Personalised medicine – how do we get there?

As the collection and analysis of increasing quantities of biodata becomes an everyday reality, new possibilities around the prediction, diagnosis and treatment of disease open up. The UK’s Life Sciences Sector Deal, published in December 2017, sees up to £210 million from the Industrial Strategy Challenge Fund committed to the “Data to early diagnostics and precision medicine” programme. In genomics, the 100,000 Genomes Project offers ground-breaking advances for patients with ambitious large-scale sequencing plans. Working with partners like Illumina, Genomics England can progress projects like this in a way that would be challenging without a centralised healthcare system.

What is personalised medicine?

“A move away from a ‘one size fits all’ approach to the treatment and care of patients with a particular condition, to one which uses new approaches to better manage patients’ health and target therapies to achieve the best outcomes in the management of a patient’s disease or predisposition to disease.”
- "Improving outcomes through personalised medicine”, NHS England, 7 September 2016

“Personalised medicine refers to a medical model using characterisation of individuals’ phenotypes and genotypes (eg, molecular profiling, medical imaging, lifestyle data) for tailoring the right therapeutic strategy for the right person at the right time, and/or to determine the predisposition to disease and/or to deliver timely and targeted prevention.”
- European Council Conclusion on personalised medicine for patients (2015/C 421/03), December 2015

The NHS is embracing personalised medicine, and in its 2016 vision statement "Improving Outcomes through Personalised Medicine" recognised the potential to transform clinical practice. NHS England envisages a move towards whole genome sequencing for specific conditions by 2020, alongside improved approaches to rare diseases and cancer. By 2025 the document sees: 

• A new taxonomy of medicine based on underlying cause and personal response. 
• Integrated clinical services taking a “whole body” approach. 
• Tailored, optimised and more effective therapies for better outcomes. 
• New NHS relationships with academia, industry, patients and patient groups.

This reflects EU-level policy. The European Council’s 2015 notice on personalised medicine called on both EU Member States and the European Commission to promote personalised medicine in a variety of ways, including the use of genomics and data sharing to enable data-supported advances. Specifically, it asked the Commission to look at “how to realise the potential of ‘Big Data’, which is used in personalised medicine, in contributing to innovative, efficient and sustainable health systems, respecting the right to protection of personal data.”

The EU’s Horizon 2020 research funding programme has a range of personalised medicine projects under way, including the International Consortium, ICPerMed. This brings in participation from Canada, Turkey and other non-EU member countries alongside EU research organisations like the UK’s Medical Research Council to drive forward research into personalised medicine in a co-ordinated way.

Increasing numbers of approvals

A study by the Personalized Medicine Coalition (PMC) highlights increasing levels of approvals for personalised medicines by US medicines regulator, the FDA. The PMC identifies personalised medicines as those where physicians use diagnostic tests to determine which medical treatments will work best for each patient. And their research shows 2017 as the fourth consecutive year in which personalised medicines accounted for more than 20 per cent of new drug approvals. Among the new molecular entities approved in 2017 were: 

• Breast cancer treatments where use is informed by the patient’s HR and HER2 biomarker status. 
• Hepatitis C treatments the use of which is informed by the patient’s HCV genotype status.

Roadblocks

While both research funders and regulators seem to be getting behind increased use of personalised medicine, there are roadblocks that need to be overcome in order to achieve rapid progress.

Alignment with companion diagnostics

In the EU, regulatory regimes for medicinal products and medical devices are separate, with different legislation and regulatory bodies. This presents an issue for personalised medicines. Where a medicine’s labelling calls for the use of a test for a predictive biomarker, then that diagnostic test will need approval under the devices system, and this is not currently co-ordinated with the route to approval of the medicine itself.

The new In Vitro Diagnostic Devices Regulation will come into play in 2022. This deals with the concept of companion diagnostics, or CDx, and envisages consultation between regulators of medicinal products and devices, but does not fully coordinate the two systems.

The problem for CDx is addressed in the EMA’s “Concept paper on development and lifecycle of personalised medicines and companion diagnostics”. This concept paper paves the way for guidelines on the development of personalised medicines alongside CDx. Public consultation on the concept paper ran to mid-November 2017 with a draft guideline expected during 2018.

Value assessment frameworks and reimbursement

Reimbursement mechanisms vary widely between countries, ranging from private and social insurance-based systems to centralised funding models such as used in the NHS. Convincing payers to allow in their value assessment frameworks for personalised medicines presents a challenge. How do we adequately capture the value of personalised medicine? And can CDx be approved for reimbursement alongside the medicinal product itself? The PMC report Personalized Medicine and Value Assessment Frameworks recognises the issue of funding for CDx, emphasising that they should be regarded as an integral part of the value assessment of a personalised medicine. Other key issues identified by the report are:

• The heterogeneity of treatment effects:
  Personalised medicines will require a greater recognition of the outcomes for individual patients, in contrast to the current approach measuring the average response to a treatment across a population. So, for example, PD-1 immunotherapy treatments work well for some non-small cell lung cancer patients, but they are less effective for others, meaning that the average response is overall less effective.
• Individual values and circumstances:
  An individual’s circumstances may affect whether a particular treatment is right for them. For example, a multiple myeloma patient, may place different emphasis on features such as risk profile, response to previous treatments, tolerance and willingness to suffer side effects, the impact on a patient’s family, and the value that an individual patient places on extending survival with hope of further advancement in science that may offer more curative value. 
• Emerging or evolving value elements:
  Value assessment evidence tends to be static and not sufficiently responsive to important new data. For example, an assessment report on PD-1 immunotherapy treatments for non-small cell lung cancer patients did not consider the potential of these therapies in patients with tumours that express “microsatellite instability,” which was still under study when the report was released. 
• Treatment efficiency:
  Value assessment frameworks generally focus on the effectiveness of the treatment, through analysis of clinical trial data, without taking account of the efficiency offered by personalised medicines. Going straight to the therapy most likely to be effective for a particular patient offers a cost-saving. Likewise, if a diagnostic test could help determine which patients should have a lower dose of a drug or avoid taking it all together due to an increased risk of severe toxicity, the value of that drug would be greater.

Use of personal data

Central to realising the potential of patient data is the need to safeguard privacy. The UK National Data Guardian’s 2016 review of data security highlights the need to retain public trust by adhering to data privacy law and best practice, particularly when making use of data collected through the NHS. Some instances of poor practice have led to adverse press comment and even a public backlash against the use of NHS data.

In Europe, the advent of the General Data Protection Regulation in May 2018 will make the controls over use of personal data even tighter, and significantly raise the price of failure. Fines of up to €20 million, or 4 per cent of worldwide turnover if higher, will potentially be added to the reputational cost of falling short.

Genetic data is specifically recognised in the new law as being a category of personal data, and alongside other health data, it falls within the class of special category data that is awarded a higher degree of protection. The permitted reasons for processing data of this kind are more restricted than for other classes of data, and the controls over activities like profiling and cross-border transfers are more stringent.

Where genetic or health data relating to an identifiable individual is collected and analysed, compliance will require privacy to be designed in from the outset, with anonymisation and tight security measures increasingly important for those holding and using information about individuals. Meeting the new requirements is possible, but requires careful planning and awareness.

A future worth striving for

So there are difficulties, but it is a future worth striving for. As NHS England’s 2016 report puts it when considering the advantages of genome sequencing:

“Whole Genome Sequencing provides a huge step forward in the diagnostic information available. It involves looking at an individual’s entire DNA, rather than looking at specific genes or groups of genes. When analysed with other information about our health and the way people live their lives, it provides much richer information about the complex interactions within a person, and between them and their environment. It offers a greater understanding of the underlying causes, triggers and drivers of disease as well as the likely success or failure of drugs and interventions”.