Building the framework for precision medicine

Published: 11-May-2015

As science unravels the genome, clinicians have the potential of linking genes to disease. Providing targeted, effective therapies still requires major efforts in data gathering, new clinical infrastructure and new regulation - but will it also mean smaller profits? Susan Birks reports

You need to be a subscriber to read this article.
Click here to find out more.

The development of targeted therapies is rapidly changing the very nature of healthcare. Known as ‘personalised medicine’ by some, or ‘precision medicine’ by others (putting emphasis on the molecular-level information involved), it is reshaping the traditional pharmaceutical sector. As a result, all those involved – from clinics, diagnostic labs, clinical trial companies, contract manufacturers, research companies to drug manufacturers – are facing change and challenges at a pace not experienced before.

According to the US FDA,1 since the approval of Herceptin, the development of targeted therapies has grown rapidly. The FDA’s Center for Drug Evaluation and Research (CDER) has approved 30 targeted therapies since 2012. In 2014 alone, eight of the 41 novel drugs approved were ‘targeted’. They are: Lynparza (olaparib), Blincyto (blinatumomab), Harvoni (ledipasvir and sofosbuvir), Viekira Pak (ombitasvir, paritaprevir, dasabuvir and ritonavir), Cardelga (eliglustat), Beleodaq (belinostat), Zykadia (ceritinib) and Vimizim (elosulfase alpha).

The move towards precision medicine has created new fields of science – genotyping, molecular diagnostics, pharmacogenetics, pharmacogenomics, pharmacoproteomics, metabolomics – as well as a raft of new industries in the making, including next generation sequencing (NGS), biomarkers, companion diagnostics, bioinformatics and databank curators, and biobanks.

The global biomarkers market, for example, is predicted to see a compound annual growth rate (CAGR) of 18.5% from 2013–2018, to reach US$40.8bn by 2018.2 The companion diagnostics (CDx) market, according to Visiongain’s latest market forecast, will reach $3.32bn in 2015.3

But there are more fundamental changes to be adopted by the healthcare arena if precision medicine is to become a keystone of future medical care: for example, how best to acquire and curate the mass of genetic data that needs to be studied and shared. Another requirement is the radical redesign of the drug approval process.

Precision therapies by their very nature involve much smaller numbers of clinical trial participants and accordingly fewer people are likely to benefit from the outcome

In the US, CDER currently employs some flexibility when reviewing applications for targeted drugs. Precision therapies by their very nature involve much smaller numbers of clinical trial participants and accordingly fewer people are likely to benefit from the outcome. According to CDER Director Janet Woodcock, it has had to adapt to smaller development programmes. Among the recent targeted therapies approved, almost 60% were approved on the basis of one main clinical trial along with supporting evidence. In addition, 9% used one or more of the FDA’s expedited programmes, such as Breakthrough, Fast Track, Priority Review and Accelerated Approval.1

These difficult regulatory factors were recognised in President Barack Obama’s 2015 Precision Medicine Initiative4 announced earlier this year. This seeks to provide research funding to expand efforts, mainly in oncology genomics, and to develop a research cohort study of at least one million Americans who would participate by sharing genomic and clinical data, biospecimens and biofluids, and other data.

This collected information is to be shared with researchers and participants in a variety of ways and it is hoped that these efforts will not only result in a more coherent and harmonised regulatory framework to facilitate precision medicine, but also build the data and recruitment resources for that improved framework.

The Initiative has a long list of other objectives, such as: addressing precision medicine data privacy and security-related issues; supporting clinical trials, in partnership with pharmaceutical companies, to test specific drug therapies using precision medicine techniques; developing a new approach to the FDA’s approval for NGS technologies; and supporting new interoperability standards for cross-system data exchanges as part of the national research cohort.

For widespread use of genome sequencing in routine healthcare, the development of accurate and reliable interpretation software is critical

Obama’s 2016 proposed budget would provide $130m to the National Institutes of Health (NIH) to aid the development of the national research cohort; $70m to the National Cancer Institute (part of NIH) to aid research and application of the genomic drivers of cancer; $10m to the FDA to support the development and regulation of databases and NGS technologies; and $5m to the Office of the National Coordinator for Health Information Technology (ONC), which helps to co-ordinate federal data privacy policy.

Most other regions are investing in similar projects, albeit on a smaller scale. Professor Tim Hubbard, Head of Genome Analysis at Genomics England (GE), said: ‘For widespread use of genome sequencing in routine healthcare, the development of accurate and reliable interpretation software is critical.’ As a result GE is funding technological innovations that will enhance genomic sequence data analysis capabilities. It has announced that £8m will be made available for 12–24 months’ product development programmes for successful companies in its phase two assessment.

These include: Congenica for its Sapientia Analytics Platform for commercialising genome diagnostics, clinical research and gene-discovery for rare diseases; Genomics for its statistical tools for population-scale clinical genome analysis; Omixon UK for HLA genotyping from NGS data; Oxford Gene Technology for NGS Interpret; and Seven Bridges Genomics UK for more accurate variant discovery using population genome graphs.

A UK Innovative Medicines and Medical Technology Review5 also aims to make recommendations to accelerate and tackle regulatory obstacles to innovations such as precision medicine. The review will be led by Sir Hugh Taylor, Chair of Guy’s and St Thomas’ NHS Foundation Trust, and he will be supported by an expert advisory group headed by Professor Sir John Bell, Regius Professor of Medicine at Oxford University. The review is supported by the Wellcome Trust and the aim is to ensure that the UK is able to support the future design, development and widespread adoption of medical innovations.

Translating gene targets into treatments

According to Beverly Merz, Executive Editor, Harvard Women's Health Watch,6 so far scientists have identified 3,600 genes responsible for relatively uncommon single-gene disorders; 4,000 genes related to more common conditions like heart disease and diabetes; and several hundred genes linked to cancer. But Merz points out that identifying disease-related genes doesn’t immediately translate into a cure. Rarely is only one gene implicated, therefore, working out how a genetic variation actually affects a person’s health, or how genes interact with one another or with environmental factors, requires considerable research and patient study groups.

While a host of new diagnostic companies have developed molecular tests for faulty genes, there is as yet only limited guidance for the healthcare payers on which tests are clinically appropriate

While a host of new diagnostic companies have developed, and now offer, molecular tests for faulty genes, there is as yet only limited guidance for the healthcare payers on which tests are clinically appropriate and, importantly for the US market, which tests insurance companies will cover. Payers faced with decisions on what types of tests to pay for and under what circumstances, require clear evidence on the clinical usefulness of molecular panels.

In the UK, the National Institute for Health and Care Excellence (NICE) made a recent decision that was seen by some as a milestone in delivering precision medicine to more UK breast cancer patients, when it decided Genomic Health’s Oncotype DX test will be ‘available to eligible breast cancer patients through the NHS in England as the only multi-gene breast cancer test recommended for use as an option to assist in chemotherapy treatment decision-making.’

The Oncotype DX breast cancer test is the only genomic test validated for its ability to predict the likelihood of chemo-therapy benefit as well as risk of recurrence in early-stage breast cancer, which has led to its inclusion in all major international guidelines (ASCO, NCCN, St Gallen and ESMO).


False positives

Improved access to a test is an important step forward to personalised care for breast cancer patients, but in other disease areas one biomarker or diagnostic test on its own may not be enough and some may produce false positives. A recent study7 by Johns Hopkins scientists suggests that sequencing tumour genomes for clues to genetic changes might misdirect treatment in nearly half of all patients, unless it is compared first with a genetic readout of their noncancerous tissue. When the researchers at Johns Hopkins University School of Medicine, working with a personal genome diagnostics testing company, did a comparison and analysis on some available tests for cancer genes, it revealed that almost a third of the cases that showed positive were false positives, where the genes actually had nothing to do with cancer growth.

One biomarker or diagnostic test on its own may not be enough and some may produce false positives

The lack of progress in finding a cure for autism offers a further example of the complexity of linking genetics to therapy. Various attempts have been made to look for a genetic cause for autism – itself not one illness but a spectrum of behavioural and developmental problems. Various groups of researchers have reported varying degrees of success. Another Johns Hopkins-led team of researchers8 claims it has identified a new genetic cause for the disorder, yet their results did not identify any genome-wide significant associations in the overall sample or in the phenotypic subgroups.

This suggests that the extreme clinical variability observed among patients with autism spectrum disorder does not closely reflect common genetic variation. Some commentators have postulated that this might mean some of the clinical variability in autism arises from causes other than genetic vulnerability, such as epigenetic changes or other responses to the environment.

Even for those less complex, genetic-related rare diseases, where precision therapies are on the horizon, the healthcare infrastructure needs rapid change to accommodate the product development and clinical trials required.

Integrated data

As clinical trials become smaller, companies are likely to have to recruit patients from far and wide and they need to be able to integrate all the necessary hereditary, healthcare data from a wide and varying range of patient records – often not in electronic format. What is more, mountains of genomic data are accumulating that are of little use because they are not tied to clinical information, such as family medical history, and they are generally confined to documents that cannot easily be searched or shared.

Genetic and patient data needs to be collected in standardised, shareable electronic formats

Genetic and patient data needs to be collected in standardised, shareable electronic formats

Meanwhile, contract research and diagnostics organisations need high quality samples from biobanks for biomarker discovery and validation studies. This will require a new infrastructure of biobanks that can provide the high quality samples to the research community. A new supply chain transporting and delivering such samples will also be key.

While governments are funding genomic programmes and electronic data harmonisation, and improving the regulatory path to approval, for drug manufacturers the road to a marketable therapy is still long and complex, and commercial success is by no means a given.


1. last accessed 23.04.15

2. last accessed 23.04.15

3. last accessed 23.04.15

4. last accessed 23.04.15

5. Med_TechReview.pdf: last accessed 23.04.15

6. last accessed 23.04.15

7. _sequencing_may_misguide_cancer_treatment_in_nearly_half_of_all_patients_study_shows: last accessed 23.04.15

8. new_autism_causing_genetic_variant_identifie: last accessed 23.04.15

You may also like