led to a change in the UK immunisation policy. At the October 2014 meeting of the Joint Committee on Vaccination and Immunisation (JCVI), the independent body which advises the Department of Health and Social Care on immunisation, PHE stated that ‘ …Of particular concern was the fact that most cases of MenW disease in England and Wales were caused by W:cc11, the same clonal complex as the outbreak MenC strain which struck the UK during the 1990s, which was associated with increased virulence, increased fatality rates and disease in younger age groups.’ [C]. Based on the findings of [4], JCVI were seriously concerned that the W:cc11 strain could cause a nationwide outbreak, leading to their recommendation that a MenACWY vaccine should be introduced into the routine UK immunisation schedule [D].

In summer 2015, the monovalent MenC vaccine given to teenagers in school year 9 was replaced with a single dose of a quadrivalent MenACWY vaccine. The MenACWY vaccine continued to be given routinely to this age group through a schools-based programme. At the same time, an emergency MenACWY catch-up programme began through GP surgeries, targeted at all teenagers aged 14-18 (since they have high carriage rates of meningococci and are a source of infection for susceptible children and adults) and those up to the age of 25 attending university for the first time (since there are high rates of meningococcal disease in this population) [E].

This was the first use of the quadrivalent MenACWY vaccine in Europe as part of a national routine immunisation schedule and arose directly from knowledge of the genomic epidemiology of the serogroup W strain provided by the Maiden laboratory, and the use of the publicly available resources his laboratory developed. Prior to this, the MenACWY vaccine had been only used in the UK for high-risk individuals, travellers to endemic regions and for controlling sporadic outbreaks.

At the point the MenACWY vaccine was introduced, serogroup W cases were rising at a rate of around 80% a year across all age groups (55 cases in 2012-13, 95 in 2013-14, 176 in 2014-15) [A,F]. In the first year of vaccination (2015 - 16) there were 211 serogroup W cases in England instead of a projected 317, representing an estimated 106 cases prevented. Over the next 3 years the incidence of serogroup W disease in the UK fell by more than 60%, with 225 cases in 2017-18, 113 cases in 2018-19 and 79 cases in 2019-20 [F].

PHE data indicate that if case numbers had continued to rise by 80% a year, in the 5 years to 2019 – 20, an additional 6,000 serogroup W cases would be expected in the UK, with 720 deaths (based on a 12% mortality rate) and a further 1,200 people (20%) of survivors left with life-changing disabilities which seriously affect quality of life and also have a significant economic impact for the sufferers, their families, and the NHS [F]. Whilst these figures are based on an assumption of a steady increase, the epidemic of the related serogroup C strain in the 1990s had shown that the serogroup W strain could cause thousands of cases and hundreds of deaths if left unchecked (there were nearly 1,000 cases and 78 deaths before the MenC vaccine was introduced in 1999), and that JCVI experts were seriously concerned about the potential for this aggressive MenW strain to do the same [C].

The rapid decline in MenW cases in all age groups, not just those immunised, corroborates the remarkable effect of herd immunity conferred by conjugate meningococcal vaccines that had been first recognised by Maiden’s laboratory through carriage studies he led during the implementation of the MenC vaccine in the UK during 1999-2000 [5].

Between 83% and 88% of each year 9 group received the MenACWY vaccine in UK schools in the 4 academic years from 2015-16 to 2018-19 (approximately 500,000 teenagers a year; each yearly birth cohort is approximately 600,000 according to ONS). In addition, in the same period, several hundred thousand of those aged up to 25 were vaccinated as part of the catch-up programme through GP surgeries. Overall, approximately 2,500,000 doses of MenACWY vaccine were given in the UK over the first 4 years of the programme [G].

The subsequent rise of serogroup W cases in other countries including France, Italy, Netherlands, Canada, Ireland and Australia led to them introduce the MenACWY vaccine into their national schedules [H]. In the case of the Netherlands, Canada and Ireland, the national research that prompted immunisation schedule changes used the PubMLST Neisseria database and the MRF-MGL to identify and characterise the strain; the Netherlands and Ireland directly referenced the research [4]. Cases in these three countries were confirmed to be related to the strain first identified in the UK. The Netherlands added MenACWY vaccination to their national schedule in 2018. From 2014-15 to 2015-16, serogroup W cases had increased more than fourfold in The Netherlands, particularly in adults over 65 years old [I]. The Public Health Agency of Canada used the Neisseria databases to characterise the strain affecting Canada and argue for the introduction of the MenACWY vaccine [J]. By 2019, individual provinces and territories in Canada had almost all had implemented MenACWY immunisation [H]. From 2013, Ireland experienced a similar rise in serogroup W cases; work by Irish public health authorities in collaboration with the Maiden group identified that this was the same serogroup W strain that was affecting the UK. Data from the study informed Ireland’s National Immunisation Advisory Committee, which made the decision to introduce MenACWY vaccination in Ireland in 2019 [K]. The latest recommendations in Australia also include MenACWY [L].

Conservation of the oceans is a global priority, not just in terms of the preservation of marine biodiversity, but because of the central role that oceans play in the regulation of almost all the systems that make life on Earth possible. This is recognised in UN Sustainable Development Goal (SDG) 14 “Life Below Water”, which states that “ Careful management of this essential global resource is a key feature of a sustainable future… Marine biodiversity is critical to the health of people and our planet” and notes that *“over three billion people depend on marine and coastal biodiversity for their livelihoods.*”

Pelagic seabirds (those that spend most of their time on or in the open ocean) play a key role in marine ecosystems. As higher predators, seabirds such as those studied at Oxford are sentinel species that provide an early warning of large-scale environmental disturbance. Monitoring and tracking seabirds has been shown to provide an accurate alert of threats such as overfishing and pollution. Pelagic seabirds are amongst the most threatened and fastest declining taxa on Earth and their conservation is a priority, covered by the Convention on Migratory Species.

To help achieve SDG14, the UN set a target of conserving at least 10% of coastal and marine areas worldwide by 2020. The purpose of SPAs and MPAs is to stop or limit fishing and other harmful activities in areas that are important for species and habitats, thereby helping to increase the resilience of marine environments to stressors outside these areas. SPAs are special sites designated under the EU Birds Directive (2009/147/EC) to protect rare, vulnerable and migratory birds. The Oxford research provided key scientific evidence for the creation and enhancements of several important protected areas in the UK, UK Overseas Territories and in international waters, thus helping to meet the targets of UN SDG14.

In 2016, the UK Government’s Joint Nature Conservation Committee (JNCC) designated a 180km² area of the Irish Sea as a new marine SPA (the Irish Sea Front SPA). The research on Manx shearwater foraging behaviour [1, 2] demonstrated the importance of this intensively-used sea area as a foraging hotspot for nearly 400,000 breeding shearwaters from all the major UK colonies situated up to 320km away. This was one of the central pieces of evidence used to support the creation of the SPA [A], which has also enabled the UK Government to meet obligations set in the EU Birds Directive. The Directive provides a framework for the conservation of wild birds in Europe. Further tracking research has contributed to consultation stages of other UK MPA processes, including extensions to the Welsh SPAs with marine components.

At-sea distribution data from the tracking of migratory and foraging seabirds formed a core component of a proposal led by BirdLife International for an OSPAR High Seas MPA (North Atlantic Current and Evlanov Seamount). OSPAR is the mechanism by which 15 governments and the EU cooperate to protect the marine environment of the North-East Atlantic. The Oxford research helped BirdLife International to argue the case for this MPA, since it showed that the proposed area was intensively used as a stopover by shearwaters during their long distance migration between the breeding areas in the North Atlantic and non-breeding areas in the South Atlantic and Oxford research is referenced throughout the nomination [B] . The High Seas MPA is one of the largest ever marine protected area proposals at around 650,000km², and has now been provisionally accepted under the Oslo-Paris agreement.

University of Oxford research data [4, 6] have provided scientific evidence for MPA designation [C] around South Georgia and the South Sandwich Islands. A review of known trends and threats coupled with the population structure of penguins in the South Sandwich Islands was presented as evidence to the Government of South Georgia (GSGSSI), and this influenced the design of an increased fisheries’ no-take zone around the South Sandwich Islands. The Oxford research highlighted where (currently unfished) quota of krill would impact penguins on the Antarctic Peninsula if this quota were released. The MPA followed these guidelines of the foraging areas of penguins and ensured the protection of key penguin populations, while also reducing the risk that fishing effort would be displaced to the Antarctic Peninsula.

The Director of Fisheries and Environment, Government of South Georgia, confirms: “On the basis of the evidence provided…, GSGSSI considered [Dr Hart's] recommendation to extend the ‘No Take Zone’ (NTZ) around the South Sandwich islands. This advice included extending the NTZ from 3km out to 50km in order to reduce the potential for competition for food resources between the region’s krill fisheries and penguin populations. The rationale behind the extension was to encompass the foraging range of the penguin species in the region as determined by Dr Hart and colleagues’ recent research… This entered into GSGSSI legislation in May 2019.” [D].

Oxford research has also contributed to important changes in conservation status and protection for the Balearic shearwater, Europe’s rarest seabird, which has a world population of 25,000 to 30,000. It breeds in the Mediterranean and migrates north into the Atlantic in the summer, with as many as 25% of birds spending time in UK waters. The species is predicted to be heading for extinction; entanglement in fishing nets is a major cause of death. In 2016, Guilford and others helped drive the designation of the Balearic shearwater becoming an ACAP (Agreement on the Conservation of Albatrosses and Petrels) Priority Population (one of eight worldwide), supporting global efforts to promote the recognition and conservation of this species [E]. SEO/Birdlife confirm the role Oxford research played in “providing assessment to ACAP […] to propose the species as a priority population for the Agreement, which was accepted by the Advisory Committee (Guilford & Arcos 2016). This has facilitated funding research and conservation action on the species afterwards, including ACAP grants as well as other sources of funding.” [F].

Since 2004 the Balearic shearwater has been designated a Critically Endangered species in the IUCN’s Red List, as it is in rapid decline. In the 2018 revision of the Red List, the Oxford research formed part of the evidence used to keep the species at Critically Endangered. However, following this there was a move to reduce its status to Near Threatened, much lower on IUCN’s scale of concern. In April 2020 the Oxford research [3] played a significant part in preventing this reduction in status [G]. The letter from SEO/Birdlife further describes the researcher’s “ *contribution to the Red List Forum of BirdLife International/IUCN to review the conservation status of the Balearic Shearwater, providing information that helped to keep the species as Critically Endangered.*” [F]

In addition, the Oxford group was involved in providing evidence that night setting of fishing activity (mainly longlining) would be effective for mitigating by-catch of shearwaters. This measure is now incorporated into Birdlife International recommendations [H]. In the UK, the Oxford group has provided expert scientific advice to the Balearic Shearwater Working Group which works to understand the distribution, abundance and conservation requirements of the UK’s only globally Critically Endangered bird species. Natural England's Principal Advisor states that the Oxford research has given them a basis on which to plan for better UK conservation of this species, describing it as ‘ a crucial step for conservation of this species in the UK’ and ‘ a major advance in the evidence required to safeguard the future of a Critically Endangered species’ [I].

In September 2014 Penguin Watch was established as a citizen science project on the Oxford-led Zooniverse platform [4, 5]. Volunteers were asked to count penguins in images and time-lapse footage captured by satellites and drones. More than 1,000,000 people have taken part over two iterations of Penguin Watch without registering, and there are currently 21,894 registered volunteers who visit repeatedly. Version 1 from April 2014 to June 2017 had 54,000 registered volunteers. In a survey of volunteers, conducted in December 2020, 97% of 513 respondents said they would suggest Penguin Watch to others [J].

Penguin Watch has regular outreach in schools and was the featured citizen project of British Science Week 2017 (BSW), with 21,650 downloads of the primary school pack and 8,770 of the secondary school pack. Over 150 schools followed up directly with a request for more data and videos. Many people asked additional questions on the project’s “talk” forum. The participation counter for the BSW challenge recorded 521,678 images classified, exceeding the target of 250,000 images. During the Covid-19 pandemic there has been a huge uptick in interest from schools, with 385,010 images analysed in April 2020, ten times as many as in April 2018 (31,246 images) and April 2019 (30,324) [K].

The education part of the Penguin Watch site has presentations, videos and data available for download. This has been very popular during 2020, as teachers have used it for online teaching, and Duke of Edinburgh Award candidates have used Penguin Watch as their volunteering component. Of 16 volunteers who identified as teachers, 87% said they had modified their curriculum, for example on climate change, because of Penguin Watch. 93% of teachers described the level of engagement of their students as Excellent or Good when volunteering for Penguin Watch [J]. One testimonial from a US high school biology teacher described: “The first time I incorporated PenguinWatch into my classes I quickly realized no one teaches about Antarctica.... I quickly began learning everything I could, I was not taught about Antarctica in school either. We learned geography, the location of bases, where Dr. Hart’s research was focused, and about several penguin species. The students were fascinated and so was I.” [L]

Some volunteers have become extremely engaged, using Penguin Watch data as training workshops or degree level projects. Three of the project’s most enthusiastic volunteers have used Penguin Watch to develop skills in biology research and academic writing skills, to the point of contributing data and content to a study [5].

Penguin Watch has seeded a number of camera-based remote studies on seabirds and other taxa. At the Seabird Conference in Liverpool in September 2018, training was given to new non-academic partners from the British Indian Ocean Territory, Ascension, Turks and Caicos and Anguilla. The project has also continued to help St Helena Environment Department, and Falklands Conservation, Chile, New Zealand and Australia. Around the UK cameras previously used in Antarctica were donated to Jersey, Orkney and Isles of Scilly, and Penguin Watch has helped Birdlife and the RSPB to establish monitoring sites on the UK mainland at Bempton Cliffs, the Isle of May, Skomer Island, Ramsey Island and Natural England on Brownsea Island. Natural England's Principal Advisor confirmed, *“This information is crucial to understand management requirements of recently designated Marine Protected Areas so that these rare and highly protected birds can breed successfully. The cameras worked well and given movement restrictions during the Covid 19 pandemic, they have helped a lot to fill gaps in our monitoring. Natural England is now an enthusiastic proponent of camera monitoring as part of the UK & Ireland Seabird Monitoring Programme.*” [I].

Penguin Watch has served as a case-study and model for many other wildlife monitoring projects that have been initiated, many also on the Zooniverse platform. For example, it has helped international government scientists from the US National Oceanic and Atmospheric Administration to build ‘Seal Watch’ and Parks Canada to build the ‘Arctic Bears’ project.

Oxford Biodynamics was spun out of the University of Oxford in 2007, based on the intellectual property around the CCS [5], to take this technology to market. Much of the early years of the company involved optimising and making reproducible the steps of the CCS measurement process, which was known to be unreliable and thus not yet suitable as a diagnostic tool. Further optimisation resulted in moving from analysis of cancer biopsies to using whole blood. The restriction enzymes, ligase and buffers were optimised and standardised, and all the steps were converted onto a robotic platform. This resulted in a platform that could detect CCS in 4 hours and conversion from a CCS test at individual loci to a microarray platform with 1,200,000 anchoring sites, representing potential CCS throughout the genome. This revolutionised the technology, since many more interactions could be detected. The resulting CCS technology was trademarked as EpiSwitch™. Akoulitchev became the CSO of Oxford Biodynamics, Ramadass the CTO, while Mellor contributed as a member of the company’s Scientific Advisory Panel. The EpiSwitch™ platform has worldwide patents, enabling technology licensing across Southeast Asia, Australia, Europe and the United States.

In July 2015, OBD raised in excess of USD7,000,000 to accelerate the commercialisation of its epigenetic platform technology – EpiSwitch™; in December 2016 the company was floated on AIM (a sub-market of the London Stock Exchange), and raised GBP20,000,000 from investors. In Aug 2018, GL Capital Group became a 5% investor with approximately GBP9,750,000 new shares in the Company; notably this was GL's first investment in Europe. In December 2019, OBD signed a master services agreement with a major US pharmaceutical company [A], who cannot be identified for commercial reasons.

The patents, licencing agreements and contracts with leading pharmaceutical companies, gave OBD a strong strategic position in the global market and led to a market capitalisation at September 2020 of more than GBP75,000,000 [B]. Revenue in the year ending September 2019 was GBP907,000 [A]. OBD has grown to 39 employees (November 2020, headcount) at sites in Oxford UK (32) and in the USA (3) and Malaysia (4) [C].

OBD won the 2015 Frost&Sullivan European Award for Technology Innovation [D(i)], being described as “the only company with a patented and established biomarker discovery platform for industrial use”. In 2019 it received a Queen’s Award for Enterprise in Innovation, denoting ‘outstanding achievement’ and citing the “Novel epigenetic biomarker technology to deliver personalised medicine” [D(ii)].

For its validated results with the EpiSwitch^TM platform, OBD is recognized and represented in USA at the Foundation of NIH (FNIH) Biomarker Steering Committees in Oncology, Inflammation & Immunity and Neurosciences. [D(iii)].

A number of projects initiated and funded by pharmaceutical companies use the EpiSwitch technology to develop biomarkers that will predict the likely success of a treatment, thus paving the way to personalised medicine.

In a commercial project funded by Merck KGaA, OBD worked with EMD Serono (the biopharmaceutical business of Merck KGaA, Darmstadt, Germany, in the US and Canada), the Mayo Clinic and Pfizer. 99 patients were profiled using EpiSwitch for CCS detection in non-small cell lung cancer (NSCLC) using samples and data from the JAVELIN Solid Tumour trial (NCT01772004), to develop and validate baseline predictive biomarkers for response to avelumab [E(i)]. By adding Lyell Immunopharma to the collaboration, they examined biomarkers for response to immune-checkpoint treatments, with samples from patients with NSCLS and melanoma [E(ii)]. The CCS approach could identify systemic cellular network deregulations associated with differences in clinical outcome. Furthermore, stratification of patients using CCS outperformed the FDA-approved immunohistochemistry test for PD-L1 expression level, as well as the genetic test based on tumour mutational burden [E(ii)].

Genentech funded the development of the first successful blood-based assay for prognostic stratification and disease subtyping in diffuse large B-cell lymphoma (DLBCL), using EpiSwitch to detect, screen and monitor the changes in CCS, in collaboration with Roche and OBD. This work drew on highly-characterized samples and data from a clinical trial funded by Hoffmann-La Roche (NCT00486759) on chemotherapy for DLBCL patients [F]. The results showed 100% effectiveness of the EpiSwitch for CCS detection in classifying DLBCL sub-types from a single blood sample, compared with the extensive tissue biopsy gene expression profiles undertaken in the original trial.

Amyotrophic lateral sclerosis (ALS) is a progressive and ultimately fatal neurodegenerative disease, where motor neurons are lost from the central nervous system, with no known cure. In 2015, OBD won funding from Innovate UK, in collaboration with the University of Oxford and Chronos Therapeutics Ltd, to develop prognostic epigenetic biomarkers for ALS. This enabled ALS diagnosis to be predicted with 75% specificity, with prognosis of fast- or slow-progressing ALS determined to 80% specificity. When the results were published in 2018 [G(i)], a commentary in EBioMedicine the following month stated that the work offered “ ... an attractive tool to detect structural related epigenetic changes in ALS” and posed the CCS as the “ new kid on the block” for ALS biomarker research [G(ii)].

Subsequently, Mitsubishi Tanabe Pharma America (MTPA) sponsored a study of biomarkers, led by the Massachusetts General Hospital, REFINE-ALS (NCT04259255), in which patient biomarker data from EpiSwitch is compared with assessment of ALS disease progression. The Senior Director of Medical Affairs, MTPA said: " Through this biomarker study we are seeking to enhance our understanding of edaravone therapy in ALS. We are proud to announce Oxford BioDynamics has joined us in this effort…." [H]. The first patient was recruited in October 2019 [A] and the study is ongoing.

EpiSwitch is being deployed in the GETAFIX clinical study funded by the Chief Scientist Office of Scotland to provide epigenetic data to help identify COVID-19 patients at risk of severe disease and to profile patients who will benefit from therapeutic anti-viral treatment with favipiravir. OBD is not a co-investigator on the trial, which is carried out by NHS Scotland and the University of Glasgow [I(i)]. A Professor of Translational Immunology at the University of Glasgow explained the significance of the technology for the study:

“*To have the greatest impact on the current clinical challenges associated with COVID-19, it is essential that we have the ability to rapidly stratify individuals into those that will progress to severe disease, and those that will respond to available therapies. Evaluation of the immunological set-point via EpiSwitch™ will provide that much needed stratification tool.*” [I(ii)]

By November 2020, OBD was actively engaged in a programme to develop prognostic tests of severity of COVID-19, which included work with multiple cohorts from UK, USA, and Latin America, for a total of over 500 patients worldwide [C]. To this end, OBD had secured agreements including a partnership with Boca Biolistics Inc., FL, USA [C,J(i)]; a collaboration with Oregon Health and Science University, Portland, OR, USA (J(ii); and participation in the PREDICT COVID UK Programme (West Hertfordshire NHS Trust) [C].

In September 2018, OBD became an industrial partner in a Horizon 2020 Innovative Training Network, ‘Predictive Epigenetics’ (PEP-NET), which was granted EUR4,000,000 in total. Akoulitchev contributed to the 2020 virtual Summer School, training early-stage researchers in entrepreneurship. Akoulitchev is also the second supervisor for one of the 15 training projects funded by the Network.

The use of viral vectors to introduce new genes into cells is a technology with wide application in biomedical research and disease treatments. Oxford Biomedica is a pioneer in the field of gene-based therapy and the production of gene therapy products and was the first company to test lentiviral therapy in humans. The REF2014 Impact Case Study on Oxford Biomedica described how University of Oxford research contributed to impacts up to July 2013 on the early growth of the company and initial development of clinical applications of lentiviral gene therapy, specifically clinical trials for Parkinson’s disease and eye disorders. Here we describe the contribution of University of Oxford research (since Jan 2000) to impacts between Aug 2013 and Oct 2020, which includes major new applications and breakthrough clinical approvals, including for blood cancers, haemophilia, immunodeficiency, and a more effective strategy for Parkinson’s disease, as well as substantial commercial benefits.

Oxford Biomedica’s lentivirus technology (LentiVector®) has proven to be a safe and flexible method of delivering gene-based therapies. It can deliver the therapeutic gene of choice to the relevant cell population without provoking a destructive immune response. For long-term expression, the expression cassette integrates into the host DNA without triggering changes in the expression of host genes that might lead to cancer. The gene can also be appropriately controlled to allow permanent expression of the desired proteins and thus long-term clinical benefit. Much of Oxford Biomedica’s clinical work is through partnership with pharmaceutical companies, utilising the vector technology to address specific clinical challenges.

Oxford Biomedica’s LentiVector® platform is based on lentiviral vector technology research by the Retrovirus Molecular Biology Group at the University of Oxford. Specifically, the research from the Kingsman group in 2000 and 2001 is essential to the platform and products. According to the Chief Scientific Officer of Oxford Biomedica in 2020, the company’s “ current success and its product portfolio has been materially contributed to by research carried out by the company and Oxford University” [A], citing University of Oxford research [2, 3]. For example, the CSO stated: “ out of over 23 product development programmes…14 depend on the codon optimization technology developed through the work of the Department of Biochemistry” [1], and these “ include leading gene and cell therapy product development programmes” [A]. The CSO confirmed “ these technological developments have led to many of the deals [Oxford Biomedica] has made, including with Novartis, Bristol Myers Squibb, Orchard Therapeutics, Sanofi, Boehringer Ingelheim and Santen” [A].

Kymriah (Tisagenlecleucel) is the most clinically advanced therapy based on the Oxford Biomedica LentiVector® platform. This CAR T cell therapy was developed by Novartis in partnership with Oxford Biomedica, with Oxford Biomedica as the sole manufacturer of the lentiviral vector used in Kymriah (agreement signed in July 2017 and extended in Dec 2019) [F, Bi]. In CAR T cell therapy, a patient’s own T cells are specifically reprogrammed to recognise and attack their cancer cells. In August 2017, Kymriah was the first CAR T cell cancer therapy to be approved by the US FDA – the first gene-based therapy to be made available in the US – and was given breakthrough designation [C]. In August 2018 it was granted marketing authorisation in the European Union [D]. By the end of 2019 there were more than 200 treatment centres using Kymriah in at least 25 countries, including UK, Japan, Australia, and Canada [A, E].

Revenue and growth for Novartis and Oxford Biomedica: Oxford Biomedica received an upfront payment of USD10,000,000 in 2017 [F], and an agreed minimum of USD75,000,000 over 5 years in manufacturing revenues, as well as royalties [Bi]. For Novartis, net sales of Kymriah were USD278,000,000 in 2019 [E]. Novartis also holds a portfolio of more than 30 patents associated with Kymriah, and in 2019 Novartis increased manufacturing capacity in Switzerland and France in response to strong global demand for this treatment [E].

Transformative treatment for patients with relapsed or refractory blood cancers: Kymriah is approved to treat patients up to 25 years old with relapsed or refractory (r/r) B-cell acute lymphoblastic leukaemia (ALL), or adults with r/r diffuse large B-cell lymphoma (DLBCL). Since 2018, Novartis has also initiated 6 trials for new or expanded indications [E]. For example, in April 2020, the FDA granted Regenerative Medicine Advanced Therapy designation to Kymriah for patients with r/r follicular lymphoma [A].

B-cell ALL is the most common cancer in children. In the pivotal trial for Kymriah (study started April 2015; published Feb 2018), approximately 83% of the 63 treated children and young adults had their cancers go into remission within 3 months, transforming their chance of survival [F]. As of Dec 2019, Novartis had provided CAR T therapies to at least 1,800 patients with leukaemia or lymphoma [G]: in each case, the cells are personalised for the patient using the University of Oxford-derived lentiviral technology. The patients receiving CAR T have poor prognosis and other treatments have failed. For example, in the UK, the National Institute for Health and Care Excellence (NICE) recommends its use for DLCBL within the Cancer Drugs Fund after failure of two or more systemic therapies, as there is no standard treatment for r/r DLBCL [H].

Evaluation of Kymriah in the real-world setting (i.e. post-trial clinical practice) has shown similar safety and efficacy compared to the pivotal trials that led to regulatory approval. Specifically, for ALL, in the trial the treatment achieved 80% remission in patients for whom all other treatments had failed, and in the real-world setting the best overall response rate (complete response, CR) was 88% (95% CI 80%-94%) [Ii]. In the real-world setting for DLBCL, the overall response rate (ORR) was 59.6% (28 of 47 patients) including 38.3% (18 patients) achieving a CR [Iii].

In 2018, the lead clinician on the Kymriah trial at the Children’s Hospital of Philadelphia, USA, stated

“Before this personalized, cellular gene therapy, the patients…had about a ten percent chance of surviving…Hundreds of patients later, we're able to say those children who safely achieve durable remissions have a good chance of long-term disease control, and our hope is this is the last treatment they ever need” [Ji].

The first child treated with this CAR T therapy remains cancer free after 8 years [Jii]. Thus, Kymriah provides transformative, life-saving benefit to a high proportion of treated adults and children for whom there was previously no treatment option.

Expanding CAR T therapies: Oxford Biomedica’s vector system is recognised as a platform for CAR T therapy that is being harnessed for a wider range of diseases. In March 2020, Oxford Biomedica signed a new license and 5-year supply deal with Juno Therapeutics (Bristol Myers Squibb) to produce a CAR T therapy for undisclosed cancers, providing Oxford Biomedica with an up-front payment of USD10,000,000. This deal agreed that up to USD217,000,000 would be paid to Oxford Biomedica in milestones [Bii].

Oxford Biomedica’s lentiviral vector technology is also the basis for gene therapy product development programmes for several debilitating or life-threatening diseases. These products have yielded commercial benefits for Oxford Biomedica and their partner companies, with benefits to patients within clinical trials.

Parkinson’s disease: Stemming from the University of Oxford-derived technological developments, since 2018 Oxford Biomedica is in partnership with Axovant Gene Therapies [A], to develop lentiviral-based gene therapy treatment for Parkinson’s disease (AXO-Lenti-PD). Parkinson’s disease affects an estimated 10,000,000 people globally. In 2019, the ongoing phase II trial yielded promising 3-month data from the first cohort, triggering progression to the second cohort and a USD15,000,000 milestone payment to Oxford Biomedica [Bi]. AXO-Lenti-PD is the most advanced of Axovant’s products, thus contributing substantially to the company’s closing public offering of USD74,700,000 in Feb 2020 [K]. In January 2020, 12-month data —considered an important timeframe for assessment of therapeutic response and durability of gene therapy— from the first patient cohort (30 patients) demonstrated 37% improvement in motor function [Bii].

Haemophilia: Haemophilia represents the largest world market for rare diseases, affecting approximately 400,000 people worldwide. In 2018, Oxford Biomedica agreed a partnership with Bioverativ, a Sanofi company, to license the LentiVector® technology to treat haemophilia with gene therapy. Oxford Biomedica received an initial payment of USD5,000,000, to be followed by milestone payments of more than USD100,000,000 [F].

Severe combined immunodeficiency: In 2018, Oxford Biomedica formed an alliance with Orchard Therapeutics for the development of ex vivo gene therapy, using LentiVector®, for indications including the life-threatening severe combined immunodeficiency disease ADA-SCID [Bi], which affects between 1 in 200,000 and 1 in 1,000,000 newborns. In 2019, 2-year follow-up results from 20 ADA-SCID paediatric patients treated with the LentiVector-based product OTL-101 showed an impressive 100% overall survival (OS) and 100% event free survival (EvFS), compared to 88% OS and 56% EvFS for an historical control cohort of 26 patients who underwent haematopoietic stem cell transplant [L].

Inherited retinal disease: In their first collaboration in Japan, in June 2019 Oxford Biomedica partnered with Santen to develop gene therapy for inherited retinal diseases, with undisclosed milestone payments and royalties on sales to Oxford Biomedica [Bi]. These diseases cause vision loss or blindness, often in children and young adults. The use of the University of Oxford-derived lentiviral vector system was essential for this application, as lentiviral vectors can deliver large genes to the eye, which is challenging with other vector systems.

Oxford Biomedica’s product development based on the University of Oxford research has contributed substantially to the company’s expansion and commercial success [A]. It maintains a portfolio of over 60 patent families and employs more than 600 people across 6 sites, of whom more than 120 have a PhD (Oct 2020) [A, Bi Bii]. Oxford Biomedica has expanded to have a footprint of approximately 21,000m², including a new approximately 3,000m² research laboratory for LentiVector® platform innovation and an approximately 8,000m² facility for clinical manufacturing in Oxfordshire, both in use from 2019 [Bi]. In 2020, Oxford Biomedica agreed with AstraZeneca to produce their COVID-19 vaccine at this manufacturing facility. In 2019, the Oxford Biomedica generated revenues of GBP64,100,000 (an increase from GBP37,600,000 in 2017) [Bi]. The company is in the FTSE 250 Index with a valuation of approximately GBP700,000,000 (10-2020) [A].

Oxitec was spun out from the University of Oxford in 2002 to develop and commercialise the use of engineered sterile males through large-scale production and release. As reported in the REF2014 case study, Oxitec raised substantial investment from 2008 onwards, enabling early trials of the OX513A GM mosquito strain in the Cayman Islands in 2009, and in various locations in Brazil from 2011 onwards. These were highly successful, reducing the wild populations of Ae. aegypti by up to 98% and confirming the effectiveness of the approach [A]. During the REF2021 impact period the research has resulted in new impacts with global reach, including large-scale trials with public health benefits, further development of the technology and regulatory framework, and realisation of major commercial impact.

A four-year programme of releasing the OX513A mosquitos took place in the Brazilian city of Piracicaba starting in July 2015, initially treating an area with 5,000 residents, but subsequently expanding to include 11 additional neighbourhoods, home to a further 60,000 residents. The releases achieved 83% or greater suppression of wild Ae. aegypti, and up to 98% suppression, compared to untreated sites [Bi].

Building on the original University of Oxford research [1,2,3], Oxitec subsequently developed an improved 2nd generation strain, OX5034 which offers more sophisticated genetics using sex-specific alternative splicing. Here, the male mosquitoes carry only two introduced genes; the male-selecting self-limiting gene and a simple marker gene that allows easy tracking in the field.

A first pilot study of OX5034 took place in four suburbs of Indaiatuba, Brazil, between 2018 and 2020. The trial demonstrated an average of 89% peak suppression of wild Ae. aegypti populations across the two communities treated with a low release rate of mosquitoes and an average of 93% across two communities treated with a higher release rate compared to the untreated control site [Bi].

Oxitec also developed a new delivery method for releasing the GM mosquitos, via soaking an egg-filled mini capsule in water. The Friendly™ technology (trade name Aedes do Bem™ in Brazil) was successfully deployed in trials in the city of Indaiatuba, Brazil, in October 2019 and extended to three different mosquito vectors of disease ( Aedes aegypti, Anopheles albimanus and Anopheles stephensi) opening the potential to tackle a number of vector-transmitted diseases such as dengue fever, zika and malaria [Bii]. Following the trial, the Health Secretary of Indaiatuba said “ After the success of the first pilot with Oxitec's Aedes do Bem™, we are eager to explore the benefits that technology can offer to our city and other communities affected by the mosquito” [C].

Following the successful trial and final approval of biosafety by CTNBio, the national regulatory authority for biosafety in Brazil in May 2020 [D], Oxitec and the Dengue Control Programme of the City of Indaiatuba rolled out further releases of the Friendly technology (“Aedes do Bem!”) in November 2020 in preparation for the 2020-2021 high mosquito season [F]. The Health Secretary of Indaiatuba emphasised the significance of the project in terms of public health management: “Indaiatuba is privileged to have Aedes do Bem!™, because without this restraint we could have two epidemics together, Covid-19 and Dengue” [E].

The Bill and Melinda Gates Foundation awarded a grant of USD5,812,666 to Oxitec in 2018 to further develop this technology for malaria control, to transfer a self-limiting genetic platform into the malaria vector for future application in Meso-America and the Caribbean to reduce or eliminate this mosquito where it transmits malaria [Fi]. The director of the malaria program at the Gates Foundation said Oxitec’s approach would complement other types of interventions being used to stop malaria’s spread. “ Vector control has played a critical role in reducing cases and deaths due to malaria over the past 15 years. With further progress stalled at present, continued innovation of new and transformational interventions is critical to realizing the goal of a world free of malaria.” [Fii].

Results from the early field trials demonstrated the technique was effective at suppressing the wild Ae. aegypti and this had a significant impact on the spread of dengue fever in the trial sites [A]. According to the Piracicaba Department of Health, in the year immediately preceding the trial release of the OX513A GM mosquito strain from July 2014 to July 2015, 133 cases of dengue were recorded in the region. In the period after the release of "Aedes do bem", from July 2015 to May 2016, there were only seven cases of the disease in the same area [G]. As a result, the Mayor of Piracicaba announced that the transgenic Aedes would be a fundamental part of the city’s strategy in combating dengue, zika and chikungunya: “Piracicaba is investing in the expansion of a solution that has already shown itself to be able to control a serious public health problem. And the best part is that we are doing this with innovative, clean, environmentally friendly and sustainable technology” [G].

In the United States there have been several outbreaks of dengue, which is spread by Ae. aegypti - an invasive species in the United States - particularly in Florida, which experienced 62 locally acquired cases in one county in 2020 [Hi]. As such the Florida Keys Mosquito Control District formed a partnership with Oxitec as part of its commitment to explore novel, safe and effective ways to keep mosquito populations at low levels and stop the spread off dengue, as well as an emerging threat from the zika virus. Oxitec received approval from the U.S Environmental Protection Agency (EPA) in May 2020 [Hii] to carry out pilot projects to try and control the spread of dengue via this vector, although activities have been delayed to 2021 due to the COVID-19 pandemic.

As highlighted in the 2014 impact case study, the University of Oxford’s research was the first to result in the development of GM insects ready to for release into field trials. At the time there were few regulatory frameworks in place to manage these activities, prompting the World Health Organisation, the US Department of Agriculture and other national bodies to develop relevant guidance.

The Special Programme for Research and Training in Tropical Diseases (TDR) published its Guidance Framework for testing genetically modified mosquitoes in June 2014 [I]. The Framework extensively quotes the research undertaken at the University of Oxford by Alphey, including [1].

In April 2014, the National Biosafety Committee in Brazil (CTNBio) gave their technical approval for commercial release of Oxitec’s OX513A strain of GM mosquito [J], primarily to try and control the spread of the dengue virus. In doing so, Brazil was the first country in the world to approve the unconstrained release of a genetically modified mosquito [J]. In April 2016 the National Health Surveillance Agency (Anvisa) of Brazil announced that following the approval by CTNBio of commercial release of OX513A, it would grant Oxitec a special temporary registration for OX513A. Anvisa also announced that it would “ create a regulatory framework capable of evaluating this and other similar products that may be developed” and that “ the Agency has already been drafting new rules, under the theme 54.1 of the Regulatory Agenda 2015-2016, ‘Evaluation of Macroorganisms for biological control of vectors and pathogens in urban environment’" [K]. The new strain, OX5034, was reviewed by CTNBio and on 25 May 2020 “ the Commission concluded that this request complies with CTNBio standards and the relevant legislation aimed at ensuring the biosafety of the environment, agriculture, human and animal health” [D].

At the start of the impact period in mid-2013, Oxitec Ltd had 37 staff all based in Oxford. Over the period 2014-2019 staff numbers averaged 56, peaking at 70 in 2017 [L]. In July 2014 Oxitec opened its first factory in Brazil and now has a large facility there - Oxitec do Brasil, Campinas, Sao Paolo, a subsidiary of Oxitec Ltd. Oxitec was purchased in September 2015 for USD160,000,000 [M] by a US company, Intrexon Inc, realising significant returns for its investors, and representing foreign direct investment into the UK. Oxitec Ltd was the Winner of Social Impact Investment of the Year , at the UK Business Angels Association Awards in 2016 [N].

The research on saiga mass mortality [1, 2] led to the development of Standard Operating Procedures for disease surveillance of wild ungulates in Kazakhstan, adapting Food and Agriculture Organisation best practice [A]. These protocols were funded by the Convention on Migratory Species (CMS) and supported by the NGO the Saiga Conservation Alliance, which Milner-Gulland had co-founded in 2006 and continues to serve as Chair and Trustee. The protocols were adopted by the Government of Kazakhstan for use in their ongoing monitoring of saiga populations and their allocation by 2020 of approximately GBP3,400,000 annually for disease surveillance, saiga conservation and anti-poaching activities for saigas [C(i)]. Since 2015, Kazakhstan's saiga population has more than tripled in size (from 100,000 to over 228,000 in 2018 [B(ii)] and over 300,000 individuals in 2019) as a result of conservation action and recovery from the 2015 mass mortality [C(i),(ii)].

The research also informed Government and NGO responses to a subsequent mass-mortality episode in Mongolia (2017-18). The Mongolian population has been hit by a viral pathogen and therefore decreased in size from around 12,000 individuals in 2016 to around 3,800 in 2019 [D(i)], but new actions were implemented (including improved control of smuggling of horn to China) that contributed to an increase in saiga population to 8,500 in 2020 [D(ii)]. The Russian population has remained stable at around 7,000 individuals, due to investment in anti-poaching (e.g. ranger teams) and protected areas [C(iii)]. GBP101,111 of this investment came from a 34% increase in fundraising by the Saiga Conservation Alliance (SCA) in 2015/16. SCA describe the research [1,2] as informing their campaigns for donor funding and enabling them to attract funding from zoos for the first time [E].

The collaborative research [1] underpinned the 2018 re-assessment of the saiga's status on the IUCN Red List [F(i)] and is cited therein. The status remained as Critically Endangered only because in order to ensure the recovery is sustained, there is a five-year lag built in between a species being eligible to move to a lower threat status, and actually moving down (in this case, to Endangered status). The Co-Chair of the IUCN/SSN Antelope Specialist Group stated that it is “particularly important that the underlying science is robust” and further explained that,

“*the research carried out by Professor Milner-Gulland and colleagues concerning the mass mortality of saiga antelopes in Kazakhstan in 2015 ... greatly improved our understanding of mass mortality events, the role of disease in saiga population dynamics, and therefore the extent of the threat which disease is likely to pose to saigas in the future. This was an important consideration in the assessment process.*” [F(ii)]

The saiga antelope is one of the few species for which a UN Convention – in this case jointly the Convention on Migratory Species (CMS) and Convention on International Trade in Endangered Species (CITES) – has an internationally agreed Memorandum of Understanding (MOU) with an action plan and Medium-Term International Work Programme, agreed and ratified by governments, who commit to carrying out its priority actions. The Work Programme is revised every 5 years. Milner-Gulland is Technical Advisor to the CMS, drafting the Work Programme and evaluating both the status of the species and progress towards fulfilling the Work Programme [G]. The Head of the CMS Terrestrial Species Team emphasises the role of the Saiga Conservation Alliance (SCA) as “the official coordinator of, and technical advisor to, the Saiga MOU” and confirmed that the SCA prepares documentation for the CMS, including the overview reports before each meeting of the Signatories” [G]. In particular they confirm that,

“Each of these outputs is founded on Professor Milner-Gulland's scientific research expertise, as she leads on the preparation of these documents, which then inform international conservation policy and the prioritization of actions by the Signatory States, donors and NGOs.” [G]

Accordingly, in April 2019 Milner-Gulland was the scientist in the organising team for a meeting of government and civil society representatives convened by CMS, CITES and the German government. This meeting revised the 2016-2020 work programme to a new draft programme for 2021-2015 [B(i),(iii)], for ratification in 2021 at a meeting of the Parties to the saiga MOU. The underpinning research outputs 1-5 informed the drafting of specific actions under the ‘3.0 Sustainable Use and Trade’ and ‘10.0 Health and Disease’ headings [B(iii),G]. This Work Programme provides the framework which structures and guides all conservation action, policy and fund-raising for this species, by governments, NGOs (local and international) and international bodies.

As an action prescribed in this new Work Programme [B(iii), item 3.10], Milner-Gulland was co-author of a report, “The Sustainable Use of Saiga Antelopes” in Dec 2020 [H], which cites [1,4]. This report is described by CMS as one of the routes to “ inform international conservation policy and the prioritization of actions by the Signatory States, donors and NGOs” [G]. It was commissioned to inform the 4^th Meeting of the Signatories to the saiga MOU (scheduled 2021).

The Oxford team's research [1,2,3] was also instrumental in the major decision of the 2019 CITES 18^th Conference of the Parties in August 2019, to change the status of the saiga in international law: a zero quota was applied on trade in the species' products, so that there could no longer be any legal international commercial trade in saiga products [I(i)]. CITES described in a letter [I(ii)] that the influencing factors were not only “ the extent and causes of the mass mortality episode in Kazakhstan” [1] but also “ *the high prevalence of ongoing consumption of saiga products in Singapore, highlighting that the demand for saiga products regulated under CITES is ongoing and substantial (based on research carried out by the University of Oxford)*”, acknowledging [3]. Similarly, CMS stated that both strands of the research influenced the decision-making process of national governments [G] at this meeting, and the Academy of Sciences in Uzbekistan described that “ the research of the Oxford team on disease and saiga product use directly informed the position taken by the Government of Uzbekistan” when their Delegation approved the proposal for a zero quota [J].

The Singapore-based behaviour change intervention [4,5] was spread through a variety of English-language and Chinese-language news outlets in Singapore including the Straits Times (the premier English-language newspaper) [Ki], The Mothership (a popular English-language news outlet among younger generations) [Kii], and Lion City News (a popular Chinese-language outlet) [Kiii]. These news articles were strategically promoted using targeted advertisements via Facebook (in collaboration with the Saiga Conservation Alliance), Google, and Outbrain (a seller of adverts on third-party websites). Across the three advertising platforms the adverts were shown almost 5,000,000 times [5]. Both Facebook and Google adverts out-performed the expected average reach for similar adverts on these platforms. The research found that 63% of publicly-viewable Facebook user-created content in response to the adverts (e.g. comments, shares with added text) were identifiably pro-conservation (e.g. shock about the message and declaring an intention not to use saiga horn products again). In contrast, only 13% were identifiably negative towards the intervention adverts. The team’s follow-up evaluation [L] found that the target audience was significantly more likely than non-target audience individuals to state that their saiga horn usage had decreased following the intervention (4% compared to 1%), and to cite the intervention as their reason for changing behaviour. The work also indicated a significant decrease across the population in misunderstandings about the saiga’s conservation status, from 28% of the total sample believing they were common in the wild down to 21% [L].

Together these research efforts have caught the imagination of the general public world-wide, with substantial coverage in print, online, TV and on radio. The research on saiga disease was very widely featured in the media [M] and as a result of this raised profile, film-makers have travelled to film saigas [Ci,E,J], from the documentary ‘Ustyurt Saigas. The right to live’, which was awarded the diploma of the Eurasian Academy of Television and Radio [J], to the BBC featuring saigas on ‘Planet Earth II’ (Episode 5: Grasslands, 2016) and ‘Nature’s Weirdest Events’ (2018) [E]. Both in-country conservation NGOs and Governmental Protected Areas within the saiga range have received substantial additional funding for their conservation work from the public worldwide [C,E,J], not only for direct action but also enabling public engagement including 6 steppe ‘wildlife clubs’ in Uzbekistan for up to 3,000 children and adults annually [J].