Is personalised medicine worth it?

Georgina Ferry attended the Personalised Medicine and Resource Allocation Conference, organised by the Centre for Personalised Medicine (CPM, a joint initiative of WTCHG and St Anne’s College) and the Health Economics Research Centre (HERC, part of the Nuffield Department of Population Health) at St Anne’s on 3 March.

DNA double helix with dataThe headline claim that a human genome can now be sequenced in 24 hours for $1000 is routinely contrasted with the 15 years and $3bn it took to obtain the first human sequence. The promise of ‘personalised medicine’ is that with the information available in individual genome sequences, doctors will be able to diagnose and treat their patients more effectively than ever before, saving lives and saving money. But does either claim stand up?

Earlier this month the CPM and HERC brought together an international group of geneticists, health economists and ethicists to debate the issues. As CPM Director Ingrid Slade put it in her introduction, just because it is possible to sequence a patient’s genome does not necessarily mean that it should, or will, be implemented in a healthcare system with limited resources. Does the test result tell you any more than a medical history and family history taken by a competent physician – and if so, what is the value that it can add to patient management? Co-convenor Sarah Wordsworth, Associate Professor of Health Economics at HERC, emphasised that genomics technologies cannot be tackled in isolation, but need input from economists and philosophers fully to evaluate the financial and ethical implications of implementation in the healthcare service.

The purpose of the conference was to identify the challenges to bridging the gap between translation of genomics and its implementation in healthcare, and to explore how those might be overcome. This is the research area covered by the first speaker on the programme, Kathryn Phillips, Director of the University of California San Francisco Center for Translational and Policy Research on Personalized Medicine. ‘Genomics is the most complicated area I’ve studied as a health economist’, she said, and went on to spell out some of the complexities.

For instance, it is more cost-effective to screen across a panel of genes than to test for individual mutations. But if secondary findings turn up along with the result of interest, is it cost effective to return them to the patient? ‘People expect something to be done if a marker is found,’ said Phillips, ‘even if there is no [current evidence of] clinical utility. There’s a sense that if we get a bunch of data we’ll figure out what to do with it – but it’s better to do the economic analysis in advance.’

It’s reassuring that researchers worldwide are tackling these questions. Wolf Rogowski of the Institute of Health Economics and Health Care Management in Munich described the EuroGentest project, which is developing a prioritisation tool that can be introduced across the EU to help decide whether a genetic test meets predefined criteria on evidence of benefit, risk of disease and cost, for example.

Others are already tackling these challenges in practice. Adrian Towse, director of the Office of Health Economics, an independent research organisation in the UK, addressed the problem of valuing platform technologies such as next-generation sequencing. He pointed out that although the cost of running sequencing machines has fallen spectacularly, the costs of preanalytics and assay, data analysis and developing an evidence base for clinical utility are as high as ever. The potential ‘value’ would be in reducing adverse effects of drugs, reducing delays in diagnosis, increasing willingness to start treatment, and gaining approval for treatments that work in defined subsets of patients (such as Herceptin for breast cancer patients with Her2 mutations).

In the UK the Medical Genetics Clinical Reference Group (CRG) works alongside the UK Genetic Testing Network to assess ‘gene dossiers’ on genetic tests submitted for adoption by commissioners in the NHS. CRG chair Frances Flinter reported that current test recommendations include 1300 disorders involving 800 genes, increasingly in the form of panel tests where many genes are sequenced at once. ‘Economic evaluation of the entire pathway [from diagnosis to treatment] is required before the potential contribution of genetic testing can be properly assessed’, she said. Encouragingly she said that several panels demonstrably save money, not only by testing for several genes at once but also by avoiding more expensive investigations or providing certainty about the choice of treatment.

The conference concluded with a reflective session on the ethical implications of genomic analysis in healthcare, particularly on questions of consent and personal autonomy. Based on his experience with the UK 100,000 Genomes Project, Mike Parker, Director of the Ethox Centre, argued that the seeking of informed consent from participants for their genetic data to be held and used for both research and commercial development did not in itself resolve the ethical issues posed by personalised medicine. For instance, he suggested that people’s autonomy should be respected: as long as they are not deceived, they ‘should be able to make decisions on the basis of their own values, in the context of uncertainty’.

Roger Crisp, Professor of Moral Philosophy at Oxford, echoed Parker in saying that the broader ethical framework should include questions of social justice such as non-discrimination and the addressing of health inequalities. He warned that personalised medicine carried the risk of translating economic inequalities into health inequalities, unless it was explicitly targeted to do the reverse.

Other speakers brought a wide range of different perspectives and examples of rigorous research into these questions. The conference was a suitably sobering antidote to some of the over-optimistic claims made for genomic medicine: however, it also made it clear that addressing the economic and ethical questions at an early stage greatly increases the likelihood that the benefits of genomic analysis in health care will eventually be realised.

This entry was posted in Translational medicine and tagged , , . Bookmark the permalink.