Regulating genomics – the next decade

The implementation plan for 2021-2022 provides an insight as to how the National Genomic Health Strategy will begin to 'transform genomic healthcare over the next ten years'. An aim of the plan is to 'incorporate the latest advances in genomics into routine healthcare'. One part of the strategy relates to reproductive genomic screening and expectations as to the developments of non-invasive prenatal testing (NIPT) and considers the foreseeability of whole genome sequencing (WGS) to 'detect inherited variants as well as de novo mutations causing severe paediatric conditions.' This links with the ambition set out in the chief medical officer's Annual Report in 2016 relating to personalised medicine, particularly the future of newborn screening.

Is genomic specific regulation indicated to help to realise this plan? Or will current healthcare regulation be sufficient to carry through these changes in ways that 'are anchored on equity of access, data security and privacy'? Is it genomics that needs bespoke regulation or is it the use of data and application of artificial intelligence which requires new regulation? For example, how much can the public sector equality duty ensure diversification of genomic datasets and equity of access? Are existing human rights and equality frameworks sufficient? Do we need more than data protection to ensure the appropriate use of genomic data?

The recent publications relating to genome editing governance by the World Health Organisation (WHO) highlight the importance of ensuring that genetic data collected is representative of the diversity of our global population. This is a challenge that needs to be met head on. The framework for governance explains that genomic data collections are 'not currently representative of the global population and genome editing innovations that make use of the available data can target variants that cause disease only in some populations to the exclusion of others' and therefore there is a need for 'the development of inclusive genome editing innovations that take note of the diversity of the human population and human experience.'

In the UK this will likely mean trying hard to ensure equality of access and removing barriers to accessing trials relating to NIPT and WGS and beyond. Any trials need to be inclusive. The WHO considers the ethical value and principle of 'inclusiveness' is critical to informing what decisions are made. To ensure global accessibility, individual countries must ensure that their own schemes and trials are as accessible as possible by design. The genome strategy has already indicated that it 'will develop robust systems of outreach and communication to diversify our genomic datasets' and address 'ethnic bias historically seen in most large genetic datasets.'

Responsible regulatory stewardship around 'data collection, storage, processing, distribution and destruction in accordance with established privacy constraints' is also key, as the WHO recognises. The implementation plan for the UK genome strategy explains the need to develop 'an ecosystem of world-leading secure genomics datasets.' Data access and data security are both critical to genomic research and clinical realisation of that research. However, so is patient voice, choice, and consent: 'it is equally important that we harmonise patient choice and consent frameworks and data standards, so that patients and research participants across the UK can have trust in the safe, appropriate and responsible use of their data.' The importance of trust and confidence in the provision of data by patients is mission critical to the success of this plan.

Data provision has been a live topic recently with the postponement of the General Practice Data for Planning and Research initiative to extract pseudonymised patient data from primary care records because of concerns at the pace of rollout without consultation, and the need to opt-out rather than opt-in. Some circumspection about data use and provision can be healthy and it shows the importance of public engagement in trying to develop buy-in to new data provision schemes. It also highlights the importance of a rethink of the process (and meaning) of patient consent in the digital age – and opt-in versus opt-out approaches to consenting. However, it also needs to be set in the context of just how much data (relating to our health and beyond) we can generate day to day in our lives (particularly for those who have been living them predominantly from behind a screen since the start of the pandemic). This is particularly so now as many primary care services require initial assessment via an app.

What regulation might need to accompany the genomic strategy?

What kind of domestic regulation is required in the coming decade? Ultimately, it is about ensuring that there is ownership, trust and understanding of any changes. Both informal and formal regulation are relevant for the 2020s.

Informal regulation will be important. Education for the public, for patients, clinicians (building on the work by Health Education England), researchers and commerce relating to genomics and data will have its place, to bring stakeholders into the vision. In the 2020 strategy security and privacy are addressed in the cross-cutting themes of 'engagement and dialogue with the public' and 'maintaining trust'. Trust is vital. As Professor Sir Mark Caulfield (chief scientist at Genomics England) recognised in his evidence to the House of Commons Science and Technology Select Committee on Commercial Genomics, '[i]f we find ourselves in a loss of trust position, people will not take up genomics when they really need it the most in the health system.' Trust is key to continuing participation in longitudinal genomics studies.

Formal regulation might see the introduction of data trusts and data tax. These could steward public use and gains from data, recognise the importance of data as the 'new oil' and seek to avoid situations such as Royal Free/DeepMind. It could ensure that data provided is deployed legally and ethically and for the benefit of the NHS and its patients. If commercially sharing datasets result in financial gains by companies (through apps or selling of AI/tech products) then data trusts and ultimately the NHS should benefit from a portion of these gains. Data tax could also be deployed to tax companies or organisations who solely trade on data – as a disincentive to do so.

Formal regulation could also see the modernising of often labyrinthine or ineffective consent provisions appropriate for the digital age. It could also see the regulation of commercial genomics through future regulations under the Medicines and Medical Devices Act 2021. There should also be greater person-based regulation for researchers/analysts who engage with genomic datasets, so that there is personal accountability to professional standards, alongside organisational accountability (as is the case for biomedical scientists, who are regulated by the Health and Care Professions Council). It could set clear national standards to ensure greater accuracy and consistency across the public or private sector. It could also see the regulation of personal wellbeing genotyping tests/nutrigenomics.

Next steps

Any approach to regulation needs to balance patient protection/empowerment with enabling swift innovation. Ultimately those with rare unidentified diseases or new conditions can be impatient to access genomic medicine, which is why proactive, swift yet careful regulation is imperative.